|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
The Effect of a Summer School Literacy Program on t he Reading Attitudes of Elementary School Struggling Readers by Katie Fradley A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Childhood Education College of Education University of South Florida Major Professor: Susan P. Homan, Ph.D. Darlene DeMarie, Ph.D. John Ferron, Ph.D. James R. King, Ed.D. Date of Approval: November 27, 2007 Keywords: ERAS, mixed-study, scripted literacy, thi rd-grade, Voyager Passport Copyright 2008, Katie Fradley
DEDICATION To my Dad who taught me the importance of persisten ce. I love you. Katie Nothing in the world can take the place of persist ence. Talent will not. Nothing is more common than unsucc essful men with talent Genius will not; unrewarded genius is almost a prov erb; Education will not. The world is full of educated d erelicts. Persistence and determination alone are omnipotent. The slogan press on has solved and always will so lve the problems of the human race. Calvin Coolidge
ACKNOWLEDGMENTS I am truly grateful to the many people who have gui ded and supported me through this process. I would first like to thank my committee, Dr. Homan, Dr. Ferron, Dr. King and Dr. DeMarie. Each of you provided me with a tremend ous amount of support and guidance. Thank you for your patience and availabil ity throughout this process. Dr. Homan thank you for the kindness, flexibility and e ncouragement you have shown. Next, I would like to thank my fellow Manatee County doc students (a.k.a. The Manatee Mafia) Melinda Lundy, Anne Juola-Rushton, Beth Severson an d Ruby Zickafoose for making all those trips to Tampa so enjoyable! I would especial ly like to thank Melinda Lundy for her friendship and help during Focus Groups. My fam ily has been extremely supportive throughout this process. First, a special thank you to my sister Polly for the free editing, support, and the countless cards and celebrations a s we celebrated each hurdle. Next, my heart felt appreciation to my sister in law Tricia for the dissertation therapy sessions on all those early morning runs. I could not have done this without the encouragement and support of my loving parents. To my mother who insp ired me to further my education by example. I would like to thank my son Henry, the mo st patient twelve-year in the universe (and a very talented skate boarder) for hi s love and understanding. Finally, and most importantly I never would have started or fini shed this process were it not for the inspiration, guidance and support of my husband and soul mate Chuck. Thank you for your love, your never ending support, and belief th at I could do this. I love you.
i TABLE OF CONTENTS LIST OF TABLES .................................... ................................................... ........................v LIST OF FIGURES ................................... ................................................... .................... vii ABSTRACT .......................................... ................................................... ........................ viii CHAPTER 1 INTRODUCTION ............................ ................................................... ..........1 Statement of the Problem ......................... ................................................... .............1 Theoretical Framework ............................ ................................................... .............4 Conceptual Framework ............................. ................................................... ............7 Purpose of Study ................................. ................................................... ................10 Research Questions ............................... ................................................... ..............11 Hypothesis........................................ ................................................... ...................11 Significance of Study ............................ ................................................... ..............11 Definition of Terms............................... ................................................... ..............12 Delimitations and Limitations .................... ................................................... .........16 Organization of Remaining Chapters................ ................................................... ..21 CHAPTER 2 REVIEW OF LITERATURE .................... ..................................................2 2 Methods of Selection ............................. ................................................... .............22 Overview and Statement of the Problem ............ ................................................... 22 Theoretical Perspectives ......................... ................................................... ............24 Reading First .................................... ................................................... ...................28 Goals and Purposes of Reading First ............. ............................................29 The Role of Reading First in the Teaching of Readi ng .............................31 Controversy Surrounding Reading First ............ ........................................32 Grade Retention .................................. ................................................... ....34 Summer School .................................... ................................................... ...44 Scripted Literacy Programs........................ ................................................53 Reading Attitude ................................. ................................................... ....68 How the Review of Literature Informed the Study .. .................................72 Organization of Remaining Chapters................ .........................................73 CHAPTER 3 METHOD .................................. ................................................... ...............75 Overview of Chapter .............................. ................................................... .............75 Design ........................................... ................................................... ......................75
ii Voyager Passport Training ........................ ................................................... .........78 Population and Sample Selection................... ................................................... .....83 Pilot Study ...................................... ................................................... .....................85 International Review Board Approval .............. .........................................85 Descriptive Statistics ........................... ................................................... ....86 Dependent Measures t-test ....................... ......................................88 Reliability of the Instrument .................... ..................................................8 8 Cronbach Alpha .................................. ...........................................88 Interrater Reliability and Scoring................ ...............................................90 Threats to Legitimation (Qualitative Phase) ...... ........................................92 Instrumentation .................................. ................................................... .....94 Quantitative Instrument ......................... ........................................94 Data Collection .................................. ................................................... ...100 Quantitative Procedures ......................... ......................................100 Qualitative Procedures .......................... .......................................101 Focus Group Analysis ............................ ..........................101 Classroom Observations .......................... ........................108 Quantitative Data Analysis ...................... ....................................110 Qualitative Data Analysis ....................... .....................................111 Focus Group Analysis ............................ ..........................111 Classroom Observations .......................... ........................114 Triangulation of Findings ...................... .........................115 Mixed Data Analysis.............................. ......................................115 Evidence of Ethical Considerations and District Pe rmission ..................115 Organization of Remaining Chapters................ .......................................116 CHAPTER 4 FINDINGS ................................ ................................................... ..............117 Overview of Chapter .............................. ................................................... ...........117 Summer School Training ........................... ................................................... .......118 Fidelity Checks .................................. ................................................... ...............119 Quantitative Findings ............................ ................................................... ............124 Descriptive Statistics ........................... ................................................... ..124 Assumptions Underlying the Dependent Measures t-te st ........................127 Inferential Statistics ........................... ................................................... ...............128 Dependent measures t-test ........................ ...............................................128 Factorial ANOVAs ............................... ................................................... ..........129 Assumptions ...................................... ................................................... ....130 Results .......................................... ................................................... .........130 Recreational Attitude ............................ ................................................... 130 Academic Attitude ................................ ................................................... 135 Tukey Test ...................................... .............................................139 Total Attitude ................................... ................................................... .....140
iii Summary of Quantitative Findings ................. .........................................144 Qualitative Findings ............................. ................................................... .............146 Focus Group Participants ......................... ................................................147 Classroom Observations ........................... ...............................................154 Fieldnote Analysis .............................. .........................................155 Case Vignettes ................................... ................................................... ...156 Horn Elementary Mr. Owl ......................... ..................................156 Dolphin Elementary Mrs. White ................... ...............................159 Carter Elementary Mrs. Fields ................... ..................................163 Lincoln Elementary Mrs. Smith ................... ................................166 Franklin Elementary Mrs. Golden ................. ..............................169 Cross Case Analysis .............................. ................................................... ............174 Pattern 1: Teachers Thoughts on Voyager ......... ....................................174 Pattern 2: Teacher Characteristics ............... ............................................175 Pattern 3: Activities During Observations ........ .......................................176 Pattern 4: Differences in Teaching Methods ...... ....................................177 Pattern 5: Students Attitudes During Observations ................................177 Focus Groups ..................................... ................................................... ...178 Analysis of Focus Group Data .................... .................................181 Summary of Focus Group Findings ................. ............................200 Triangulation of Findings ........................ ................................................202 McKennas Model of Reading Attitude .............. .....................................208 Normative Beliefs ............................... .........................................209 Instructional Approach........................... ......................................210 Gender .......................................... ................................................211 Testing.......................................... ................................................211 Age and Reading Attitude ........................ ....................................211 Organization of Remaining Chapter ................ ................................................... .212 CHAPTER 5: SUMMARY, CONCLUSIONS, RECOMMENDATIONS AN D IMPLICATIONS ...................................... ................................................... ..............213 Overview of Chapter .............................. ................................................... ...........213 Summary of Study ................................. ................................................... ...........213 Research Questions ............................... ................................................... ............214 Quantitative Findings ............................ ................................................... ............214 Descriptive Statistics .......................... ................................................... ...215 Dependent Measures t-test ....................... ................................................215 Factorial ANOVAs .............................. ..................................................2 15 Qualitative Findings ............................. ................................................... .............216 Limitations, Conclusions and Implications......... .................................................21 8 Recommendations for Practice .................... ................................................... ....220 Fidelity of Implementation ...................... ................................................221
iv Lack of Norming Data ........................... ......................................221 Group Size ..................................... ..............................................222 Prior Exposure to Program....................... ....................................222 Sample ..................................... .....................................223 Length of Intervention ........................... ..................................................2 23 Supplementing of Curriculum....................... ...........................................223 Implications...................................... ................................................... .....224 National Level .................................. ............................................224 Local Level ..................................... .............................................225 Recommendations for Future Research .............. .................................................22 5 Closing Thoughts ................................. ................................................... .............228 LSIT OF REFERENCES .. .................. .............................233 APPENDICES ........................................ ................................................... ......................248 Appendix A: The Elementary Reading Attitude Survey (ERAS) .......................249 Appendix B: Informed Consent Form ........................................255 Appendix C: Sample Voyager Lesson ................................................25 7 Appendix D: Voyager Passport Training Agenda ..... ..........................................259 Appendix E: ERAS Teacher Training Protocol ...... ............................................260 Appendix F: ERAS Group Protocol ................. ..................................................2 63 Appendix G: Informed Consent Pilot Study ........ ...............................................264 Appendix H: ERAS Pilot Study Group Protocol ...... ...........................................266 Appendix I: ERAS Pilot Study Scores .............. ..................................................2 67 Appendix J: Coding Training Protocol ............. ................................................... 268 Appendix K: Focus Group Protocol ................ ................................................... 270 Appendix L: Attitude Scale for Classroom Observati ons....................................272 Appendix M: Example of Transcript Coding ......... .............................................274 Appendix N: Permission Letter from School District ..........................................275 Appendix O: Request Letter for Study to School Dis trict ...................................276 Appendix P: Summer School Day Structure .......... .............................................277 Appendix Q: Voyager Fidelity Measure ............. .................................................27 8 Appendix R: Protocol for Gathering and Analyzing F ield Notes ........................280 Appendix S: Focus Group Clarification and Probing Routines ...........................282 ABOUT THE AUTHOR .................................. ................................................... .. End Page
v LIST OF TABLES Table 1 Definition of Reading Attitudes .................................................. ................6 Table 2 Reading Attitude Models ........ ................................................... ................28 Table 3 Meta-analyses of Retention Resea rch ............................................... .........37 Table 4 Summary of Grade Retention Resea rch................................................ .....43 Table 5 Summer School Research ......... ................................................... ..............52 Table 6 Summary of Research on Voyager U niversal Systems .............................66 Table 7 Descriptive Data Pilot Study.... ................................................... ...............87 Table 8 Summary of Fidelity Visits ...... ................................................... ..............123 Table 9 Descriptive Statistics........... ................................................... ...................127 Table 10 Results of Paired samples t-Test ................................................... ...........129 Table 11 Variable Differences in Recreation al Attitude ....................................... ...132 Table 12 Differences in Recreational Attitu de by Site and Gender .........................133 Table 13 Variable Differences in Academic A ttitude ........................................... ...136 Table 14 Difference in Academic Attitude in Regards to Gender and Site .............137 Table 15 Tukeys Studentized Range for Diff erence in Academic Attitude ...........140 Table 16 Variable Differences in Total Atti tude .............................................. .......142 Table 17 Difference in Total Attitude in Re gards to Gender and School Site ........143 Table 18 Number of Focus Group Participants by Site .......................................... .149
vi Table 19 Information on Focus Group Partici pants ............................................. ....150 Table 20 Student Observations During Summer School ........................................17 3
vii LIST OF FIGURES Figure 1. Number of Students Promoted After Summer School 2005-2007 ...................9 Figure 2. Changes in Attitudes Pilot Study ........ ................................................... .........87 Figure 3. Mean Scores ERAS ........................ ................................................... ...........126 Figure 4. Difference in Recreational Attitude in Re gards to Gender and School Site .............................................. ................................................... ...............134 Figure 5. Difference in Academic Attitude in Regard s to Gender and School Site .............................................. ................................................... ...............138 Figure 6. Difference in Total Attitude in Regards t o Gender and School Site ............144 Figure 7. Gender of Focus Group Participants ...... ................................................... ...153 Figure 8. Ethnicity of Focus Group Participants ... ................................................... ...153 Figure 9. Summary of Attitude of Classroom Observat ions of Targeted Students .......................................... ................................................... ...........178 Figure 10. What part of summer school do you like b est? ............................................18 3 Figure 11. Do you like reading? Why or why not? ... ................................................... .186 Figure 12. What types of books do you like to read? .................................................. ..189 Figure 13. Does anyone read to you at home? ....... ................................................... .....192 Figure 14. Have you noticed any changes in your fee lings about reading? .................194 Figure 15. Tell us about Voyager Passport. ........ ................................................... ........198
viii THE EFFECT OF A SUMMER SCHOOL LITERACY PROGRAM ON T HE READING ATTITUDES OF ELEMENTARY SCHOOL STRUGGLING R EADERS Katie Fradley ABSTRACT This mixed-method study explored and examined the r eading attitudes of thirdgrade struggling readers (n=91) following six weeks of summer school using a scripted literacy program ( Voyager Passport ). During the quantitative portion of the study students (n=91) from five different summer school s ites were given the Elementary Reading Attitude Survey (McKenna & Kear, 1990). The survey, which was administered by the classroom teacher the first day of summer sc hool and the last week of summer school, provided scores for academic, recreational and total reading attitude. Following data collection the results of the ERAS surveys wer e analyzed using a dependent measures t-test as well as descriptive statistics. Results revealed no significant differences in recr eational or total reading attitude following summer school using a scripted literacy p rogram. Gender and school site were both examined using a multivariate analysis. Result s indicated no statistically significant differences based on gender. However, when academic attitude was examined the results for school site were found to be significant F (4, 90) = 2.87, p = .03. A follow-up Tukey test revealed that although there was a difference in academic attitudes between the school sites, the variation could not be pinpointed to particular sites.
ix The qualitative portion of the study relied on both field notes gathered through classroom observations (n=113) and focus groups. On e focus group was held at each of the five summer school sites. During focus groups a group moderator asked the students a series of six questions. Results were analyzed usin g semantic content analysis (Stewart & Shamdasani, 1990) to identify themes related to stu dents attitudes about reading. After a cross case analysis of the targeted classrooms was conducted, triangulation was used to compare the findings from the ERAS survey, classroo m observations, and focus groups. The qualitative findings revealed that following su mmer school students liked to read, felt they were better readers, and felt prepared to take the standardized test. However, only 29% of the students passed the alternative ass essment. The results also revealed questions regarding the fidelity of the implementat ion and concerns with the lack of norming data on the fidelity measure.
1 CHAPTER 1 INTRODUCTION Statement of the Problem Each year approximately 2.4 million children in th e United States are retained, ultimately costing taxpayers 14 billion dollars a y ear (Jimerson & Kaufman, 2003). Though three meta-analyses, containing 700 analyses of achievement from more than 80 studies published between 1925 and 1999, fail to su pport the use of grade retention as an early intervention to enhance academic achievement, retention is currently being used with struggling readers (Holmes, 1989; Jimerson, 20 01; Jimerson & Kaufman, 2003). In the state of Florida, where 14 percent of the third graders in 2006 failed to meet promotion criteria, promotion to fourth grade is di rectly tied to performance in reading on the Florida Comprehensive Achievement Test (FCAT) ( Institute for School Innovation, 2006 ) Retained third-grade students in the state of Flori da are frequently encouraged to attend a district summer reading camp where they receive intensive reading instruction. Counties are responsible for writing their own summ er reading camp schedules, which incorporate state guidelines. The first guideline concerns the time requirements for summer school. According to state guidelines, stude nts are to attend summer school for six hours per day, four days per week. The duration of summer school is from six to eight weeks. The next guideline concerns instructional re quirements. Intensive reading
2 instruction must last a minimum of two hours per da y (i.e., one-third of the total instructional day). The remainder of the student da y is allocated for reading enrichment in the form of read alouds, independent reading, mento ring and technology. Additionally, the state stipulates that formal assessment last no more than 30 minutes per day. In this study in a school district on the west coas t of Florida during the summer of 2007, students attended summer school five days per week for five hours each day for a total of 150 total hours. The research-based interv ention program used was Voyager Passport. According to Passport ( Voyager Expanded Learning, 2004), the program is designed for use with struggling readers in kinderg arten through third grade, who are performing below grade level, and for whom the co re-reading program is not working. The Florida Center for Reading Research (FCRR) defi nes core-reading programs as, comprehensive reading programs that are intended f or use as the initial instruction in kindergarten through third grade classrooms (Flori da Center for Reading Research, www.fccr.org ). In this study, during the 4.5 hours of actual da ily instructional time, students received two 45 minute Voyager Passport lessons. Voyager states that the goal of instruction is to accelerate struggling readers to grade-level proficiency through 26 weeks of targeted, explicit, systematic instruction ( Voyager Principals Handbook, 2006). Third-grade lessons consisted of instruction from two modules. The first module focused on comprehension and vocabulary, while the second module focused on fluency. An additional optional component focused on word st udy and was designed for use with students who were reading less than 44 words per mi nute on a grade level passage. An additional component of Voyager Passport is ongoing progress monitoring. Progress monitoring is defined as, a process of ev aluating individual student reading
3 progress between benchmark periods in order to make instructional decisions ( Voyager Expanded Learning, 2004). Progress monitoring occur s every fifth lesson and utilizes VIP (Vital Indicators of Progress), a progress moni toring system made up of brief, oneminute measures for evaluating students developmen t of grade-level reading skills such as initial sound fluency. When Voyager Passport is used during the traditional school year, these results are logged into an online syste m called V-Port which tracks the students growth. V-Port was not utilized during su mmer school. Further information on V-Port can be found in Chapter Two. Voyager Passport defines the program as explicit and systematic ins truction. Specifically, Voyager defines explicit instruction as, A direct instruc tional approach in which the teacher states the reason for learning th e skill, models it, gives the students guided practice, and provides independent practice with feedback. Additionally, Voyager ( Voyager Expanded Learning Systems, 2004) defines systemati c instruction as, An arrangement of skills in a logical order from t he easiest to the most difficult. Combined explicit and systematic instruction provid es repeated practice of clearly stated skills delivered in a way that ensures understandin g and minimizes confusion (p.33). Although Voyager Passport does not refer to the program as scripted, it does demonstrate many characteristics matching the defin ition of a scripted program. Scripted reading is a reading program characteriz ed by an explicit teachers manual with instructions for teachers to follow verbatim w hen using the program with their students (Moustafa & Land, 2005). In a scripted c lassroom, all activities are to be followed in the order presented, and the teachers instructions are to be read word-forword from the manual (Meyer, 2002). This approach c an be contrasted with a non-
4 scripted program, which describes activities, provi des examples, and expects teachers to choose activities they deem to be the most appropri ate to use with their students (Moustafa & Land, 2005). Specifically, Voyager Passport requires minimal preparation; encourages the teacher to closely reference the curriculum guide; uses explicit language; has carefully sequenced and paced skills; uses a pace that is bri sk and business-like; includes teacher modeling and monitoring of students understanding; engages students through eye contact, hand signals, brief verbal reminders; and uses corrective procedures ( Voyager Expanded Learning, 2004). However, although the man ual provides specific teacher dialogue, Voyager encourages teachers implementing the program to, Become familiar with both the directions and implementation and to refrain from reading the script verbatim (p. 30). Reading attitude can influence factors such as eng agement and practice (Mathewson, 1994; McKenna, Kear, & Ellsworth, 1995) Because summer school is currently being used as an intervention to assist t he struggling reader, it is important to examine any relationship that may occur between sum mer school using a scripted reading program and struggling readers attitudes. Theoretical Framework This study, which focused on the reading attitudes of third-grade struggling readers, was grounded in theory on attitude. An exa mination of the literature revealed varying definitions of attitude depending on the in vestigator. Table 1 offers a summary of the various definitions of reading attitude. Theore tical underpinnings primarily stem from two reading attitude models: the Mathewson (1985) m odel and the McKenna (1994)
5 model. Mathewsons (1985) model focuses on attitude and the role it plays during both the act of reading and during the period of time wh en one learns to read. Additionally, this model predicts attitude development over time (McKenna et al., 1995). When applying Mathewsons (1985) model, attitude is just one of a set of factors that influence an individuals intention to read (McKenna et al., 1995). McKennas (1994) model strives to examine the longterm development of reading attitudes. This model examines three princi ple factors influencing attitudinal change: (a) beliefs about the outcomes of reading i n light of the judged desirability of those outcomes, (b) beliefs about the expectations of others in light of ones motivation to conform to those expectations, and (c) the outcomes of specific incidents of reading (McKenna et al., 1995). Additionally, McKenna (1994 ) sees reading attitude as being broken down into two different dimensions: attitude toward recreational reading and attitude toward school related academic reading.
6 Table 1 Definitions of Reading Attitude ___________________________________________________ ___________________________________________________ ___ Investigator Year Definition of Attitude ___________________________________________________ _____________________ Ajzen 1989 An individuals disposition to respon d favorably or unfavorably to an object, person, institution, even t, or any discriminable aspect of their world (p. 241) Alexander and Filler 1976 A system of feelings rel ated to reading which causes the learner to approach or avoid a read ing situation (p.1). Beck 1976 A positive or negative evaluation of so me person, object, or thing (p. 302). Fishbein and Ajzen 1983 A learned predisposition t o respond in a consistently favorable or unfavorable manner with respect to a given object (p.6). Petty and Cacioppo 1981 Positive or negative feeli ng about some person, object or thing (p. 302).
7 Conceptual Framework For the qualitative component of this mixed method study, the conceptual framework was driven by the effects that the No Child Left Behind Legislation has on third-grade students, and the role that retention, summer school, and a scripted reading program have on struggling third-grade readers att itudes toward reading. McGuire (1989) speculated that the tri-component view of at titude could be measured more effectively through the use of open-ended responses This view supports the use of qualitative research as a way to gather open-ended data when investigating the reading attitudes of students using a tri-component view of attitude, such as that adopted by McKenna (1994). In formulating his model, McKenna s ynthesized the work of Mathewson (1985) and others and in doing so McKenna s model (1994) identifies three principal factors that contribute to attitudinal ch ange. These factors are: beliefs about the desired outcomes of reading, beliefs about the expe ctations of others, and the outcomes of specific incidents of reading. Even though statistics on retention reveal the nega tive impact retention has on children (Jimerson & Kaufman, 2003; Jimerson, 2001; Parker, 2001), it is currently being used as a strategy that is intended to assist the s truggling reader. Third-grade students who do not pass the reading portion of the FCAT exa mination are retained unless they qualify for a good cause exemption. One of these good cause exemptions involves the strategic use of summer school. During summer s chool the students receive instruction using a reading program that has state approval based on the states definition of scientifically -based reading research (SBRR), o r research which uses rigorous, systematic, and objective procedures to obtain kno wledge about reading development,
8 reading instruction, and reading difficulties. Foll owing the intervention, the students are given an alternative reading assessment, the Stanfo rd 10 Test. If they score in the 45th percentile or higher on the alternative assessment they are eligible for promotion to the fourth grade. Voyager Passport was used as an intervention tool in the district w here the study took place during summer in 2005, 2006, and 2007. The scores have varied each year (see Figure 1). In 2005, when Voyager Passport was used as an intensive summer school reading program with 263 third-graders, 72 percent of students who otherwise would have been retained achieved proficiency on the SAT9 (alternative reading assessment) and thus were promoted to fourth-grade. According t o the Florida Department of Education, the targeted district achieved the highe st summer gains of any county in the state of Florida. In 2006, 547 third-grade students in the same distr ict scored at Level 1 on FCAT, thus qualifying them to attend a summer reading cam p. Third-grade students who scored at Level 1 were encouraged, but not required, to at tend summer school. Once again Voyager Passport was used as an intensive summer school reading pro gram. Of the 241 third-grade students who completed summer school an d participated in SAT 10 testing, 43 students (17.8 percent) were promoted based on t his score. In order to be placed in fourth-grade, Florida statute requires a third-grad e student to place in the 45th percentile or higher on the Stanford 10 assessment. In 2007, a total of 285 students completed summer school and took the Stanford 10. Of those st udents, 83 students (29 percent) scored in the 45th percentile or higher and were placed in fourth gra de as a good cause exemption.
9 The promotion findings from 2005 to 2006 differed g reatly. However, it is difficult to compare the results from 2005 to 2006 because in 2006 the district changed the assessment tool and began using the Stanford 10 as the alternative assessment. The decision to change the assessment was made at the s tate level because the norms from the Stanford 9 were out of date. The Stanford 10 was no rmed in the fall of 2002 with a large sample of the nations K-12 student population. Aft er the update, all of the test questions on the Stanford 10 were new and the creators claime d that the Stanford 10 reading test had more items that addressed skills in critical an alysis and strategies (Harcourt, 2003). Additionally, selections of poetry were added at al l levels of the test. An additional change from the Stanford 9 to the Sta nford 10 was that the Stanford 10 was not timed. Figure 1. Number of Students Promoted After Summer School 20 05-2007 nnrnnrrn nrn n
10 Purpose of Study The goal of No Child Left Behind (No Child Left Behind Act, 2002) is for every third-grade student to receive effective reading in struction so that they are able to read on grade level. This legislation promotes the use of g rade retention, summer school, and reading programs based on scientifically based read ing research (SBRR) as interventions to assist the struggling reader. Although researche rs have previously examined the effects of summer school as an intervention strategy (Borma n & Dowling, 2006; Cooper, Charlton, Valentine & Muhlenbruck, 2000 and Duffy, 2001), and the use of summer school has been promoted as a way to accelerate the reading development of struggling readers (Allington, 1998; Duffy, 2001), there remains only limited research that focuses on summer school that uses scripted reading program s and the progress that struggling readers can be expected to make. Voyager Universal Literacy System is marketed as a comprehensive reading system. As the program has experienced success, the company has added additional components. One of these components is Voyager Passport Although there is research available to support Voyager Universal Literacy System as a core reading program, because Voyager Passport is a new program completed early in 2003, there is only scant research that is specific to the intervention ( www.readingfirstsupport ). Additionally, although numerous researchers have ex amined factors that influence childrens attitudes toward reading (Fishbein & Ajz en, 1975; McKenna & Kear, 1990; McKenna, et al., 1995), little research has focused specifically on struggling third-grade students attitudes toward reading. Presently there is only minimal research that is
11 specific to the intervention, Voyager Passport Therefore, the purpose of this research study was to determine the effect a summer school l iteracy program using a scripted reading program, Voyager Passport had on the reading attitudes of struggling thirdgrade readers. Research Questions The following research question was addressed in th e quantitative portion of this study: What is the effect of a scripted summer scho ol reading program on the reading attitudes of third-grade struggling readers? The fo llowing research question was addressed in the qualitative portion of this study: What do third-grade struggling readers perceive to be the effect of a scripted summer scho ol reading program on their attitudes toward reading? Hypothesis The Quantitative Hypothesis was: There is a positiv e relation between struggling third-grade readers attitudes toward reading succe ss and the completion of a scripted summer school reading program for third-grade strug gling readers. Significance of the Study Given the present political climate supporting the use of grade retention, and considering the amount of research that has shown t he negative effects grade retention can have on a child, it is important to find interv entions such as summer school to assist the struggling reader. However, it is of equal impo rtance to consider struggling readers attitudes toward reading and how the use of a scrip ted reading program affects struggling readers attitudes toward reading. This study has t he potential to contribute to the field of literacy education. It is hoped that findings from this study will help county
12 administrators with decision making in regards to s ummer school and what instructional materials to use with struggling readers. Definition of Terms Core-Reading Programs. The Florida Center for Reading Research (FCRR) defines core-reading programs as comprehensive read ing programs that are intended to be used as the initial instruction in K-3 classroom s (Retrieved 12/2/2006 from Florida Center for Reading Research, www.fccr.org ). DIBELS. The Dynamic Indicators of Basic Literacy Skills (D IBELS) are a set of standardized individually administered measures of early literacy development. DIBELS is widely used to monitor early reading progress wi thin classroom settings and measures initial sound fluency, letter naming fluency, phone me segmentation fluency, nonsense word fluency, and oral reading fluency. DIBELS was largely unknown before Reading First, yet DIBELS is now the primary assessment too l promoted by the Department of Education under Reading First. DIBELS measures init ial sounds fluency, phoneme segmentation fluency, nonsense words fluency, and o ral reading fluency. Initial Sounds Fluency assesses a child's skill to identify and pr oduce the initial sound of a given word. Phonemic Segmentation Fluency assesses a child's sk ill to produce the individual sounds within a given word. Nonsense Word Fluency assesses a child's knowledge of lettersound correspondences as well as their ability to b lend letters together to form unfamiliar "nonsense" (e.g., fik, lig, etc.) words. Finally, O ral Reading Fluency assesses a child's skill of reading connected text in grade-level mate rial. ( http://dibels.uoregon.edu/dibels_what.php ).
13 Elementary Reading Attitude Survey (ERAS) The Elementary Reading Attitude Survey is designed to provide quantitative estimate s of childrens attitude toward both recreational and academic reading and can be admini stered to an entire class in a manner of minutes. (McKenna & Kear, 1990). ESEA Elementary and Secondary Education Act of 1965. The No Child Left Behind Act (NCLB) reauthorized and amended federal education programs established under the Elementary and Secondary Education Act (E SEA) of 1965. Explicit Instruction A direct instructional approach in which the tea cher states the reason for learning the skill, models it, gives the students guided practice, and provides independent practice with feedback ( Voyager 2006, p. 33). FCAT. The Florida Comprehensive Achievement Test, or the FCAT, is the standardized test used in the primary and secondary public schools of Florida. FCRR The Florida Center for Reading Researchs mission is to conduct basic research on reading, reading growth, reading assess ment, and reading instruction that will contribute to the scientific knowledge of read ing and benefit students in Florida and throughout the nation; to disseminate informati on about research-based practices related to literacy instruction and assessment for children in pre-school through 12th grade; to conduct applied research that will have a n immediate impact on policy and
14 practices related to literacy instruction in Florid a; and to provide technical assistance to Florida's schools and to the State Department of Education. No Child Left Behind Legislation (NCLB) The No Child Left Behind Act of 2001 (Public Law 107-110), commonly known as NCLB, is a United States federal law that reauthorizes a number of federal programs that aim to improve the performance of U.S. schools (ed.gov, 2006). NRP The National Reading Panel Group commissioned by Congress to review the growing body of reading research K-3. It was co mposed of some of the nations leading experts in reading research. Report of the National Reading Panel Report that reflects the findings of the National Reading Panel, a congressionally funded st udy in 2000. An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. Reading First. A grant program for schools that fail to meet stan dards set forth by the national government under No Child Left Behi nd (NCLB). Under Reading First, qualifying schools receive federal money ove r a three-year period to provide teacher education, programs, materials, remedial pr ograms, and ongoing monitoring of student progress. Scientifically Based Reading Research (SBRR). A term defined by Reading First as a scientifically based reading research th at uses rigorous, systematic, and objective procedures to obtain knowledge about read ing development, reading
15 instruction, and reading difficulties. This type of research is defined by Reading First as employing systematic and objective procedures to obtain valid knowledge relevant to reading development, reading instruction, and re ading difficulties; and includes research that employs systematic, empirical methods that draw on observation or experiment. It also involves rigorous data analyses that are adequate to test the stated hypotheses and justify the general conclusions draw n; relies on measurements or observational methods that provide valid data acros s evaluators and observers and across multiple measurement and observations; and h as been accepted by a peerreviewed journal or approved by a panel of independ ent experts through a comparably rigorous, objective, and scientific review (Title 1 Part B, Section 1208(6) of the ESEA). Scripted Reading Program. Scripted reading programs are characterized by very explicit teachers manuals with instructions f or teachers to follow verbatim when using the program with their students (Moustafa & L and, 2005). In a scripted classroom, all activities are to be followed in the order presented, and the teachers instructions are to be read word-for-word from the manual (Meyer, 2002). Systematic Instruction An arrangement of skills in a logical order from the easiest to the most difficult. Vital Indicators of Progress (VIP) VIP was developed by Dr. Roland Good and Dr. Ruth Kaminski and is an alternative form of DIB ELS, which is widely used to monitor early reading progress within classroom set tings. VIP measures initial sound
16 fluency, letter naming fluency, phoneme segmentatio n fluency, nonsense word fluency, and oral reading fluency. Voyager Passport Program. Voyager Passport is a reading intervention program for struggling readers in kindergarten through thir d-grade who are performing below grade level. The program provides intensive, explic it instruction in phonemic awareness, phonics, fluency, vocabulary, and comprehension tha t can be delivered by a teacher, reading specialist, trained paraprofessional or stu dent teacher. Lessons are highly structured, using clear, succinct language and leav ing little flexibility for teacher decision making. VPORT. A tool that allows district administrators, princi pals, literacy coaches, and teachers to monitor student progress, compare s tudent data against a trajectory of desired learning, and make instructional decisions to match student needs. Access to VPORT is set up through the district or school. Delimitations and Limitations of the Study There were delimitations to the study. The study on ly focused on third-grade struggling readers whose reading performance was me asured by FCAT. Therefore, the researcher limited the participants of the study to struggling readers who scored at Level 1 on FCAT reading. Furthermore, because the sample was a convenience sample, the researcher also limited the study to struggling thi rd-grade readers who were attending summer school in the targeted district. In 2006, 54 7 third grade students scored at Level 1 on the reading portion of FCAT and qualified for su mmer school. Of those students, 241
17 completed summer school and took the alternative as sessment, Stanford 10. Although the Elementary Reading Attitude Survey (Mckenna & Kear, 1990) (see Appendix A) was administered to all of the students, data were only collected from those students who returned a yes informed consent. There were several potential threats to both intern al and external validity for both the qualitative and quantitative portions of this s tudy. According to Johnson and Christensen, (2004), internal validity refers to, the ability to infer that a causal relationship exists between two variables (p.230). One possible threat to the internal validity of this mixed study concerned researcher b ias. Onwuegbuzie (2003) contends that researcher bias can occur when the researcher also is the person collecting the data. The researcher attempted to prevent researcher bias by making her intentions clear with the participating students and summer school teache rs. Johnson and Christensen define reflexivity as, Self-reflection by the researcher on his or her biases and predispositions (p.249). The researcher actively engaged in reflexi vity, by engaging in critical selfreflection about potential biases and predispositio ns (Johnson & Christensen, 2004). Through reflexivity the researcher became more self -aware which helped to monitor and control for biases concerning the use of scripted l iteracy programs. An additional limitation evolved during the study a nd involved politics and the conflict of interest that arose because the researc her was also a third-grade teacher in the district where the study took place. This meant tha t the researcher had to modify certain aspects of the study to comply with district reques ts. For instance, the district would not allow the researcher to conduct the fidelity checks This was because the district felt that
18 it was a conflict of interest for the researcher to be in a position where she was evaluating fellow teachers. Another threat to internal validity involved instru mentation. Instrumentation can cause a threat to internal validity if the instrume nt used during pre-testing is different than the instrument used in post testing. This threat wa s controlled for by using the same instrument for the pre-test and post-test. Addition ally, another potential threat to internal validity concerns mortality. Mortality is often a t hreat to validity when studying at-risk students who often are more likely to drop out of a study (Onwuegbuzie, 2003). This was controlled for by attempting to obtain as large a s ample as possible. Because attendance at summer school was optional, and participation in the study was optional, some of the students elected not to participate, and others dro pped out of summer school before the researcher had given the post-assessment. (Specific information on participants can be found on p. 82). An additional mortality issue that occurred was that some of the students were not present on the first day of summer school and were never given the initial ERAS (McKenna & Kear, 1990) survey. Additionally, there were several threats to externa l validity. According to Johnson & Christensen, (2004), external validity is referre d to as The extent to which the results of a study can be generalized to and across populat ions, settings and times (p. 242). Ecological validity refers to the extent to which f indings can be generalized across settings, conditions, variables, and contexts (Onwu egbuzie 2003). Because the findings from the study reflected the views of struggling th ird-grade readers, it might not be productive to generalize the results across differe nt populations. Because the participants were aware that they were participants in a researc h study, the researcher needed to
19 consider reactivity, which is defined by Johnson an d Christensen (2004) as, an alteration in performance that occurs as a result of being awa re of participating in a study (p. 245). To control for this possible threat, the researcher collected both quantitative and qualitative data. The qualitative data, gathered th rough participation in focus groups and field notes gathered during classroom observations, were then compared with the findings of the quantitative data gathered through the ERAS (McKenna & Kear, 1990). An additional threat to external validity arose whe n the district became involved in student selection for the focus groups. Because the district wanted to impact as few classrooms as possible, the researcher was asked to select students from just one classroom at each site. This meant that the researc her was limited in participant selection. There were very few children per site with permissi on to participate in the study who had a low reading attitude initially. In fact, only 10 of the 91 students had an initial low attitude score. The majority of the students who pa rticipated in focus groups had an average or high attitude. Population validity refers to the ability to gener alize results of the study to individuals not included in the study (Johnson & Ch ristensen, 2004). In this study, the targeted population consisted of struggling readers However, the accessible population consisted of third-graders who were enrolled in sum mer school. To help control for population validity, the researcher invited all of the retained third-graders attending summer school in the spotlighted district to partic ipate in the study. However, in order to meet IRB requirements, participation hinged on whet her or not the students returned a yes informed consent (see Appendix B). Johnson and Christensen (2004) define temporal validity as, the extent to which the stud y results can be generalized across
20 time (p. 245). This was an issue because the data for this study was collected during the period of one summer. Therefore, although the data were valid for this time period, there was no assurance that the same results would hold v alid across time (Johnson & Christensen, 2004). Maturation refers to physical or mental changes tha t may occur within individuals over time (Johnson & Christensen, 2004). Because th ese changes over time may affect an individuals performance over time, another possibl e threat to internal validity involved maturation. The researcher attempted to control for this possible threat to validity by limiting data collection to the 30 days the student s were enrolled in summer school. Additionally, the researcher conducted a Pilot Stud y to determine if the Elementary Reading Attitude Survey (see Appendix A) measured c hange with a homogeneous population of struggling readers following a major event. These results showed that the ERAS measured differences during a short timeframe, since the Pilot Study was an examination of students attitudes about reading bo th before and after the FCAT. When examining the assumptions for possible violati ons, because the criterion variable, attitude, was assessed using a continuous variable there was no reason to believe that the assumption level of measurement had been v iolated. Additionally, because the same measure of attitude is used both as a pretest and as a posttest, there was no reason to believe that the assumption of paired observations had been violated. All of the students scores were independent, so the assumption independ ent observations had not been violated. An examination of the normal probability plot reveals that the scores appeared to be normally distributed, thus the normal distrib ution for different scores assumption
21 had not been violated. Because the sample selected for this pilot study was a convenience sample, the researcher knowingly violated the assum ption of random sampling. Organization of Remaining Chapters The remaining chapters present information which i s pertinent to this study. Chapter 2 begins with an examination of literature on struggling readers focusing on federal and state initiatives, grade retention, and summer school, scripted literacy programs, and students attitudes toward reading. The topic of Chapter 3 is methodology. This chapter begins with information on the Voyager Passport training and the instructional fidelity measure. Next, descriptions of the design of the study, the population and sample selection, instrumentation, d ata collection, and the manner in which the data were analyzed and interpreted are pr esented. Chapter 4 summarizes the findings of the study. Both quantitative and qualit ative findings are reported. Chapter 5 presents a summary of the study, conclusions and im plications derived from the research findings, recommendations for practice based on the study conclusions and implications, and recommendations for future research.
22 CHAPTER 2 REVIEW OF LITERATURE Methods of Selection This chapter begins with an overview and a statemen t of the problem. Following the problem, this chapter is organized into five re search strands presented in five sections. The first section will examine the research on the federal initiative, Reading First The second section will examine research on the detrime ntal effects of grade retention. The third section will look at summer school as an inte rvention with struggling readers. The fourth section will examine scripted literacy readi ng programs, specifically Voyager Passport and the role it presently has under NCLB. The fif th section will consist of an examination of factors contributing to the reading attitude of struggling readers. Finally, the researcher provides an overview of how the lite rature informed the study. Overview and Statement of the Problem During the 1998 State of the Union Address, President Clinton brought retention into the national spotlight when he called for an end to social promotion. Three years later, and just three days after President Bush took offic e in January 2001, the No Child Left Behind act began with the intent to improve student achiev ement and change the culture of U.S. schools (www.ed.gov). Six years later, as s tates work to implement No Child Left Behind reading has become a top political issue and a ma jor focus of the Bush presidency. President Bush describes this law as th e, cornerstone of my administration (Retrieved 12-06-2006 from www.ed.gov ).
23 The ultimate goal of No Child Left Behind is for every child in the United States to read on grade level by the end of third grade. I n Florida, an additional component of this legislation involves assessment using The Flor ida Comprehensive Achievement Test (FCAT). Reading FCAT scores group students into fiv e levels, with one being the lowest level and five being the highest. Floridas current law requires that students must earn above a level 1 on the reading portion of the third grade FCAT to be promoted to fourth grade unless they qualify for one of the following six good cause exemptions. These exemptions include: (a) Students with limited Engli sh proficiency (LEP) who have less than two years of instruction in an English for Spe akers of Other Languages (ESOL) program; (b) Students with disabilities for whom pa rticipation in the statewide assessment program is not appropriate; (c) Students who demonstrate through a student portfolio, that they are reading on grade level bas ed on the Sunshine State Standards; (d) Students with disabilities who were previously reta ined in grades K-3; (e) Students who were previously retained in grades K-3 for a total of two or more years; and (f) Students who demonstrate an acceptable level of performance on an alternative standardized reading assessment. (Office of Program Policy Analy sis and Government Accountability, 2006). The last good cause exemption involves sending re tained third-graders to summer school, also called summer reading camps. Under Florida state law, retained students must be given the opportunity to participa te in the districts summer reading camp (Office of Program Policy Analysis and Governm ent Accountability, 2006). Instruction in these summer reading camps require s the use of a reading program based on scientifically based reading research (SBRR). In the district used in this study students
24 attending summer school received remediation using Voyager Passport Upon completion of summer school, the students were give n an alternative reading assessment, Stanford 10. If the student scored at or above the 45th percentile, he or she was then placed in fourth grade as a good cause exemption. Little research has been undertaken on the impact s ummer school has on struggling readers attitudes. Additionally, althou gh there has been research on Voyager Universal Literacy Systems, much of that research has been conducted by the co mpany. Because Voyager Passport is a relatively new intervention program, there is limited research on it. This research will be presented in the section on Voyager Furthermore, because retained students are at greater risk for l ong-term negative outcomes such as dropping out of school, increased substance abuse, fewer employment opportunities, and more behavioral problems (Holmes, 1989; Jimerson 20 01; Jimerson & Kaufman, 2003; Parker 2001) and because certain instructional appr oaches may produce positive experiences, which may in turn contribute to attitu de influences; it is important to consider not only alternatives to retention, but ho w they impact the struggling readers attitude towards reading. Theoretical Perspectives The theoretical underpinnings of this study stem fr om research on reading attitudes. Much of the research on attitude began i n the 1930s and increased during the 1960s and 1970s as researchers began to examine o ther variables that influenced the attitude-behavior relationship (Fishbein & Ajzen, 1 975; Liska, 1984). During the mid 1970s Fishbein and Ajzen (1975) developed a causal relationship model, the Fishbein/Ajzen model. Under the model, Fishbein and Ajzen provided a general
25 definition of attitude as, a learned predispositio n to respond in a consistently favorable or unfavorable manner with respect to a given objec t (p.6). Specifically, when viewing attitude using this recursive model, attitude was b roken into three concepts: affect, cognition, and conation (behavioral intentions). Th e model assumed that behavior was directly caused by conation (behavioral intentions) which was caused by attitudes, which in turn reflected the beliefs about the consequence s of behavior. Also during the 1970s Alexander and Filler (1976) offered a definition fo r attitude that was specific to reading, but different from that offered by Fishbein and Ajz en. Alexander and Filler defined reading attitude as, a system of feelings related to reading which causes the learner to approach or avoid a reading situation (p.1). This definition suggested that attitude could be thought of as existing on a continuum, with both positive and negative extremes. During the 1990s, and building on the work of the earlier attitude theorists, two new models of reading attitude emerged, The Mathews on model (1985) and The McKenna model (1994). An examination of both models revealed attitude as one of a set of factors influencing an individuals intention to read (Mathewson, 1994; McKenna et al., 1995). Unlike the Fishbein/Ajzen model (1975) Mathewson (1985) did not adopt a causal relationship and instead adopted a tripartit e approach to attitude towards reading. Under this approach, Mathewson (1985) views attitud e towards reading as being made up of three components: prevailing feelings about read ing (personal values), action readiness for reading (goals), and evaluative beliefs about r eading (self-concepts). The two other factors that contributed to the decision to read, o r not to read, were external motivators (cognitive) and the individuals emotional (affecti ve) state.
26 McKenna (1994) synthesized the work of other theor ists in an effort to, construct a model more conducive to considering the long-term development of reading attitudes (p. 938). McKennas model (1994) adopted three prin cipal factors which influence attitudinal change: (a) beliefs about the outcomes of reading in light of the judged desirability of those outcomes; (b) beliefs about the expectations of others in light of ones motivation to conform to those expectations; and (c ) the outcomes of specific incidents of reading (p. 938). McKennas model (1994) predicted that as children g ot older and had more options available to them during their leisure time their attitude towards reading would worsen (Anderson, Tollefson & Gilbert, 1985; Martin 1984). Further research also supported that struggling readers attitudes about r eading declined as they got older (Ishikawa, 1985; Ross & Fletcher, 1989). McKennas model (1994) also predicted that reading attitude was linked to reading ability (Wal berg & Tsai, 1985; Wallbrown, Brown, & Engin, 1978). Although there are varying opinions on what impact instructional methods have on childrens attitudes about reading, there is lit tle evidence to support the effects of methods and materials on reading attitudes. McKenna Stratton, Grindler and Jenkins (1995) reported no difference in the attitudes of 1 -5 students taught using a whole language approach as compared with their peers who received reading instruction from a basal reader. In a nationwide survey McKenna et al., (1995) teste d McKennas model (1994) as they investigated the reading attitudes of 18,185 s tudents in grades 1 through 6. Attitude
27 was measured using the Elementary Reading Attitude Survey or ERAS (McKenna & Kear, 1990). The survey was made up of 20 questions and used a 4 node, pictorial rating scale using Garfield. The survey results were broke n down into two subscales: academic reading attitude and recreational reading attitude. Results of the national survey showed that childrens attitudes decreased with age, begin ning positive in first grade but ending in indifference in sixth grade. Additionally the r esearchers found a relationship between students reading ability and their attitude towards reading. The researchers also examined gender and ethnicity and found that girls as a grou p possessed more positive attitudes about reading and that gender differences did not p lay a role in ability. Regarding ethnicity, it appeared to play little role in stude nts negative trend toward reading attitude. Finally, the researchers examined teachers relianc e on the basal reader to see if it impacted childrens attitudes about reading. The re sults showed that there was not a relationship between time spent in basal readers an d childrens attitudes about reading. These findings offer support for McKennas model (1 994) and argued for more studies in this area. These results offered further support for the McKen na model (1994) which postulated the attitudinal impact of a childs read ing experiences. It is important for educators to use early intervention with struggling readers in an effort to curb the attitudinal decline (McKenna et al., 1995). Little research has been conducted which specifically examines different reading methods and materials to determine their impact on the attitudes of struggling readers. This invest igation applied the theoretical underpinnings of McKennas model (1994) and sought to determine if a scripted summer
28 school reading program had an effect on the reading attitudes of third grade struggling readers. Table 2 offers a summary of these key atti tude models. Table 2 Reading Attitude Models ___________________________________________________ __________________________________________________ Name of Model Year Purpose of Model ___________________________________________________ __________________________________________________ Mathewson 1976 Clarify relationships between attit ude and reading using a small set of variables: attitude, motivati on, attention and comprehension. Fishbein and Ajzen 1980 Attitudes do not affect be havior directly but are mediated by intention. Attitudes toward reading gives rise to intention to read, which then lead s to reading itself. Mathewson 1985 To increase the scope of the 1976 m odel. McKenna 1994 Postulates that an i ndividuals attitude toward reading will develop over time principally as the result of three factors: normative beliefs, beliefs about the outcomes of reading and specific reading experiences (p. 939). Reading First The purpose of this section of the literature revie w is threefold. First, the researcher will examine the goals of Reading First and how they specifically relate to the struggling
29 reader. Next, the researcher will examine the role that Reading First plays in the support and interventions struggling readers receive. Final ly, this section will examine the controversy that presently plagues Reading First Goals and Purposes of Reading First The No Child Left Behind Act (NCLB) focused reading instruction on findings compiled by the National Reading Panel. In 1997 the National Reading Panel was charged by Congress to, Convene a national panel t o assess the status of research-based knowledge, including the effectiveness of various a pproaches to teaching children to read (National Reading Panel, 2000). Under the umb rella of NCLB there are many state and federal initiatives that focus on improving rea ding for young students. According to the OPPAGA Report (Office of Program Policy Analysi s and Government Accountability, 2006), these programs, stress the importance of identifying struggling learners, providing intensive remediation, and ensu ring that low performing students do not fall further behind (p. 3). In 2002 the U.S. Department of Education implemente d the largest and most focused early reading initiative ever undertaken in the Unites States, Reading First The purpose of Reading First was, to ensure that all children in America learn to read well by the end of third grade (Guidance for the Reading First Program, 2002). In 2002 Florida began receiving Reading First grant funds. That year the state received approximately 43 million dollars. During 2003 and 2004 the state received an additional 100 million. Reading First has five purposes, which are described in Title 1, Part B, section 1201 of the Elementary and Secondary Educat ion Act (ESEA).
30 These include: 1. To provide assistance to State educational agencies (SEAs) and Local educational agencies (LEAs) in establishing reading programs for students in kindergarten through grade 3 that are based on scie ntifically based reading research (SBRR), to ensure that every student can r ead at grade level or above not later than the end of grade 3. 2. To provide assistance to SEAs and LEAs in preparing teachers, including special education teachers, through professional de velopment and other support, so that teachers can identify specific rea ding barriers facing their students and so the teachers have the tools to effe ctively help their students to learn to read. 3. To provide assistance to SEAs and LEAs in selecting or administering screening, diagnostic, and classroom based instruct ional reading assessments. 4. To provide assistance to SEAs and LEAs in selecting or developing effective instructional materials (including classroom based materials to assist teachers in implementing the essential components of reading instruction), programs, learning systems, and strategies to implement metho ds that have been proven to prevent or remediate reading failure within a st ate. 5. To strengthen coordination among schools, early lit eracy programs, and family literacy programs to improve reading achieve ment for all children. (p. 4 The Reading First Programs Grant Application Pro cess, Final Inspection Report, 2006).
31 The Role of Reading First in the Teaching of Readin g The Reading First initiative provides states, districts, and schools with funding to implement scientifically based reading instruction (SBRR) in grades K-3. States must apply to the Department of Education for funding, w hich comes in the form of large formula grants to State Education Agencies (SEAs) t hat submit approved applications. Although the funds are only used with K-3 students, they are distributed based on a formula that calculates the number of children age 5 to 17 living below the poverty line in that state. From the SEA, the funding is distribute d to the Local Education Agencies (LEAs) who can apply for competitive sub-grants. Al l 50 states, the District of Columbia, and several U.S. territories participate in Reading First grants. The key to this funding is that it must be used to purchase programs that are based on scientifically based reading research (SBRR). Sc ientifically based reading research (SBRR) is defined by Reading First as research that, uses rigorous, systematic, and objective procedures to obtain knowledge about read ing development, reading instruction, and reading difficulties. This type o f reading research involves controlled experiments with data analysis and a thorough peerreview process (www.Reading firstsupport.us.org). The National Reading Panel (2 000) used the following guidelines to determine what research was considered scientifical ly based reading research (SBRR). First, the research had to address achievement in o ne or more skills in reading. Next, the findings had to be able to be generalized to the la rger population of students. Third, the research had to examine the effectiveness of an app roach by comparing it to other types of instruction. Finally, the research had to be pub lished or scheduled for publication in a
32 refereed (peer reviewed) journal (Report of the Nat ional Reading Panel, 2000; www. Readingfirstsupport.us.org). The phrase, scie ntifically based reading research is an important part of Reading First In fact, the phrase appears more than 100 times i n the NCLB 2001 law (Grunwald, 2006). Under Reading First in order for early literacy instruction to be eff ective, reading programs must provide explicit and systematic instr uction in the following five key areas of reading: phonemic awareness, phonics, vocabulary fluency, and comprehension. According to Reading First states and districts are allowed to select their own textbooks and programs as long as they are backed by sound sc ience. Controversy Surrounding Reading First Presently a great deal of controversy surrounds Reading First Immediately following the release of The Report of the National Reading Panel Teaching Children to Read (2000), questions about the research surround ing the report began to surface primarily amongst educators (Allington, 2002; Cunni ngham, 2001; Krashen, 2001; Krashen, 2005). During the past several years, que stions and concerns with Reading First have escalated, and a myriad of critics of Reading First have emerged. Since 2002, Education Week has reported concerns amongst researchers and educ ators. Numerous articles and editorials have been published that ar e centered around Reading First In recent years these articles and editorials have multiplied and are no longer restricted to educational journals and books. The c ontroversy surrounding Reading First can now be found in a variety of journals, newspape rs, books, and on the Internet. An example of this is a recent report in The Washington Post Billions for an Inside Game on Reading (Grunwald, 2006), which attacks scienti fically based reading research.
33 Grunwald purports, Reading First had little to do with science or rigor. Instead, t he billions have gone to what is effectively a pilot p roject for untested programs with friends in high places (p.1). Grunwald is referring to the allegations that there were ulterior motives behind the reading programs that were appro ved under Reading First Allegations such as those by Grunwald correspond w ith the inspector generals findings. In 2006 a long awaited Final Inspection R eport on the Reading First grant application process was published by The U.S. Depar tment of Education. The executive summary of the report (The Reading First Programs Grant Application Process Final Inspection Report, 2006), concludes that the Panel s method of screening panel members for possible conflict of interest issues was not ef fective. In fact, the report uncovered six panelists (serving on the National Reading Panel) w hose resumes revealed, Significant professional connections to a teaching methodology that requires the use of a specific reading program (p.1). Additionally, the findings state that the department did not follow its own guidance for the peer review process with some states applications funded without documentation that they met all of the crit eria for approval. Further, the findings state because, Criteria developed by the departmen t included language that was not based on the statutory language, state applications were forced to meet standards that were not required by the statute (p.1). Ultimatel y the findings reveal that federal officials may have overstepped provisions of the NC LB act thus the program officials failed to maintain a control environment that exemp lifies management integrity and accountability (p.2). Under Reading First a majority of funds have gone to support tradition al textbook publishers such as Scott Foresman, Macmillan, McGra w-Hill, Harcourt, Houghton
34 Mifflin, and Open Court. To meet the needs of strug gling readers under Reading First many schools have adopted supplemental materials, i ncluding Voyager Passport Voyager Passport is one of the companies that experienced tremendou s success under Reading First The company that produces Voyager was estimated to be worth $5 million before Reading First Recently the company sold for $380 million dollar s (Grunwald, 2006). In June 2005 Reid Lyon, who was chief of the child development and behavior branch of the National Institute of Child Health an d Human Development began working for Randy Best, the entrepreneur who founded Voyager (Manzo, 2006). Grade Retention Ironically, although the goal of No Child Left Behind (NCLB) is for every child to read on grade level by the end of third grade, No Child Left Behind (NCLB) has left countless struggling students behind as retained st udents. Grade retention, the act of having a child repeat a grade, is also referred to as flunking, non-promotion, and being held back (Jimerson & Kaufman, 2003). In addition t o low achievement, retained students frequently have the following characterist ics in common: low parental IQ, lack of parental involvement (Jimerson, Carlson, Rotert, Egeland, & Sroufe, 1997), typically boys and often minorities. In addition, retained st udents are likely to have missed a greater percentage of school days than their peers who have been promoted (Jimerson et al., 1997; Jimerson & Kaufman, 2003). When examinin g race, the children most likely to be retained are African-American or Hispanic childr en (Rafoth, 2002). The purpose of this section of the literature revie w is to uncover research findings on grade retention as an intervention to assist the struggling reader. In an effort to question why retention is still being used as a str ategy to assist the struggling reader, the
35 researcher examined the findings of three meta-anal yses examining the effectiveness of grade retention as an intervention. The first systematic, comprehensive overview on gra de retention was provided by Jackson (1975) and included 30 studies conducted be tween 1911 and 1973. Jacksons goal in conducting the review was to determine whet her low-achieving students or those with socio-emotional maladjustment benefited more f rom grade retention or promotion to the next grade (Jackson, 1975; Jimerson, 2001). Ja ckson included students of all ages in his review and used design type as a way to categor ize studies into three groups: naturalistic (retained compared to promoted), pre-p ost (retained performance before retention compared to performance after retention), and experimental (potential retainees randomly assigned to be retained or promoted). Although Jackson (1975) concluded that, There is n o reliable body of evidence to indicate that grade retention is more beneficial than grade promotion for students with serious academic or adjustment difficulties (p. 62 7), he cautioned researchers against concluding that promotion is better than retention. Rather, the results of his review of research showed that, research evidence is so poor that valid inferences cannot be drawn concerning the relative benefits of these two optio ns (p. 627). Additionally, Jackson recommended that more research was needed, but caut ioned that the research would take years to complete. Following Jacksons systematic review other researc hers began to examine the vast amount of research on the subject of grade ret ention. McAfee (1981) supported Jacksons conclusion regarding the quality of reten tion research and purported that in order for researchers to determine whether or not r etention was beneficial, it was
36 necessary to conduct research with experimental des igns. However, he went on to state, Unfortunately, it seems that most school districts will be unwilling to adopt such a strategy because of the political ramifications (p 22). After examining eight matched studies where retained students were matched to pro moted students on the basis of achievement test scores Holmes (1983) concluded tha t although the purpose of retention is for retained students to catch up, research does not support this practice. Holmes (1983) further suggested, Retained pupils fall beh ind the year they are retained and spend the rest of their academic careers in vain tr ying to catch up (p.4).
37 Table 3 Meta-analyses of Retention Research Researcher Years # Studies Criteria Met hod Participants Ag es Holmes 1929-1981 44 original research e ffect size 11, 132 elem / jr high & Mathews promotion vs. retention Holmes 1989 63 origi nal research effect size n/a kinder/elem/ comparison group jr high Jimerson 1990-1999 20 original research ef fect size 2, 806 comparison group K-high school
38 A total of three meta-analyses containing 700 analy ses of achievement, from more than eighty studies published from between 1925 and 2006 fail to support the use of grade retention as an early intervention to enhance academic achievement (Jimerson & Kaufman, 2003; Jimerson, 2001; Holmes, 1989; Silber glitt, Appleton, Burns & Jimerson, 2006). The results are indicated in Table 3. A meta-analysis was conducted by Holmes and Matthew s (1984) to determine the effects of retention on elementary and junior high school students using both achievement and socio-emotional outcomes. The meta-analysis inc luded 44 studies published between 1929 and 1981. After calculating 575 individual eff ect sizes, the mean effect size was (-.37), meaning that on average the retained group scored (.37) units lower on the outcome measures than the promoted group. Outcomes included: academic achievement, language arts, reading, mathematics, word study ski lls, social studies, personal adjustment, social adjustment, emotional adjustment behavior, self-concept, attitude toward school and attendance (Holmes and Matthews, 1984). Holmes and Matthews (1984) concluded that, Educational professionals w ho continue to retain students do so despite cumulative evidence demonstrating that the potential for negative effects consistently outweighs positive outcomes (p. 232). These results confirmed the findings of Jackson (1975). A second meta-analysis was conducted by Holmes (198 9). Findings gleaned from this meta-analysis once again indicated that after examining 63 studies from between 1925 and 1989, the results showed the overall negat ive effects associated with grade retention. Although findings from 9 studies yielded positive results, the benefits of
39 retention diminished over time (as cited in Jimerso n, 2001, p.422). The results of this meta-analysis offer further support for both Jackso n (1975) and Holmes and Matthews (1984) meta analyses. Jimerson (2001) provided a systematic review as wel l as a meta-analysis of research from 1990-1999. The studies included in hi s review utilized a combination of IQ, academic achievement, socio-emotional adjustmen t, SES, and gender to match groups of control analyses between the comparison g roup and the retained students (Jimerson, 2001). Supporting previous findings, Jim ersons meta-analysis contains similar results to findings reported over the prece ding 90 years (Holmes, 1989; Holmes & Matthews, 1984; and Jackson, 1975). Specifically, analyses which focused on the repeated year produced a mean effect size of (.09) in favoring the retained students. However, longitudinal results demonstrate a mean ef fect size of (-.31), meaning initial gains from repeating a grade often disappear over t ime. Only four of the 20 studies examined exploring the efficacy of grade retention, support the use of retention. The other 16 studies failed to support retention. Jimer son (2001) contends, Researchers, educators, administrators, and legislators should c ommit to implement and investigate specific remedial intervention strategies designed to facilitate socio-emotional adjustment and educational achievement of our nations youth (p. 435). Thus, findings from the past decade reports result s that are consistent with the converging evidence and conclusions of research for m earlier in the century that fail to demonstrate that grade retention provides g reater benefits to students with academic or adjustment difficulties than does promo tion to the next grade (p. 434).
40 Findings from the three meta-analyses fail to demon strate that retention provides greater benefits to struggling students, than promo tion to the next grade. Furthermore, results show that grade retention actually can be d etrimental to a childs future. Shephard and Smith (1990) conducted a synthesis of research on grade retention and concluded that although retained students may appear to do better in the initial year following the retention, they are at much greater risk for futur e failure than their equally achieving, non-retained peers (p. 84). Grade retention has e ven been identified as the single most powerful predictor of dropping out of high school ( Jimerson & Kaufman, 2003; Parker, 2001; Rumberger, 1987). Approximately 60 percent of students retained once drop out of high school by Grade 12. Even more disturbing, stud ents who are retained twice have a 90 percent chance of dropping out before high schoo l (Mann, 1987; Parker, 2001). In addition, grade retention has been linked to other long term, negative outcomes including fewer employment opportunities, substance abuse, ar rests, more behavior problems, higher level of emotional distress, and reckless be havior (Jimerson & Kaufman, 2003). According to Darling-Hammond (1998), Students who are retained essentially do worse in the long run than comparable students who are promoted, in part perhaps because they do not receive better or more suitable teaching when they are retained, and in pa rt because they give up on themselves as learners (p. 18). A longitudinal study (Ferguson, Jimerson, Dalton, 2 001) followed 106 kindergartners through 11th grade examining the effects of family characterist ics, school readiness, socialization, and student demographics on academic achievement and behavioral adjustment outcomes. Students were class ified into one of four categories:
41 students retained in kindergarten, students retaine d in first or second grade either through a transitional program or by traditional early grad e retention, students recommended for a transitional class who were promoted, and students who were promoted on schedule. The study examined within group factors. Dependent vari ables were represented by specific academic and behavioral outcomes. Independent varia bles were socioeconomic status, mothers level of education, parental value of educ ation, age and kindergarten personal social functioning. The researchers employed descri ptive statistics, multiple regression, Analysis of Variance, and Chi Square statistical an alyses. The participants consisted of students who were ret ained as well as those who were recommended for retention, yet were promoted. Results indicated that students, who were recommended for retention, yet were promoted a nd experienced academic success had certain factors in common. These included: moth ers who graduated from college, only minimal delays on the Gessell Developmental D elay index, no kindergarten personal-social functioning deficits, strong scores on standardized tests and participation in a ninth grade sport. Students who were retained and did not experience success after retention also had certain characteristics in commo n. Of the retained students, results showed that older students who had demonstrated ear ly personal-social deficits were especially disadvantaged by retention as were retai ned students whose mothers had a low level of education, lower socioeconomic status, or low parental view of education. Although these findings are noteworthy, there are possible limitations to the study. The small sample size may make it difficult to generalize the findings to a larger population. Another possible limitation relates to the number of independent variables, representing a specific level of contextual analysi s.
42 Further evidence of the negative effects of grade r etention comes from a longitudinal study examining the effects of grade r etention on student reading performance (Silberglitt et al., 2006). The study f ound that retained students did not experience a benefit to being retained. Further res ults showed that when retained students were compared to similar performing peers that were not retained, the researchers found no difference in slopes, thus showing the treatment had no effect. Additionally, when the retained students were compared to a randomly selec ted group of students, they made less progress. Strengths of the study included a large s ample size and data collection that was extensive and lasted for years. However, upon close r examination of the data, 92% of the population used in the study was Caucasian. This ma kes it difficult to generalize these findings to a broader population because the childr en most likely to be retained are African American and Hispanic students (Rafoth, 200 2). A summary of grade retention research can be found in Table 4.
43 Table 4 Summary of Grade Retention Research Researcher/Year Goal Size Design Findings Jackson, 1975 Determine benefits to retention. Review of 44 studies n/a Retention not beneficial. Peterson, Gracie, Ayebe, 1987 Determine the long term effect of retention/promotion on academic achievement. 106 first, second and third graders Matched comparison groups Students performed better the year following the retention, lose it in the second or third year. Roderick, 1994 Explore influence retention has on grad. 707 drop outs and graduates Event History Analysis Retention increases odds of dropping out increase Jimerson, 1999 A 21 year examination of the long term effects of retention. 21 years Longitudinal Retained students more likely to drop out than promoted peers who are performing equally. Jimerson, Anderson & Whipple, 2002 Examine retention as a predictor of drop out status. 17 studies Systematic Comprehensive Review of Research Retention associated with subsequent school withdrawal. Ferguson, Jimerson & Dalton, 2001 Explored factors associated with longitudinal and academic behavioral outcomes 107 Kindergarten-Eleventh Grade Prospective Longitudinal Lower SES, lower level of mothers ed, lower parental value of ed., inter. skills, students age all risk factors. Silberglitt, Appleton, Burns & Jimerson, 2006 Used HLM to compare retained students to promoted students. 147 First-Eighth Grades Longitudinal Retained stude nts did not experience a benefit or deficit in their reading growth.
44 After thoroughly examining the bulk of research on the issue of grade retention, (Darling-Hammond 1998; Holmes 1989; Jimerson 2001; Jimerson & Kaufman 2003; Parker 2001) it is evident that it is more benefici al to focus on instructional strategies to assist the education of children at risk of academi c failure, rather than to retain them (Jimerson, 2001; Jimerson & Kaufman, 2003). Despite these conclusions, retention is not only being used today but has become an important p iece of The No Child Left Behind legislation. Under NCLB, one of the interventions offered to ret ained third-grade students is summer school. Retained third-graders have the oppo rtunity to attend free reading summer school for six weeks. At the end of the six weeks the students are given an alternative reading assessment. Students who succes sfully master the alternative assessment are then placed in fourth grade as a go od cause exemption. Summer School Under No Child Left Behind legislation summer schoo l has taken an important role as a strategy to assist the struggling reader. In the state of Florida reading summer school is provided at no cost to second and third g rade students who do not achieve mastery on standardized tests. The purpose of this section of the literature revie w is to examine the research on summer school and to specifically examine summer sc hool as an intervention for struggling readers. Summer programs to remediate le arning deficits can be grouped into four categories: summer programs to help students m eet minimum competency for graduation or grade promotion, summer school as an opportunity to retake a course,
45 summer school as a way to provide a program beyond the school year for children with disabilities, and summer school as a way to prevent summer learning loss (Cooper, Nye, Charlton, Lindsay, and Greathouse, 1996). Many researchers have documented the summer reading slide (Cooper et al., 1996). The analogy of a faucet is used by Entwisle, Alexander and Olson (1997) to describe the summer reading slide and the effect it has on students from different socioeconomic backgrounds. When school is in sessio n, all of the students receive equal resources. However, during the summer, the faucet i s turned off. During this drought students from poor families do not receive the same resources that the school provides during the school year. Parents of higher SES statu s, however, provide extra resources such as vacations, summer camp and trips to the lib rary for their children. Thus the faucet theory suggests that this lack of resources for the child of lower SES status could lead to inequality when his performance is compared to that of his peers. After reviewing 39 studies Cooper et al., 1996 conc luded that achievement test scores declined over summer vacation. Key findings from an additional part of the review involved a meta-analysis of 13 of the studies showi ng that summer loss for a typical student was equal to about one months worth of kno wledge in math and reading. Further findings revealed that summer break was more detrim ental on math and spelling progress than on reading. Offering support for the faucet t heory, research shows that the summer slide is particularly harmful to students from low socio-economic status (Cooper et al., 1996; Heyns, 1978). Downey, von Hippel, and Broh (2 004) concluded that there was a correlation between SES and summer reading loss. Sp ecifically, Downey et al., (2004) estimated that the reading level of a student with a family income of $40,000 fell 2.5
46 months behind a student with a family income of $10 0,000. This offers additional support for Heyns 1978 findings which showed that not only is there an achievement gap for students as a result of summer school, but this gap tends to be greater among the havenots than among the haves (Borman & Boulay, 2004 ). In todays era of accountability, summer school is presently being used as a core pr ogrammatic component of the high stakes testing initiative (Borman, 2000). Although summer school is currently being used as a n intervention with struggling students, much of the research on summer school is non-experimental. During the early 1970s researchers found that summ er programs in math, reading and language-communication showed modest achievement ga ins as well as having a positive effect on students attitude about school and learn ing (Austin, Rogers, & Walbesser, 1972). Results of a large-scale national study on s ummer learning, The Sustaining Effects Study (SES), found that although there were reading gains over the summer there may have been math losses. Results from data from over 120,000 students revealed that, overall in comparing the achievement gains of stude nts who attended summer school with those who did not attend, no differences were found (Carter, 1984). The Teach Baltimore program began in 1992 and has p rovided summer instruction to more than 2,100 Baltimore City publi c school students, as well as recruiting 287 college students from a variety of majors. The mission of Teach Baltimore is, To create high-quality summer learning opportunities f or students from high poverty communities and to improve teacher recruitment and retention in Baltimore City (Borman & Dowling, 2006, p. 3). A longitudinal stud y examined the effects of a
47 multiyear summer school program in preventing summe r reading loss and promoting longitudinal achievement growth (Borman & Dowling, 2006). The purpose of this voluntary summer school was to avert the summer achievement slide and have a positive impact on stu dents learning. Participants included 438 students from high poverty schools. The goal of the research was threefold. First, the researchers wanted to study the effectiveness of Th e Teach Baltimore Summer Academy on summer learning loss. Next, the researchers want ed to transform collegiate volunteerism into a focused and effective commitmen t. Finally, the researchers hoped to create a successful prototype that could be replica ted. Class size was limited to eight students. Summer school lasted for seven weeks and included breakfast, 3 hours of intensive reading and writing, lunch, physical acti vity, hands on math and science projects, arts and crafts, and enrichment activitie s. Additionally, the students attended weekly field trips to museums and cultural events. The volunteers attended an extensive training as well as working closely with a mentor t eacher. The method involved contrasting longitudinal outcomes for the participa nts with 248 children in the control group. Findings from the longitudinal study showed that al though summer school can improve the achievement of at risk students, encour aging and sustaining students long term participation was a challenge. According to th e researchers, approximately 50 percent of the students assigned to the program att ended with enough regularity to make a difference. Students who attended at least two of t he three years had achievement scores at least one standard deviation higher than those s imilar peers in the control group.
48 A meta-analysis by Cooper, Charlton, Valentine, and Muhlenbruck (2000) examined the findings of 93 studies of summer schoo l. The researchers used quantitative synthesis to analyze the findings. Results showed t he average effect size for remedial summer school programs equal to approximately one-f ifth of a standard deviation (d=0.19). Findings from the meta-analysis showed th at among students attending summer school, those children who were middle class benefi ted more than those who were disadvantaged. Additionally, researchers found the following characteristics of summer school programs to be related to achievement: small group/individualized instruction, early intervention, parent involvement, and treatme nt fidelity. In contrast to these characteristics, Austin et al., (1972) found the fo llowing characteristics many ineffective summer school programs have in common: short durati on, limited academic focus, and low academic expectations. Summer Bridge is a summer school program in Chicago for third, sixth and eighth graders not meeting minimum score requirements on t he Iowa test of Basic Skills (ITBS), and in danger of being retained. Unlike Teach Balti more, attendance in Summer Bridge is required. A study using multiple methods examined h ow low performing students attending summer school perceived their summer lear ning environments (Stone, Engel, Nagaoka, & Roderick, 2005). Although this study onl y examined data from 1999, the program, has been in place since 1997. It consists of six weeks of instruction for three hours a day for third and sixth graders. Eighth gra ders attend class for four hours a day over seven weeks. Similar to Teach Baltimore, key c haracteristics of the Summer Bridge program are low class size and a remedially focused and highly structured curriculum. At Summer Bridge, students receive a great deal of per sonal attention, with the average class
49 size at only 16 students, as compared to 30 student s during the regular school year. Additionally, students receive even more support fr om tutors and aides. However, students and teachers at Summer Bridge know that th e students must pass the test given at the end of summer school in order to avoid reten tion. The study explored three questions: (1) How do students describe academic press and per sonalism in Summer Bridge? (2) To what extent do these descriptions differ fro m those of the school year? (3) How do these perceptions vary by student demogr aphic and performance characteristics? (p. 938) Results of the quantitative portion of the study (S tone et al., 2005) showed that on average, between 1997 and 2000, third graders gaine d about (.20) grade equivalents in reading, sixth graders gained (.40) and eighth gra ders gained (.80). Additional support was gathered both from surveys of students who had attended Summer Bridge in 1999 and semi-structured interviews with students who at tended Summer Bridge that same year. Results from the qualitative portion of the s tudy showed that 52 percent of the students had a positive experience overall. These s tudents touched on four themes: teachers covered more content and made the content easier to understand; teachers paced instruction and made sure the students understood; one-on-one time with the teacher was available; and skills were improving. An additiona l 35 percent of the students had a neutral experience, and the final 13 percent had a negative experience. Also, the researchers combined the quantitative dat a with the qualitative data to conclude that, Over half of the 48 students in the qualitative sample characterized their experiences as more positive in the summer than in the school year (p. 952). However,
50 it is difficult to generalize the findings of the s urveys to a larger population since the samples under represent African American students. The researchers state that this was because African American students were less likely to complete surveys. Additionally, it is difficult to judge the effectiveness of Summer B ridge, since data on how many students met promotion requirements following Summer Bridge is not presented. Rather, the researchers only offer data on students with whom t hey conducted the interviews, of those 48 students, 21 were promoted. Using a different approach to literacy instruction with summer school students, than that of Teach Baltimore or Summer Bridge, Duff y (2001), examined the effects of a balanced, accelerated, and responsive literacy prog ram on the reading growth of elementary school struggling readers by looking at 10 second-grade children enrolled in an elementary summer school program. Duffy (2001) a sserts that the purpose of the research was, To address the significant real worl d teaching problem of accelerating the reading growth of elementary school struggling read ers (p. 68). The study was conducted as a formative experiment with a mixture of quantitative and qualitative methods used. Duffy took on the role of the teacher and the researcher, and modified the program based on her students needs and progress. During the summer program, which lasted 30 days, Duffy used a variety of reading mat erials. There were 21 instructional days in the program and students attended summer sc hool from 9:00 a.m. to 12:00 p.m. Monday through Friday. Students received two and a half hours of instructional time each day, during which time students received instructio n in whole group reading and word sorting, individual reading and writing, book talks and read alouds, and small group instructional level support reading.
51 Results from analyzing the six categories that eme rged from the content analysis revealed that students demonstrated growth in six a reas of reading, as well as perceptions, positive attitudes toward reading, and increased in structional levels. The six areas where students showed growth were: word identification ab ilities, fluency, reading comprehension, self-perceptions, attitude towards r eading and instructional reading level. Like Summer Bridge, these results support the use o f summer school as an alternative to retention for the struggling reader. Results from the study showed that on average, students increased their reading levels on QRI passages and running records an average of 1.3 years in just 30 days. Additionally, through interviews these results support summer school as having a positive impact o n students attitudes toward reading. However, on a cautionary note, other factors may ha ve contributed to the success of the students. Teacher expertise may have been a factor in the results because the researcher, who also was a college professor, was t he summer school teacher. Duffy concludes her article by offering support for a balanced approach to teaching, as opposed to one that relies on a commercial reading program, like that used in Summer Bridge. Duffy purports: Rather than purchasing fixed, commercial reading p rograms and training teachers to use these programs, perhaps a better in vestment of school districts time and resources would be help teachers understan d how principles of balance, acceleration, and responsive teaching can be utiliz ed in multiple, purposeful ways in classrooms with struggling readers (p.92). Tab le 5 offers a summary of summer school research.
52 Table 5 Summer School Research Researcher/Year Variables Studied Sample Size/Ages Design Findings Austin, Rogers & Walbesser, 1972 Review of research from Title 1/ESEA n/a Review of findings Summer program showed gains in math and reading/did not persist over time. Heynes, 1978 Summer achievement 42 Atlanta Schools Grades: 4,5,6,7 Longitudinal Achievement gaps increase during summer. Carter, 1984 Compensatory education 120,000 student s from 300 schools over 3 years 5 separate studies No difference in performance of kids who attended summer school. Cooper, Nye, Charlton, Lindsay & Greathouse, 1996 Summer vacations effect on achievement scores n/a Narrative and MetaAnalytic Review Scores decline over summer/math and spelling effected most. Cooper, Charlton, Valentine and Muhlenbruck, 2000 Benefits of SS 93 studies Narrative and MetaAnalytic Review Middle class students benefited more from SS than disadvantaged. Duffy, 2001 Balanced Literacy in SS Second Grade (1 0 stud.) Multiple Methods QRI increased 1.3 yrs. Stone, Engel, Nagaoka & Roderick, 2005 Perceptions of summer learning environments Grades: 3, 6 and 8 Multiple Methods More than half students felt SS more positive than school year. Borman & Dowling, 2006 Multiyear SS and effects on summer slide 686 students high poverty schools Longitudinal HLM Describes effects of summer slide.
53 However, fixed, commercial scripted reading progr ams are exactly what many states are requiring in summer school. In the stat e of Florida, students attending summer school must receive intensive reading instruction f or a minimum of two hours per day using a research-based intervention program. In the targeted district the research-based intervention program used is Voyager Passport a scripted literacy program. Scripted Literacy Programs The purpose of this section of the literature revie w is to examine scripted literacy programs. First, the researcher will provide an ove rview of direct instruction and scripted literacy. Next, the author will present research on DISTAR, the father of scripted literacy. Finally, the author will present research on Voyager Universal Literacy Systems, the form of scripted reading that will be used during the pr esent study. Direct Instruction stresses basic skills and breaks them down into minicomponents. Additionally, Direct Instruction follow s a Bottom-Up approach to literacy instruction with children learning the sounds of th e letters before letters and words. The curriculum is fast paced with highly structured and scripted reading lessons. The stimulus response interaction between the teacher and studen ts is extremely important and requires that teachers ask 200-300 questions each day. The l essons are scripted, making each sequence predictable with little variation. The Dir ect Instruction Model is defined by Meyer et al., (1983) as having the following compon ents: a) a consistent focus on academic objectives; (b) h igh allocations of time to small group instruction in reading, language, and math; ( c) the tight, carefully sequenced DISTAR curriculum which includes a task analysis of all skills and
54 cognitive operations and numerous opportunities for review and practice of recently learned skills; (d) ongoing inservice and preservice training that offers concrete, hands-on solutions to problems arising in the classroom; and (e) a comprehensive system for monitoring both the rate a t which students progress through the curriculum and their mastery of the mat erial covered (p. 243). Although states and districts make choices about re ading instruction, under NCLB, the programs and materials must be based on s cientifically based reading instruction which is defined as rigorous, systemati c, and objective procedures to obtain knowledge about reading development, reading instru ction, and reading difficulties. Materials that are considered scientifically based reading research consist of curriculum that includes instruction in the five areas of read ing: Phonemic Awareness, Phonics, Vocabulary, Fluency, and Comprehension. Scripted li teracy is defined as reading programs characterized by very explicit teachers m anuals with instructions for teachers to follow verbatim when using the program with thei r students (Moustafa & Land, 2005). In a scripted classroom, all activities are to be followed in the order presented, and the teachers instructions are to be read word-for-word from the manual (Meyer, 2002). Scripted literacy programs can be traced back to th e late 60s to Siegfried Engelmann and Wesley Becker. Project Follow Through began in 1967 and continued until the summer of 1995. Head Start began in the s ummer of 1965. The purpose of Project Follow Through was to follow through on H ead Start and help children from kindergarten through third grade continue the progr ess they had made in breaking the cycle of poverty through better education. Although Project Follow Through was initially conceived as a comprehensive social services progra m, before the program got underway
55 budget cuts forced a re-conceptualization. Thus, Pr oject Follow Through was converted to a longitudinal experiment aimed at finding effec tive methods for teaching disadvantaged children. Project Follow Through invo lved 120 communities and 10,000 children each year from 1968 to 1976. It continued as a service program until funding was eliminated in 1995. One of the models developed and implemented under Project Follow Through was the Direct Instruction System fo r Teaching Arithmetic and Reading (DISTAR or Direct Instruction). DISTAR has been found to be successful when working with disadvantaged students (Kuder, 1990; Meyer et al., 1983). Sexton (2001) compared DISTAR (Engelmann, Haddox & Bruner, 1984) to a basal readi ng program as a way to increase language ability and reading comprehension. Partici pants included 40 first grade students who were all African American. The effectiveness of the program was measured by the Slosson Intelligence Test. Results of the study rev ealed that students using DISTAR earned a score on the Slosson Intelligence Test tha t was 9 points higher than the average of the basal group. Additionally, the researchers c oncluded that the DISTAR program was equally effective with students of low language ability as those with high language ability. On a cautionary note, it is difficult to g eneralize the findings of the study to other populations since all of the participants were Afri can American. Finally, although the research study was attempting to measure the effect iveness of DISTAR, the tool used to measure growth was actually an intelligence test. I t may have been more effective to use an assessment tool that measured language ability a nd reading comprehension. A similar study by Kuder (2001) compared the effect iveness of DISTAR to a basal reading series when working with children wit h learning disabilities. Once again
56 the researchers compared the effectiveness of DISTA R to a basal reading program. The participants were 48 students identified as learnin g disabled representing 3 urban schools. The children were in two different classes. One cla ss was taught using DISTAR, while the other class used a basal to teach reading. Like the Sexton study, this study compared the results of the experimental group to those of t he control group. After seven months of training, the results from the study showed that in reading subtests the DISTAR group performed better on word comprehension and word att ack, while the basal group scored better on letter identification. However, the resea rchers reported that there were no statistically significant differences between the t wo groups. During the present study, students attending summer school received instruction using Voyager Passport It shares many characteristics with scripted lit eracy programs. Voyager Universal Literacy System began as an after school tutoring program aimed at struggling readers. Voyager has experienced great success under No Child Left Behind. In fact, the company went from being worth 5 millio n before Reading First to a net worth of over 350 million dollars in 2005. Voyager Universal Literacy Systems is the umbrella under which Voyager Passport comes under. Although there have been some studies conducted that examine Voyager Universal Literacy Systems (Frechtling, Zhang & Si lverstein, 2006; Roberts & Alan, 2003; Hect & Torgesen, 2002), little research has been done yet using Voyager Passport However, many schools are using Voyager Passport as a reading intervention with struggling readers. These interventions take p lace in small groups during the school year, as well as with students attending summer sch ool. Voyager Passport is an approved
57 supplemental reading program under Reading First and falls under the category of scripted literacy programs. Voyager Passport is described under Reading First as a K-3 reading intervention that is grounded in scientifically based reading re search. In an effort to get students on grade level in reading, the goal of Voyager Passport is to accelerate students reading. Voyager Passport lessons are designed to be taught explicitly at a quick pace in a small group setting. Voyager Passport lessons last between 30-45 minutes. Instruction us ing Voyager Passport is designed to be delivered five days a week. Duri ng a Voyager lesson every minute of instructional time is structured. A ll of the lessons in each reading component provide explicit instruction on every ste p of the reading process, with teacher modeling followed by multiple practice opportunitie s. Third grade lessons consist of instruction in two modules. The first module focuse s on comprehension and vocabulary, while the second module focuses on fluency. An additional component of Voyager Passport is ongoing progress monitoring. Progress monitoring is defined as, a process of ev aluating individual student reading progress between benchmark periods in order to make instructional decisions. Voyager uses both choral and individual student responses i n an attempt to add extensive practice for all students. According to the publisher, Voyager Passport provides a complete reading intervention program to give struggling rea ders the tools they need to read on grade level (Voyager Expanded Learning, 2004). For a detailed summary of a typical daily lesson refer to Appendix C. Voyager assesses using Vital Indicators of Progress (VIP). VIP is an alternative form of Dynamic Indicators of Basic Early Literacy Skills (DIBELS). DIBELS is defined
58 as a set of standardized individually administered measures of early literacy development. They are designed to be short (one minute) fluency measures used to regularly monitor the development of pre-reading and early reading sk ills ( www.dibels.uoregon.edu ). DIBELS was largely unknown before Reading First, ye t DIBELS is now the primary assessment tool promoted by the Department of Education under Reading First. DIBELS measures initial sounds fluency, phoneme seg mentation fluency, nonsense words fluency, and oral reading fluency. Initial So unds Fluency assesses a child's skill to identify and produce the initial sound of a given w ord. Phonemic Segmentation Fluency assesses a child's skill to produce the individual sounds within a given word. Nonsense Word Fluency assesses a child's knowledge of letter -sound correspondences as well their ability to blend letters together to form unfamilia r "nonsense" (e.g., fik, lig, etc.) words. Finally, Oral Reading Fluency assesses a child's sk ill of reading connected text in gradelevel material. ( http://dibels.uoregon.edu/dibels_what.php ). Ken Goodman describes himself as a practical theori st, researcher and teacher educator whose work has centered on literacy proces ses, how they are learned, and how best they can be taught. Goodmans socio-transactio nal theory of the reading process demonstrates that reading is a unitary process in w hich readers actively construct meaning, that is they make sense of print. Goodman has written a critical review of DIBELS (2006). According to Goodman (2006), The tests reveal that competent reading is the abi lity to read words rapidly, accurately, and that comprehension is the result of such rapid, accurate reading. They also believe that what happens in one minute o f reading happens in all of reading. Its likely that they do not explicitly st ate their definition of reading
59 because they dont see any need to define reading s ince they have not considered that there could be any other definitions (p.9). An additional component of VIP is VPORT, an online data management system that allows teachers to analyze class data and comp are student data against a trajectory of desired learning. In 2003 the Texas legislature ma ndated an appropriations bill to spend 12 million dollars on a single intervention program for struggling readers. All districts had to use the one program chosen by TEA (Texas Edu cation Agency) or pay for their own ( www.edu.cyberpg.com ). Although there are intervention studies underway, b ecause Voyager Passport is a relatively new program that has only been in existe nce since 2003, the researcher was unable to find any research that was specific to th e intervention. Furthermore, although there is research on The Voyager Universal Literacy System much of the research is not longitudinal. The research which is provided by Voyager was conducted by researchers who were directly associated with Voyager ( www.edu.cyberpg.com ). Further, in many cases the key researchers in each of the studies ar e in some way connected with Reading First. The Voyager website provides four different categories of rese arch: scientific research studies, independent impact studies, white papers and stories, quotes and testimonials. Voyager states that the scientific research studies were, conducted by nationally renowned researchers using quasi-experim ental and comparative designs ( www.voyager.com ). A study which lasted eleven weeks was conducted in 2002 and evaluated the effectiveness of The Voyager Universal Literacy System. The participants were 108 economically disadvantaged kindergartners (Hect & T orgesen, 2002). During the study,
60 58 students were given The Voyager program as a part of their school day. An additiona l 50 students made up the control group. These studen ts were not given the Voyager curriculum. Student performance was measured by a n umber of tests that measured phonemic decoding ability, letter sound knowledge, print concepts, phonemic segmenting, and phonemic building systems (VIP). Pr etests were given in February and posttests were given in April and May. Results from the study showed that students receivi ng instruction in the Voyager Universal Literacy System made larger gains from pretest to posttest in all a reas, except word identification and spelling letters, which sta yed the same. On a cautionary note, when looking at the study participants the control classrooms had 10 more students that were limited English proficiency than the Voyager classrooms. Thus, the classrooms may not have been equally matched. The findings of a second study (Roberts & Alan, 200 3), used data from 865 first grade and kindergarten students from 13 schools in Virginia. The nine schools that used Voyager during the 2001-2002 school year were all describe d as low achieving and lowincome schools. An additional four high performing schools used an alternative reading program. Performance was measured using Phonologica l Awareness Literacy Screening (PALS). Results from the study demonstrate that onl y 21% of kindergartners and 29% of first graders began the year on grade level. Howeve r, by years end 70 % of kindergarteners and 68% of first graders were on gr ade level as measured by a Phonological Awareness Literacy Screening. The researchers state, There were no changes from fall to spring for children attending nonVoyager schools Although the nonVoyager schools are described as
61 high performing, the results showed that 87% of t he kindergarten and 79% of the first graders (in the nonVoyager schools) were on grade level at the end of the yea r. However, because there is no equally matched compar ison group, it is not clear why the scores of the high performing schools students a re included in the study, since they did not receive Voyager training. The researchers state that the high perf orming schools were included to demonstrate that the gap between kinder garten and first grade students attending high to average schools in the sample was narrowed significantly over the course of the 2001-2002 school year. The researcher s also state, Children attending Voyager classrooms made large gains that they would probab ly not have made if Voyager had not been part of their school experience. On a cautionary note, one would expect students to progress during the school year no matt er what reading program was used. The study lists Greg Roberts, Ph.D. as the Program Evaluation Consultant. Since the report gave no background information on the au thor, a Google Search was conducted to locate more information. The findings were inter esting, given the recent controversy with Reading First The results indicate that Dr. Roberts is the prin cipal investigator and director of the Special Education Strand of the Cen ter on Instruction. Additionally, Roberts is connected with Reading First, as co-dire ctor of the Central Center for Reading First Technical Assistance (CCRFTAC). Furthermore, he has a Texas connection as director of dissemination for the Texas Center on L earning Disabilities. Considering the recent controversy surrounding Reading First, it ma y be a conflict of interest that Roberts was the supervisor of the Voyager Universal Literacy System research. A third study, also examined Voyager Universal Literacy Systems Like the previous study, this study also lists Dr. Roberts a s the program evaluation consultant.
62 This study looked at the data of 16, 443 students e nrolled at 291 schools across the U.S. Data were collected using the Vital Indicators of P rogress system (VIP). VIP is defined in the study (Roberts, 2002) an alternative form of DI BELS that is, a standardized, individually administered test of accuracy and flue ncy with connected text (p. 6). Once again the researchers compared kindergarten student s in Voyager classrooms to children in nonVoyager classrooms. The results of this study were also di fficult to interpret; therefore it is difficult to form conclusions from the data given. Although the researchers are open and up front regarding concerns with inter nal validity and state that the scores were positively skewed at all three time points, th e researchers still conclude that most of the first grade students achieved benchmark status and purport that questions related to sustainability of effects will be addresses in subs equent studies. The researchers state that at that time, student and school level data will be more accessible. A final study (Frechtling et al., 2006) examined 39 8 kindergartners representing 4 Voyager schools and 4 comparison schools to determine the efficacy of The Voyager Universal Literacy System Like the other studies this study focused on Voyager Universal Literacy System which is the umbrella that Voyager Passport comes under. There were three parts to the quasi-experimental st udy. The first part compared the performance of kindergartners in Voyager Universal schools to those in comparable nonparticipating schools. The second part of the resea rch looked at how the level of implementation affected student achievement. The la st part of the study looked at the effectiveness of the program with students from dif ferent backgrounds, gender, race, ethnicity, economic status, and English language sk ills.
63 When the researchers returned in the spring they co nducted site visits at both the Voyager Universal schools and the comparison schools. During these v isits they interviewed teachers and principals to gather data on the general classroom environment, the presence of other educational reforms in the sc hools, parental involvement, students pre-kindergarten exposure to reading, and principal s assessments of the general strengths and weaknesses of the program. Researchers suggeste d that both the Voyager Universal and nonVoyager schools were comparable in all but one way. The re searchers found that the teachers at the Voyager Universal schools were spending 90-120 minutes a day on reading, whereas the non Voyager classrooms were spending less time on reading (6090 minutes). However, the researchers note that all te achers seemed to be integrating reading in other subjects. Results from the 8-month study revealed that in thr ee out of the four schools examined, a significant difference was found in fav or of the Voyager Universal students. The seven test instruments used to measure growth w ere DIBELS letter naming fluency, CTOPP Ellision, CTOPP Blending Words, CTOPP Segment ing Words, Woodcock Word Identification, and Woodcock Word Attack. All of th e students were assessed in the fall prior to the intervention using the above mentioned literacy assessments and no differences were found between the control or treat ment groups. Following the intervention, the researchers used a paired-sample t-test to analyze the results from pretest to posttest at the Voyager Universal schools. The results revealed effect sizes ranging from 1.51 (CTOPP Elision) to 8 .3 (Woodcock Word Attack) in 7 test instruments at the p<.01 level. This can be interpr eted as the Voyager Universal schools gaining from 1.51 to 8.3 standard deviations in one school year. Although these effect
64 sizes are large it is important to note that some o f the gains may be explained by the natural developmental growth of the kindergartners, rather than just the program. When the Voyager Universal schools were compared to the comparison schools an independent sample t-test was used. The gains in the Voyager Universal schools were significant with an average effect size of 0.62. The researchers con cluded that three out of the four Voyager Universal schools outperformed their comparison schools. Res earchers stated that the Voyager Universal school that did not show a significant difference had inadequate implementation and that some teachers fa iled to use parts of the curriculum or substituted materials that were not part of the cur riculum. Researchers then developed an ANCOVA model to asses s the effectiveness of the program. Using the model the dependent variable was the gain score for each student. The main independent variable was the Voyager Universal program. Control variables included gender, class size, teacher experience, an d percentage of free and reduced lunch. Although the researchers looked at race, LEP, IEP, attendance rate, and student mobility rate, these variables were excluded from the model because of a lack of variability or missing data. Researchers measured implementation effects by usi ng the Voyager Universal Instructional Fidelity Checks, and making an ANCOVA model. This time the main independent variable was implementation score. This score was determined from the final implementation score from the Voyager Universal Fidelity Measure. They grouped the scores into three clusters, high (10-12), medium (7 -9), and inadequate (0-6). Using implementation as an ordinal value, the researchers found that the level of
65 implementation had a positively significant effect (p<.05) on student achievement on all seven assessments. On a cautionary note, when reviewing the available research on Voyager it is of interest that much of the research on Voyager Universal Literacy Systems uses kindergarten and first grade students (Hecht & Torg esen, 2002; Roberts, 2002; Roberts & Allen, 2003), when students reading development sh ows the most growth. The researcher was not able to locate any research stud ies using second or third grade students. However, there are 41 independent Impact Studies on the Voyager website which report the findings of individual schools and the results they had with Voyager These studies, which appear in a summarized format, offer a brief one page report which highlights the findings. One of these independent i mpact studies is specific to the targeted county summer school students. According to an inde pendent impact study on the targeted county completed in the summer of 2005, 26 3 third-graders attending summer school received the Voyager Passport intervention. After six weeks in the summer school program, 72 percent of the third-graders passed the SAT 9 and were promoted. The results are compared to the previous summer when on ly 27 percent of the students achieved proficiency on the assessment.
66 Table 6 Summary of Research on Voyager Universal Systems Researchers Variables Studied Sample Size/Age Desig n Findings Hect & Torgesen Reading Performance 108 kindergartn ers Pretest/Posttest Treatment group on grade level after Voyager Greg Roberts Gains in Reading Fluency 16, 433 students Longitudinal, Pretest 1st graders gained 30 WPM Roberts and Allen Phonological Awareness 864 K-1 st udents Pretest/Posttest/Control Group 70% K and 68% 1st on grade level WESTAT Reading Achievement using Voyager 255 Kindergartners Pretest/Posttest/Control Group Voyager students showed greater gains than comparison group. WESTAT Year2 Year 2 Voyager Intervention Same students as prior year now 1st graders Pretest/Posttest/Control Group Voyager students showed greater gains than comparison group. Frechtling, Zhang & Silverstein Reading Performance 398 kindergartners Pretest/Post test/Control Group Voyager schools scored higher than comparison schools; Level of implementation contributed to gain in reading scores.
67 Although under Reading First, scripted literacy is being used as an intervention to assist the struggling reader, not everyone supports scripted literacy in the classroom. Allington (2002) summarizes his view on scripted li teracy with this statement, A veritable trove of scientific research tells us tha t effective teaching is not standardized and cannot be scripted (p.28). Allington agrees th at the five pillars of scientific reading instruction set forth in the National Reading Panel Report (2001) are critical aspects of reading acquisition (phonological awareness, phonic s, comprehension, fluency and vocabulary). However, Allington purports that ther e are an additional 5 pillars missing from the National Reading Panel Report. These inclu de: access to interesting text and choice; matching kids with appropriate texts; writi ng and reading; classroom organization; balancing whole class teaching with s mall group and side-by-side instruction; and expert tutoring availability ( http://teachersread.net/pdf/FivePillars.pdf ). There are alternative ways to provide supplemental tutoring to struggling readers in the classroom other than relying on the use of s cripted programs. Taylor, Short Shearer and Frye (1995) examined how first grade te achers could work with their lowest achieving readers to provide effective early readin g intervention in the classroom. The goal of EIR (Early Intervention in Reading) was to accelerate the learning of the lowest readers by providing them with an additional 20 min utes of reading instruction by the classroom teacher. Teacher training consisted of ha lf-day workshops at various times throughout the year. Using EIR, instruction was don e in small groups utilizing books at the childrens reading level. Teachers implemented a variety of reading techniques with the assistance of a part time resource teacher. The role of the resource teacher was to provide feedback and suggestions to the classroom t eacher as well as to assist with
68 dissemination of materials. Some of the reading te chniques utilized during the lessons included read alouds, oral retellings, spelling usi ng elkonin boxes, paired reading, choral reading, writing, and phonological awareness activi ties. The teacher provided scaffolded support during the lessons with the ultimate goal b eing to create independent readers. The first year the program was implemented the rese archers found that 72% of the students were on grade level by the end of second g rade. The next year the program was implemented district wide in both first and second grade classrooms. Results revealed that 78% of the children were reading at least at a pre-primer level and 36% of the students were reading on grade level or better. Reading Attitude The purpose of this section of the literature revie w is to examine the many factors that contribute to the reading attitudes of struggl ing readers. Presently, under Reading First, many struggling readers are receiving scripted lit eracy as an intervention. McKennas model states that the factors that contri bute to reading attitude are not only complex, but are subject to change and influence on e another as well as attitude (McKenna et al., 1995). McKenna (1994) postulates t hat based on his model one can predict that, Certain instructional approaches may produce more successful experiences contributing directly and cumulatively to attitude, and they may also lead to more positive beliefs about the outcomes of reading, con tributing to attitude indirectly (p.939) Reading attitude can be also be seen as one part of a broader construct, motivation to read (Sainsbury & Schagan, 2004). Guthrie and Wi gfield (2000) describe five aspects
69 of motivation. These include: learning orientation (understanding the content of what is read), intrinsic orientation (enjoyment of reading/ disposition to seek out activities, selfefficacy, and social motivation). Reading attitude the continuum of positive or negative feelings toward reading, plays an important role on both the level of ability attained by a given student and through its influence on reading engagement and practice (McKenna et al., 1995; McKenna & Kear, 1990). Likewise, a poor reading attitude may contribute to aliteracy, a condition when fluent readers choose n ot to read when other options exist (McKenna et al., 1995). The relationship between reading ability and attitu de has been explored by several researchers (Askov & Fishback, 1973; Walberg & Tsai 1985). Using multiple regression, reading achievement and attitude scores of a Nation al Assessment of Educational Progress (NAEP) were compared to home environment v ariables such as amount of television watched, presence of newspapers, spare t ime reading, dictionary use, kindergarten attendance, socioeconomic status, ethn icity, school characteristics, and other variables (Wahlberg & Tsai, 1985). Using a sample o f 1,549 nine-year-old students the researchers found that variance in reading achievem ent and attitude could be accounted for by home environment, quality of instruction, an d leisure-time television watching. Canonical correlation of reading achievement and at titude with the independent and control variables is .48, which is significant at t he .001 level. This correlation is highly significant and shows that the relationship between the two sets is very likely not to have occurred by chance. An international survey was conducted in England to determine if childrens attitudes about reading had changed over the five y ear period between 1998 and 2003.
70 Results revealed that although students were perfor ming well in relation to their peers in other countries, the attitudes toward reading held by English children were lower than those of children in other countries (Twist, Gnaldi & Morrison, 2004). A comparable study focused on the reading attitudes of upper pri mary pupils in The United Kingdom (Sainsbury & Schagen, 2004). The study presents res ults of a survey given to 5,076 nine and eleven year olds. Results presented in the sur vey displayed similar results to the findings by Twist et al. (2004), showing that while the students reading confidence increased, their enjoyment of reading declined. Swanson (1982) administered a reading attitude surv ey (Heathington, 1975) to 116 first graders in northeastern Georgia and corre lated the findings with the students reading scores on the Metropolitan Achievement Test (Durost, Bixler, Wrightstone, Prescott, and Balow, 1970). Findings from the resea rch revealed a correlation of .18 (p < .05). This means that only 3% of the variance in achievement scores can be accounted for by scores on the reading attitude inv entory. Findings from the study indicated that children had positive attitudes towa rd reading in the initial stages of learning to read and that negative attitudes did no t surface until reading became more of a task. Kush and Watkins (1996) offer support for these fi ndings and suggest that attitude towards reading declines as children grow older. The researchers tested the longterm stability of childrens attitudes toward readi ng by administering The ERAS with 189 elementary age students (grades 1-4). The researche rs administered the survey two times over a three-year period. Following the three years reading attitudes in both academic and recreational scores dropped significantly. Addi tionally, the results of a 2x4 factor
71 analysis of variance showed that girls had a more p ositive attitude about reading than boys did. It is interesting to note that as childre n grow older they make the transition between reading to learn and learning to read. Its possible that this may impact their attitude about reading, and account for the decreas e in reading attitude as children get older. Although some studies suggest that teaching techni ques can influence reading attitudes, it is difficult to substantiate this cla im. Researchers have looked at the effects on students attitudes when high quality literature was used and found positive effects (Morrow, 1983). Additionally, although researchers have undertaken the chore of examining basal readers and the effect they have on students attitudes (McKenn et al., 1995), there are no conclusive findings. Other stud ies suggest that although classroom teachers see attitudes toward reading as important, most teachers spend little time fostering childrens attitudes (Heathington & Alexa nder, 1984). Principle results from a national survey conducted in The United States to determine childrens attitudes about reading reveal ed the following findings (McKenna et al., 1995): (a) Recreational and acade mic reading attitudes begin at a relatively positive point in Grade 1, but end in re lative indifference by Grade 6; (b) negative recreational attitude is related to abilit y and the trend is most rapid for least able readers; (c) gender differences favored girls posi tive attitudes toward reading; (d) ethnicity played little role in reading attitude; a nd (e) Teachers reliance on a basal reader did not appear to play a role in reading attitude ( p. 951). These findings offer support for the McKenna model (1994) on reading attitude acquisition. Because the findings show that as chil dren grow older their attitude about
72 reading declines, this suggests the importance of a ssisting the struggling reader at an early age. The researchers conclude that, the grea test potential for further research lies in the matter of instructional techniques (p. 953). T he researchers findings that the use of a basal did not appear to play a role in reading atti tude, does offer support for other methods of teaching reading. How the Review of Literature Informed the Study From the review of research, the following conclus ions were drawn. Retention has not been shown to be an effective way to assist the struggling reader (Jimerson, 2001; Jimerson & Kaufman, 2003; Parker, 2001). Furthermor e, retention has been proven to have detrimental effects on the retained student (P arker, 2001). Yet in light of this, retention continues to play an important role under NCLB as an intervention to assist the struggling reader. Although there is research to su pport different instructional techniques and approaches used by effective summer school prog rams (Stone et al., 2005; Duffy, 2001; Borman & Dowling 2006), under NCLB students m ust attend summer school using a scientifically based reading research (SBRR) appr oved reading program. In the targeted district the approved intervention program used in summer school is Voyager Passport The recent emphasis on reading performance as defin ed by performance on the FCAT examination has ignored the important role tha t childrens attitudes play in the process of becoming literate. When considering McKe nnas (1994) model of reading attitude, the decision to read or not to read is ul timately determined by three factors: the expectation of others; both physical and time const raints as well as competing options; and the desirability of reading outcomes. Unfortuna tely, research on attitudes has shown
73 that students attitudes about reading have been sh own to decrease with age (Sainsbury & Schagen, 2004; Swanson, 1982). Third-graders in Florida who earn a Level 1 on the FCAT Reading Test are retained. These retained students are then encourag ed to attend reading summer camps, or summer schools. Summer school is now the last g ood cause intervention available to these third grade students. Although summer school is being used as an alternative to retention, and research has examined different summ er school models, very little research has been conducted on the use of scripted literacy programs with third-grade retained students. Additionally, there is a need for research that exp lores the impact different instructional methods have on childrens attitudes about reading. Because the researcher did not find any research that examined the impact that a scripted literacy summer school had on the attitudes of third-grade struggling read ers, there was a need for further research in this area. Additionally, there is very little research on The Voyager Passport program as a reading intervention. However, under Reading First legislation many districts nationwide are currently using the progra m with struggling readers. All of these findings revealed a need for research focusing on t he reading attitudes of retained thirdgraders during summer school using The Voyager Passport program. Organization of Remaining Chapters The topic of Chapter 3 is methodology. This chapter begins with information on the Voyager Passport training and the instructional fidelity measure. N ext, descriptions of the design of the study, the population and samp le selection, instrumentation, data collection, and the manner in which the data were a nalyzed and interpreted are presented.
74 Chapter 4 summarizes the findings of the study. Bot h quantitative and qualitative findings are reported. Chapter 5 presents a summary of the study, conclusions and implications derived from the research findings, re commendations for practice based on the study conclusions and implications, and recomme ndations for future research.
75 CHAPTER 3 METHOD Overview of Chapter The purpose of this mixed method study was to deter mine the effect a summer school literacy program had on the reading attitude s of elementary school struggling readers. This chapter describes the methodology use d to conduct the research and contains eight sections. The first section describe s the design of the study. The second section describes the Voyager Passport training. The third section describes the population and sample selection for the study. The fourth section includes a discussion of the Pilot Study, the validity of the instrument, an d interrater reliability. The fifth section describes instrumentation used in the study. The s ixth section describes data collection. The seventh section explains the manner in which th e data were analyzed and interpreted. The final section contains Evidence of Ethical Cons iderations. Design The intent of this mixed study was to address the f ollowing two research questions: 1. What is the effect of a scripted literacy program o n the reading attitudes of elementary school struggling readers? 2. What do elementary school struggling readers percei ve to be the effect of a scripted summer school literacy program on their at titudes toward reading?
76 This design of this non-experimental, longitudinal, mixed method study includes both quantitative and qualitative methods, dependin g on the question being analyzed. Because random assignments to groups were not possi ble, and because there was no manipulation of an independent variable, the study is considered non-experimental (Johnson & Cristensen, 2004). Although data were ga thered at multiple points in time, data collection only lasted six weeks making the st udy short-term longitudinal. Depending on the purpose for mixing methods, there are different purposes for mixed method designs. (Greene, Caracelli & Graham, 1989). Greene et al., (1989) offer this theoretical base for understanding triangulati on as a study design: Triangulation refers to the designed use of multipl e methods with offsetting and counteracting biases, in investigations of the same phenomenon in order to strengthen the validity of inquiry results. The cor e premise of triangulation as a design strategy is that all methods have inherent b iases and limitations, so use of only one method to assess a phenomenon will inevita bly yield biased and limited results (p. 256). Because the purpose for conducting this mixed-metho ds design was to seek corroboration from the results of the quantitative attitude survey, the classroom observations and the qualitative interview, the res earcher selected triangulation as the design of the study (Greene, et al., 1989; Lincoln & Guba, 1985; Onwuegbuzie, 2002). Triangulation was achieved through the use of quant itative data from the ERAS surveys and qualitative findings from the focus groups and classroom observations.
77 The dependent variable for this study was students attitudes toward reading. The independent variable was Voyager Passport The hypothesis was that time spent in a scripted literacy summer school program would affec t the attitudes of third grade struggling readers, specifically that their attitud e about reading would improve. The researcher expected this change in just six weeks t ime because of the intensity of the intervention. During summer school the students rec eived two 30 to 45 minute lessons daily of Voyager Passport Instruction. Over the course of the 30 days the ch ildren were in summer school they actually received between 180 0 and 2700 minutes of the intervention which equates to 12 weeks of instructi on. This equivalence was determined by multiplying the number of minutes times the numb er of intervention days. Mixed method research is defined as, Research in w hich quantitative and qualitative techniques are mixed in a single study (Johnson & Christensen, 2004; p. 410). Because both quantitative and qualitative met hods are used in this study, the study takes on a mixed design. Tashakkori and Teddlie str ess the importance of, mixing methods in a way that has complementary strengths a nd non-overlapping weaknesses (Tashakkori & Teddlie, 2003; p. 299). Based on this in an effort to answer both research questions, the researcher collected data concurrent ly. The first question, What is the effect of a script ed literacy program on the reading attitudes of elementary school struggling r eaders? is quantitative in nature. Data were collected from The Elementary Reading Attitude Survey (McKenna & Kear, 1990). This question was analyzed using descriptive statis tics, a dependent measures t -test and three factorial ANOVAs. The three dependent variabl es for the factorial ANOVAs were
78 recreational reading attitude, academic reading att itude and total reading attitude. The independent variables were gender and school site. The second question, What do elementary school str uggling readers perceive to be the effect of a scripted summer school literacy program on their attitudes toward reading? is qualitative in nature. The qualitative analysis was completed using findings from focus groups as well as field notes gathered d uring classroom observations. The qualitative analysis completed following focus grou ps was done using a stance of objectivist grounded theory, allowing the students responses to define the categories used in the analysis (Charmaz, 2000). The results o f the classroom observations were coded using a priori codes and analyzed for pattern s. The findings from the classroom observations are presented first as vignettes repre senting each classroom. Next, a cross case analysis was completed to find patterns throug hout the different school sites. Following quantitative and qualitative data collect ion, the researcher triangulated the findings and formed conclusions and recommendations Voyager Passport Training In preparation for the study, the researcher review ed the Voyager Passport materials available for principals and teachers and attended the training the summer school teachers attended on Voyager Passport The researcher had access to all of the training materials provided by Voyager Passport. The summer school training took place the Friday before summer school was to begin and la sted four hours (see Appendix D). The training was facilitated by the summer school c oordinator for the district and three representatives from Voyager Passport During the four hour training each teacher
79 received a Teachers Resource Kit and the materials necessary to implement Voyager Passport in their summer school classroom. The resource kit consisted of a box with a curriculum guide on fluency, comprehension and voca bulary, and targeted word study. There was also a VIP assessment guide, a packet of benchmark assessments, and test prep masters. Additionally, each kit included a stop wat ch. Although there were 35 teachers in the initial trai ning, in the end a total of 29 teachers participated in the study. Some of the tea chers were released due to lower than expected enrollment. Another teacher was absent for the initial training and never received the materials. A fifth teacher was absent at the end of the study and was unable to administer the post assessments. During the training the third-grade summer school t eachers received specific guidelines and instructions on how to structure the ir summer school day as well as how to implement Voyager Passport in their summer school classrooms. The teachers al so received tips and suggestions for implementing the program with enthusiasm and fidelity. This was done utilizing a lecture style format whic h was structured around a Power Point presentation. During the training maintaining stude nts enthusiasm during Voyager Passport lessons was stressed. Video clips were shown that showed teachers implementing the program with enthusiasm and others who lacked enthusiasm and displayed very flat affect. Teachers were then give n time to analyze video clips for strengths and weaknesses in regards to the fidelity of the implementation. Following this the teachers had time to discuss what they had seen with peers. Then the Voyager
80 consultant again spoke of the important role that e nthusiasm would play in the implementation of the program. There were differences in the knowledge of the clas sroom teachers in regards to the Voyager Passport program. Many of the summer school teachers had us ed Voyager Passport in their classrooms during the school year or the previous year in summer school and therefore were familiar with the program However, others had never used Voyager Passport Regardless of previous experience everyone attend ed the same training. Some of the teachers in attendance voiced their frustration at having to attend a training they did not think was necessary. The district allowed a portion of the training to b e devoted to the researchers study. This meant that during the training the teac hers were able to meet the researcher and receive an overview of the researchers study a nd training on how to administer the Elementary Reading Attitude Survey (McKenna & Kear, 1995). The researcher began by presenting an overview of her study, during which t ime she passed out materials and trained the summer school teachers on how to admini ster the Elementary Reading Attitude Survey (ERAS). The training followed the E RAS Teacher Training Protocol (see Appendix E). Overall, the teachers were receptive to administering the reading survey. The researcher went over the ERAS group protocol wi th each of the teachers (see Appendix F). Each of the teachers in attendance agr eed to participate in the study. The district recommended that the ERAS be administe red to all of the thirdgraders attending summer school at the five partici pating sites. This recommendation was made because the district administrators felt that giving the survey to all of the students
81 would be less confusing than only giving the survey to specific students. In regards to data collection, this meant that the survey was giv en on the first day of summer school first thing in the morning. Therefore, during the t raining the researcher provided each teacher with a set of surveys (ERAS) and informed c onsent forms (Appendix B). She explained that although the surveys would be given to all the students, data would only be collected from those students who returned a yes in formed consent letter. Next, the researcher explained how focus group participants w ould be selected and that focus group participants would be observed on Wednesdays at var ying times. Finally, the researcher allowed time for questions. There were many questio ns concerning where the teachers could obtain copies of the survey to use in their o wn classrooms during the school year as well as how to analyze the findings. Following the Voyager Passport training the teachers returned to their summer school site where they met with the acting principa l of the site. In all there were five summer school sites that participated in the study. Each of the summer school teachers received additional reading resources in the form o f books for independent reading time and big books for shared reading lessons. These inc luded leveled books, Harcourt Trophies books (Harcourt, 2004), big books and Elem ents of Vocabulary (Beck, 2005). Additionally, the teachers were able to meet the Re ading Coach and ELL teacher at each site. Both of these teachers assisted the classroom teachers during summer school. Following that meeting the teachers were given time to work in their classrooms and prepare for Monday morning when the students would arrive.
82 The elementary schools from the district were broke n into six clusters, which represented the six sites. Which site children atte nded for summer school depended on where their home school was located. Generally, chi ldren attended summer school at the site closest to their house. The researcher only us ed five of the six sites in the study. One of the sites was located in a rural area outside of the town. Due to low enrollment at this site, the district recommended that the researcher just use five sites. The acting principals of the summer school sites were actually assistant principals from the district. Two assistant principals shared each site and decided w ho would work what weeks. Therefore, in most situations the assistant principals systema tically changed midway through the study. Summer school in the targeted county followed the g uidelines set by the state. Students attended summer school five days a week fo r a total of five hours each day. Class size varied from 10 to 12 students. According to state guidelines, intensive reading instruction was to last a minimum of two hours of t he total instructional day. Additionally, the state recommended that summer sch ools not exceed a teacher to student ratio of one to twelve. Each of the summer school c lassrooms the researcher visited met these requirements. Regarding instructors, the state recommended that c ounties hire teachers who had successful teaching experience as well as reading c ertification or endorsement. An additional recommendation from the state was that c ounties involve mentors in their summer reading camp as a way to reinforce reading s kills and to enhance a students selfesteem. However, the state stipulated that although mentors could provide one-on-one mentoring for a student in the classroom, students were not allowed to leave the
83 classroom during reading instruction ( http://www.justreadflorida.org/camps/ ). During the observations at the different summer school sites t he researcher observed mentors working with students both one on one and in small groups. Additionally the researcher observed reading coaches working with small groups of students and ELL teachers assisting the classroom teacher. Population and Sample Selection The school district from which the sample was drawn encompassed a county on the west coast of Florida with a population of over 313,298. There were 33 elementary schools in the district during the summer of 2007. Although summer school was held at six different sites, only five sites were used in t his study. Students attended one of the summer schools based on which cluster their home school was located in. Determination as to which summer school site studen ts attended was made by the district office who arranged the elementary schools into six different clusters. A summer school site was then designated for each group. Therefore, although it was likely that the summer school site the students attended was not th eir home school, most likely it was the site closest to their home. The convenience sample was limited to third-grade s tudents who earned a Level 1 on the FCAT reading test and were attending summer school. The population the researcher made inferences about for both the quant itative and qualitative portions of this study were retained third-grade struggling readers who scored a Level 1 on the reading portion of the FCAT test in 2007. There were 336 th ird-grade students who attended summer school. Complete data were collected on 91 o r 27 percent of the students. The low return rate might have been due to a lack of pa rental involvement which is
84 characteristic of retained students (Jimerson et al ., 1997; Jimerson & Kaufman, 2003). Of the 336 students, 285 students completed summer sch ool and participated in Stanford 10 testing the last day of summer school. Retained students frequently have the following cha racteristics in common: low parental IQ (Jimerson et al., 1997), lack of parent al involvement, are boys, and are minorities. In addition, retained students are like ly to have missed a greater percentage of school days than their peers who have been promoted (Jimerson & Kaufman, 2003). The decision to use third-grade students was made becau se of the current political climate supporting retention as an intervention with third grade struggling readers. The purpose of the qualitative portion of the study was to use focus groups as well as field notes collected during classroom observati ons to gain a more in depth understanding of childrens attitudes about reading and to compare these findings to the quantitative findings. When selecting participants for the qualitative portion of the study, the researcher used a nested portion of the sample for focus groups. What the researcher is referring to by a nested portion, is that the sa mple members selected for one phase of the study represent a subset of those participants chosen for the other part of the research study. In this study, the participants for the qual itative part of the study came from the participants of the quantitative part of the study (Onwuegbuzie & Collins, 2004). In this particular study, focus groups provided more than i solated interviews because the group members were able to react to and build upon the re sponses and comments in others. This helped the researcher to obtain the voice of the struggling reader (Langford & McDonagh 2003; Lincoln & Guba, 1985).
85 For this study, data collected through the use of f ocus groups supported quantitative data collection. In an effort to ensur e that the voice of the struggling reader was representative of the group, the researcher att empted to intentionally select participants for the focus groups whose initial att itude on the Elementary Reading Attitude Survey (ERAS) represented three distinct levels (full scale attitude in the lowest third, middle third, and in the highest third). How ever, when the district intervened and said they wanted the least amount of classrooms imp acted by visitors, the researcher was forced to select classrooms based on how many child ren had returned their informed consent letters. Specifically, the researcher was a sked to limit focus group participants to just one classroom at each of the five summer schoo l sites. This made it impossible for the researcher to use within case sampling to selec t a nested portion of the struggling readers. (See Threats to Limitation in this chapter for more information). Pilot Study When considering instrumentation and population, be cause the ERAS was normed using a heterogeneous population and the par ticipants in this study were homogenous, specifically third-grade struggling rea ders, it was important for the researcher to collect additional empirical reliabil ity and validity data. Therefore, the researcher conducted a Pilot Study to determine how the survey performed with thirdgrade struggling readers. International Review Board Approval The researcher obtained IRB approval in the winter of 2006 to administer a Pilot Study to determine if the Elementary Reading Attitu de Survey measured changes in attitude with a homogeneous population. The sample size for the Pilot Study consisted of
86 15 third-grade students enrolled in a third grade r emediation class at an elementary school on the west coast of Florida. The students f or the Pilot Study were located in the same district where the actual study took place. Ad ditionally, the students selected for The Pilot Study came from the same population as th e students for the actual study, since they were all retained third-grade struggling reade rs. Each of the students involved in the Pilot Study returned a signed consent form (see App endix G). The intent of the Pilot Study was to determine the reliability of the Elementary Reading Attitude Survey with a homogeneous populati on. Since it was not possible to conduct the Pilot Study during summer school, the r esearcher selected the FCAT as an event that might impact the attitudes of third-grad e struggling readers. The participating students were given the Elementary Reading Attitude Survey the week before the FCAT examination. The researcher administered the survey in the students classroom while the regular classroom teacher was present using the ERA S Pilot Study Group Protocol (see Appendix H). The researcher then returned two week s later on the day that FCAT testing ended. Once again the researcher administered the E lementary Reading Attitude Survey. The individual scores are broken down by recreation al attitude, academic attitude, and total attitude and are represented in Appendix I. Descriptive Statistics The distributions of attitude scores were examined separately for recreational reading attitude, academic reading attitude and tot al attitude using descriptive statistics. A summary including the mean, standard deviation, ske wness and kurtosis is provided in Table 7 and illustrated in Figure 2. For the differ ence in recreational reading attitude the skewness and kurtosis both suggest approximately no rmal distributions. For the
87 difference in academic reading attitudes the distri bution appears to be negatively skewed. This can be interpreted as the distribution has a l ong tail in the negative direction, or there was variability in the scores in the negative direc tion. Table 7 Descriptive Data Pilot Study Group N Mean Standard Deviation Skewness Kurtosis Recreation 15 -4.13 5.67 -.58 .70 Academic 15 -3.07 5.90 -1.34 3.0 Total Score 15 -7.00 10.56 -1.41 2.70 Figure 2. Changes in Attitudes Pilot Study nnr !" #n $#n %nr rnrrn
88 Dependent Measures t-Test The results of the recreational, academic and total attitudes as measured by the ERAS were analyzed using a dependent measures t -test. This analysis revealed a significant difference between mean levels of commi tment observed in two of the three conditions. Sample means of 4.13, -3.07, and 7.00 for the recreational reading attitude, academic reading attitude, and total reading attitu de respectively show enough variation to be of practical importance. The results of a dependent measures t-test indicate d that although recreational attitude [ t (14) = -4.13, p = .014] and total reading attitude are statistical ly significant, [ t (14) = -7.00, p = .02], academic reading attitude does not show en ough variation to be of statistical significance [ t (14) = -3.07 p = .06]. The sample means are displayed in Figure 2, which shows the students attitude scores decreased after the students took the FCAT reading test. These results demonstrate that a lthough the ERAS can be used to measure change with a homogeneous population, chang e in a small sample size may not always be statistically significant. Reliability of the Instrument Cronbach Alpha When considering the reliability of an instrument i t is important to consider, the degree of consistency with which it measures whatev er it is measuring (Ary, Jacobs & Razavieh, 1996, p. 273). One way to measure reliab ility involves assessing the extent to which all items are measuring the same thing. This involves assessing a tests internal consistency. The Cronbach Alpha coefficient is used to measure internal consistency when test items are not scored as right or wrong, b ut rather are given a range of scores, as
89 is the case with the Elementary Reading Attitude Su rvey (McKenna & Kear, 1995). The items on the survey were scored on a scale of one t o four for questions related to both academic and recreational reading. This scale makes the Cronbach Alpha an appropriate measure of the reliability. Cronbachs Alpha represents a measure of internal consistency amongst a set of items. The more consistent the score of a set of it ems, the higher the reliability of the measure. The maximum value is 1.0 (Cronbach, 1951). Following the guidelines of Guilford and Fruchter, (1978), a reliability coeffi cient of .70 or higher is considered acceptable. Following a Pilot Study, the Cronbach Alpha coefficient was computed for recreational attitude, academic attitude and total attitude respectively. The raw coefficients for the pilot administration of the te st for reading attitudes following the FCAT test ( n= 15) were: pre-test recreational att itude .30, post test recreational .81, pretest academic attitude .49 and post-test academic a ttitude .65, and pre-test total attitude, .65 and post-test total attitude .66. All of the co efficients from the Pilot Study, with the exception of post-test recreational are of interest One explanation as to why most of the coefficients are low is that the low Alpha levels m ay indicate several attributes and dimensions rather than just one and thus the Cronba ch Alpha is deflated. Another reason might be due to limited numbers of items in the sca le. Yet another explanation might be due to the sample size (n=15). After analyzing the ERAS results during the full st udy, it was important to once again assess the reliability of the measure. When t he Cronbachs Alpha coefficient was computed for pre-test recreational attitude, post-t est recreational attitude, pre-test academic attitude post test academic attitude, an d pre-test total attitude, and post-test
90 attitude for the full study the following coeffici ents were reported. The raw coefficients for each of those variables were .82, .83, .85, .47 .91, and .73 respectively. Each of these numbers except post test academic attitude is consi dered satisfactory following the guidelines of Guilford and Fruchter (1978). The raw Alpha for post-academic at a .47 is of interest because it is well below a .70 which is considered an acceptable Alpha coefficient (Guilford and Fruchter, 1978). An Alpha score of .47 indicates that the scores on the academic questions of the ERAS were not cons istent. This inconsistency can also be seen by looking at the difference in mean academ ic scores based on school site. This also may be why the Factorial ANOVA for school site and academic attitude was significant, yet a follow up Tukey test could not p inpoint the significance. Interrater Reliability and Scoring Another reliability issue involved the consistency of scoring the test items. To control for this the researcher double checked the scores or the ERAS before entering them into an Excel Spreadsheet. After entering the information into Excel, once again the researcher double checked the information for accur acy. Additionally, to ensure equitable representation for each school and classroom in the study sample a random sample of 20 percent of the ERAS surveys were double scored to c heck for accuracy. The second scorer was a fellow graduate student with extensive experience in reading. Prior to double scoring the second scorer was trained by the resear cher in how to score the ERAS. The training process began with an overview of the ERAS Next, the researcher explained how to score the items based on a four point scale and how to transfer the scores to the score sheet. The second rater then observed the res earcher score a survey before
91 attempting to score a random sample of 20 percent o f the surveys. Surveys from each of the school sites were represented. Additionally, inter rater reliability measures were used during the qualitative part of the study. The second scorer also verified the codi ng of the focus group data by double coding 20 percent of the answers to the focus group questions to see if the codes were the same as those assigned by the researcher. Following the collection of focus group data, the researcher conducted a coding training with the additional scorer (See Appendix J). The coding training followed these steps: 1. The training began with an overview of the coding p rocess. 2. Next, the researcher used a sample question (that w ould not be scored by the secondary coder) as a practice question. 3. Next, the researcher and the double scorer examined the sample question and reviewed the sample codes from the typology that th e researcher had already established. 4. The researcher answered any questions the double sc orer had about the process. 5. Finally, the second scorer reviewed the transcripts from 20 percent of the questions and coded the answers. To ensure equitabl e representation for each school and classroom in the study, the 20 percent o f transcripts that were scored were made up of one question at four of the five sites. Two questions were verified from the fifth site. 6. The researcher then explained to the second coder w hat would be done if there was a discrepancy amongst the two coders. In the ev ent of a discrepancy the
92 two coders would discuss the specific response and come to agreement as to how to code it. In the event that a consensus could not be agreed on, both answers would be accepted. Threats to Legitimation (Qualitative Phase) There were possible threats to legitimation in the qualitative portion of the study. The first threat concerned researcher bias. First, when considering descriptive validity it was important that the researcher report the accoun ts of the focus groups factually. In this study, in order to control for this, the researcher did not act as the moderator during focus groups, but rather the researcher selected a modera tor. The researcher then served as an observer taking notes and observing. Additionally, the researcher tape recorded all focus group sessions, which she later transcribed. These results can be found in Chapter Four. The moderator for the focus groups was a graduate s tudent trained in qualitative research. The moderator used a group protocol (Appe ndix K) during the focus groups. The graduate student and researcher practiced using focus group protocols prior to collecting data. The use of multiple observers allo wed for cross checking of observations (Johnson & Christensen, 2004). Additionally, the us e of an outside moderator helped the researcher ensure that the informants did not provi de only socially desirable responses to the questions (Onwuegbuzie, 2002). The use of simu ltaneous triangulation helped control for this bias. This triangulation was accom plished by gathering quantitative data from the ERAS, qualitative data from the focus grou p interviews, and classroom observations. Additionally, the researcher left an audit trail which included oral tape recordings (with permission) of the focus groups as well as analyzed transcripts of the
93 students oral responses to the questions on readin g preference. Finally, in chapter 5 the researcher made clear her position and any biases t hat may have impacted the investigation. One possible bias concerned the Voyager Passport program, the reading program the students used in summer school. Because the researcher prefers to view reading instruction using a balanced approach to li teracy, it was important for her to state this position in her final report. Another threat to legitimation involved interpretiv e validity. Because the researcher only gathered data from the students dur ing the beginning of summer school and the end of summer school she was concerned that it might be difficult to obtain an adequate representation of the voice under study (Onwuegbuzie, 2002). Therefore, in an effort to capture the voice of retained thirdgraders and to get inside the heads of the participants (Johnson & Christensen, 2004, p. 251), the researcher used participant feedback, or member checking during data collection This was done informally by having the group moderator restate key statements a t the end of the focus group session. Additionally, member checking helped to clear up an y areas of miscommunications. Another way the researcher controlled for this bias was through field notes gathered during weekly classroom observations of targeted st udents. Another threat occurred during the actual study. St udents were originally targeted for participation in the focus groups based on thei r score on the ERAS. However, the district requested that the observations and resear cher visits impact the fewest number of classrooms as possible. Therefore, the researcher h ad to select classrooms where the most students had returned a yes informed consent. In al l situations this meant that the researcher was unable to obtain an equal representa tion of attitudes (low, average and
94 high). Specifically, of the 22 focus group particip ants only two of the students had low attitudes at the beginning of the study. The remain ing twenty had average or high attitudes. However, there were only 10 students wit h low attitudes at all of the summer school sites. This may have been the result of a lo w permission return rate for students with low attitudes. A lack of parental involvement is characteristic of retained students (Jimerson et al., 1997; Jimerson & Kaufman, 2003). Additionally, due to a request from the district that the researcher not conduct the fi delity checks herself, the researcher had a lack of access to the data in regards to the fideli ty checks. Additional information on the fidelity measure can be found in Chapter Four. Classroom visits were conducted weekly by the resea rcher and took place at varying times throughout the school day. Although t he majority of the observations took place during Voyager lessons, some of the observations took place durin g other literacy activities. The results of these field notes gather ed during observations and vignettes of each of the targeted classrooms can be found in Cha pter Four. Instrumentation Quantitative Instrument The instrument used in the quantitative portion of this study was the Elementary Reading Attitude Survey (McKenna & Kear, 1990) (See Appendix A). This public domain instrument was selected because it is design ed to be used with elementary age students, can be administered to an entire class of students in a manner of minutes, and provides the researcher with three different scores : recreational reading, academic reading, and a composite score. The purpose of the ERAS is to examine the reading attitudes of elementary age students so that teache rs can estimate the attitude levels of
95 their students efficiently and reliably. The conten t consists of 20 items that assess students in regards to two sub-scores: recreational reading and academic reading. A high score would represent a positive attitude toward re ading, a low score a negative attitude towards reading. The developers of the assessment a re Michael C. McKenna and Dennis J. Kear. The format of the assessment is a pictoria l questionnaire, using the Garfield cartoon, and designed to be given orally. To administer the assessment, the test administrato r begins by telling students he/she wishes to find out how the student feels abo ut reading. In an effort to prevent students from giving a response that will please th e test administrator, the administrator emphasizes to the students that this is not a test and that there are no right answers. Additionally, the administrator stresses sincerity and explains that she is going to read some statements about reading and that the students should think about how they feel about each statement. Next, the administrator distr ibutes the surveys and discusses the pictures of Garfield and the mood he appears to be in and what that means. Class consensus was then achieved as to the predominant m ood characterized by each illustration. In an effort to minimize the possible effects of decoding difficulties, each item was then read orally 2 times slowly and distin ctly, as students followed along and marked their responses. For each item, a total of 4 responses are given ran ging from 1 to 4. A score of 4 represents Garfield looking the happiest. A score o f 1 on the other hand represents Garfield looking very frustrated with his hands cro ssed. To score the survey, the administrator accounts 4 points for the happiest Ga rfield, 3 points for each slightly smiling Garfield, 2 points for each mildly upset Ga rfield, and 1 point for each very
96 frustrated Garfield. Three scores can be obtained f or each student: the total for the first 10 items, the total for the second 10 items, and a com posite total. The first 10 questions on the survey relate to attitude towards recreational reading and the second 10 questions relate to attitude toward academic aspects of readi ng. To interpret the scores, a formal approach involves converting raw scores into percentile ranks using a table provided in the dire ctions. If the researcher prefers to interpret the scores informally, they can note info rmally where the scores fall in regards to the four different nodes of the scale (Garfield pictures). Responses are quantified by assigning 1 to 4 points to each item, from most neg ative (1) to most positive (4) respectively. Thus, scores on each of the two subsc ales can range from 10 to 40 total points. A total score of 50 would represent a score that was directly in the middle, which could be interpreted as an indifferent score. For t he purpose of this data analysis, the researcher relied on the use of raw scores. Because the instrument is easy to administer and thorough directions are provided with the instr ument, the creators state that it is not necessary for the administrator to receive training before administering the assessment (McKenna & Kear, 1990). However, to ensure that eac h of the summer school teachers administered the survey the same way, the researche r used a Group Protocol (see Appendix K) to train the summer school teachers. The developers of the survey created norms by condu cting a large-scale study in January 1989. The survey was administered to 18,138 students in Grades 1-6. Steps were taken to ensure that the sample was sufficiently st ratified; specifically participants were drawn from 95 school districts in 38 states. There were five more girls than boys. The ethnic distribution was similar to that of the U.S. population at that time.
97 Overall developmental trends in attitude were addre ssed by means of two separate one-way designs for recreational and academic readi ng attitude by grade. The F test for recreational attitude was significant (F=104.1, p < .001). Academic reading was also significant F=266.0, p <.001. The researchers then ran a post-hoc Scheffes test to determine whether the mean drops between successive grade levels were significant. Recreational reading drops were all significant (p <.05) except between second and third grades. Whereas for academic reading all five decli nes between successive grades were significant (p < .05). Thus, as children progressed from first to sixth gr ade their attitude toward reading both recreationally and academically declined. Spec ifically an examination of first grade means revealed a 31 for recreational attitude and a 30.1 for academic attitude. Visually both of these scores would be located near the slig htly smiling Garfield on the ERAS. By sixth grade the two means had fallen to 27.9 recrea tional and 24.6 academic. Visually these scores fall between the slightly smiling and the slightly frowning Garfield which might suggest virtual indifference. The long-term e ffect sizes of .54 and .80, respectively from grade 1 to grade 6 are considered significant (McKenna et al., 1995). The F statistic was calculated with effect sizes of .20 for recreat ional and .27 for academic attitude, which can be interpreted as moderate effect sizes ( Cohen, 1965). The researchers also looked at attitude and ability attitude and gender and attitude and ethnicity. When the researchers examined abilit y they found that a negative recreational attitude is related to ability. Furthe rmore, the Attitudinal gap among ability levels widened with age. However, when the research ers looked at academic reading attitude they found a similar negative trend regard less of ability (McKenna, et al., 1995).
98 These results support the McKenna model which sugge sts that a readers history of success or frustration plays a central role in shap ing the readers attitude (McKenna, 1994). When gender was examined, girls had more pos itive attitudes towards both academic and recreational reading at all grade leve ls. Further more in the case of recreational attitude the gap widens with age. This is similar to the ability findings. Academic attitude remained relatively constant. Eth nicity did not appear to play a role in the negative trends of either recreational or acade mic reading. The same thing was found when it came to the teachers reliance on a basal r eader. Reliance on a basal reader did not appear to be meaningfully related to recreation al or academic reading. Reliability was measured using Cronbachs Alpha (Cr onbach, 1951). It was calculated at each grade level for both subscale sc ores and for the composite score. These coefficients ranged from .74 to .89, and of 18 coef ficients computed (for the two subscales and the full scales at each of six grade levels), 16 were at least .80 (McKenna et al., 1995). A value of .70 is considered acceptabl e. The majority of the coefficients were .80 or higher. Two coefficients were lower, recreat ional subscales at Grades 1 and 2. The researchers suggest that this may mean that the sta bility of young childrens attitudes toward leisure reading grows with their decoding ab ility and as they become more familiar with reading as a hobby or pastime. To gather evidence of construct validity, on the re creational subscale the researchers began by sorting children in the normin g group into sub-groups. The first sub-group separated children based on access to a p ublic library. The students to whom a library was available were then divided into two mo re groups, those with library cards and those without library cards. Cardholders were f ound to have a significantly higher
99 recreational attitude score (p <.001, M=30) than no n-cardholders (M =28.9). This provided evidence that the scores varied predictabl y with an outside criterion. Next, the researchers compared students who presently had boo ks checked out from their school library to those who did not. In this case the comp arison was limited to those children whose teachers reported not requiring them to check out books. Once again the means of the two groups varied significantly (p <. 001). Chi ldren with books checked out had a higher mean (M =29.2) than those who had no books c hecked out (M =27.3). The next test of the recreational subscale compared students who reported watching an average of less than one hour of televi sion per night with students who watched more than two hours per night. This time th e recreational mean for the low television group (M=31.5) significantly (p <.0001) exceeded the mean of the heavy television viewing group (M=28.6).Thus the research ers concluded that the amount of television watched varied inversely with students attitudes towards recreational reading. Next, the researchers examined the validity of the academic subscale. This time the researchers categorized their children based on reading ability (low, average, high). The high ability readers (M =27.7) significantly (p < .001) exceeded the mean of the low ability readers (M =27.0). This provided evidence t hat the scores were reflective of how students truly felt about reading for academic purp oses. The relationship between the two subscales was also examined. The researchers found, The inter subscale correlation coefficient was .64, which meant that just 41 perce nt of the variance in one set of scores could be accounted for by the other. It is reasonab le to suggest that the two subscales, while related, also reflect dissimilar factors-a de sired outcome (McKenna & Kear,
100 1990). Finally, the researchers conducted a factor analysis and found evidence that the surveys two subscales reflect discrete aspects of reading attitude. Data Collection Quantitative Procedures For the quantitative portion of the study data were gathered from The Elementary Reading Attitude Survey (McKenna & Kear, 1990). Aft er training the summer school teachers using The ERAS Teacher Training Protocol ( see Appendix E) all of the thirdgrade summer school teachers used the ERAS Group Pr otocol (see Appendix F) to administer the survey to each of their students. Th e survey was administered during the morning of the first day of summer school. Followin g implementation the surveys were bundled and sent to the front office to be placed i n a large manila envelope that the researcher had dropped off with the office staff Mo nday morning. This same procedure was followed at each of the summer school sites. Ad ditionally, also on the first day of school the summer school teachers were asked to sen d the Informed Parental Consent forms home with all of the students (see Appendix B ). At the end of the first week of summer school, the researcher returned to each site and picked up the completed surveys and returned informed consent papers. After the first week, the researcher then went thro ugh the returned Parental Consent forms and determined which students had par ental permission to participate in the study. A total of 547 third-graders qualified f or summer school. However, just 336 students (61%) actually attended summer school the first day. In all, 115 students out of 336 students had permission to participate in the s tudy. Complete data were collected on 91 students. Next, the researcher calculated the fu ll scale, recreational and academic
101 attitude scores for each of the participating stude nts surveys. Then the researcher assigned a numerical code to each student. Data wer e then placed on an Excel spreadsheet so that it could be entered into SAS. A fter entering data, the researcher double checked to be sure all data were entered on the spreadsheet accurately. Next, the researcher had a second person double check the acc uracy of 20% of the scores to be sure they were entered correctly. Finally, the researcher color coded the students ID numbers based on their full scale scores into three categories 0-40 low attitud e, 41-60 average attitude, and 61-80 high attitude. Those scores were used to help the r esearcher determine the participants for the subsequent qualitative portion of the data coll ection. The researcher administered the post-test the last week of summer school following the same procedures and protocols as the pre-test prior to the Stanford 10 assessment. Qualitative Procedures The qualitative research approach relied on the use of focus groups and field notes collected through classroom observations. Focus Groups. Focus groups are defined as, a type of group inte rview in which a moderator leads a discussion with a small group of individuals to examine in detail how the group members think and feel about a topic (Jo hnson & Chistensen, 2004, p. 185). Advantages of focus groups include: data collection can be done quickly, the researcher interacts directly with participants, it allows for rich data collection through open response, synergy of group, flexibility, its appro priate for use with children, and the results are easy to understand (Stewart and Shamdas ani, 1990). The use of focus groups offered a more in-depth understanding than could ha ve been obtained through a survey
102 alone (Barbour & Kitzinger, 1999; Billson, 1994; Ed munds, 1999; Langford & McDonagh, 2003; Morgan, 1988). Through the use of f ocus groups, the researcher gained a more in depth understanding of third-graders att itudes toward reading. Additionally, it allowed the researcher the opportunity to interact directly with the third-graders and to attempt to get inside their heads (Langford & McD onagh, 2003; Johnson & Christensen, 2004). A fellow graduate student trained in qualitative re search and working with struggling readers assumed the role of the group mo derator, leading the focus group discussion. Edmonds (1999) recommends the following qualities in a good moderator: ability to learn quickly, experience, organizationa l skills, flexibility, good memory, good listening skills, strong probing skills, time manag ement skills, and a good personality. Additionally, Stewart and Shamdasani (1990) recomme nd that the moderator is adaptable, alert, ambitious, assertive, cooperative decisive, dependable, persistent, tolerant of stress, and willing to assume responsib ility. Additionally, children are often more comfortable with a female moderator (Stewart a nd Shamdasani, 1990). The moderator worked as a reading coach at an elementar y school and was accustomed to working with struggling readers. The researcher exp anded on Edmonds recommendation of a good personality, and selected a moderator wit h the following traits: an outgoing personality, a good sense of humor, approachable, a nd a knack for making children feel comfortable in her presence. After selecting a group moderator who would conduct the focus groups, in an effort to fine-tune the group protocol (See Appendi x K), prior to summer school the
103 group moderator and the researcher conducted an inf ormal pilot of the focus group protocol with a group of second grade students from the researchers own class. The purpose of the pilot was to answer the following qu estions: Were the questions appropriate? Would the children answer them? Would the children understand what they were being asked? Were the questions worded in such a way that usable data were gathered? As a result of the Pilot Test a few modifications w ere made in the original protocol. Results of the pilot revealed the importa nce of the moderators ability to get the children to talk. Many of the students were very sh y and only gave one-word responses. This emphasized the importance of the child feeling comfortable with the moderator. It really helped that the moderator had a good sense o f humor and knew how to talk to kids. The primary modification to the protocol centered a round the informal warm up. Originally the researcher had planned on beginning with a discussion on foods. After conducting the Pilot Test, the researcher realized that for the most part all of the children had eaten the same thing for breakfast, since most of them had eaten in the school cafeteria. After realizing that in summer school al l of the children would be receiving free breakfast and lunch and would have had the exact sa me thing to eat, the researcher decided to ask the children about pets (Have you ev er had a pet? Does anyone have a dog? ). When the warm up was changed, the researche r found that the children got very excited and that although some of the pet stories w ere tragic, the children wanted to talk about their pets.
104 An additional result of the Focus Group Pilot revea led that it was helpful to change from a monologic to a dialogic interaction i n order for the focus group to feel more conversational. Additionally, allowing the fle xibility to vary the order of the questions allowed for more conversational patterns. The moderator supported the style most adaptive to the comfort level of the children. Some of the groups did better taking turns answering one question at a time. Other group s displayed more cohesiveness and piggybacked off each others responses. The researc her assumed the role of an observer. Stewart and Shamdasani (1990) support this flexibil ity and encourage the researchers to understand that, groups take on lives of their own and the interview guide is just that a guide, which the moderator and group should be allo wed to modify if it proves desirable (p.62). The protocol (see Appendix K) provided a framework for the discussion. The nature of the discussion was to find out the studen ts attitudes about reading after they had been in a scripted summer school environment fo r 30 days. The researcher began by introducing herself and the moderator and making su re that the children felt comfortable participating in the focus group. The researcher al so explained that she would be tape recording the session and showed the children the t ape recorder before placing it in the center of the group. The moderator began by having the children become more comfortable by introducing themselves and their ho me school. Next the moderator introduced herself again and explained her role and what to expect during the session. An informal warm up helps the moderator to get an idea of the participants interaction style (Billson, 1995). Therefore, the moderator began wit h a discussion on pets (Have you ever
105 had a pet? Does anyone have a dog? ). Next, the mod erator reviewed the following ground rules: speak clearly one at a time, no right or wrong answers, need for active participation, sharing the floor. When formulating the questions for the focus groups the researcher followed Stewart and Shamdasanis recommendation that questi ons be ordered from general to specific if possible. They also recommend that ques tions be ordered by their relative importance to the research agenda. This explains wh y question #2, Do you like reading? Why or why not? is positioned second. Finally, the moderator asked the following question s: 1. Tell me about summer school. Which part of summer s chool do you like best? 2. Do you like reading? Why or why not? 3. What types of books do you like to read the most? 4. Does anyone read to you at home? 5. Have you noticed and changes in your feelings about reading this summer? 6. Tell me about Voyager Passport. Because participants in focus groups do not alway s say everything they think, it was necessary for the moderator to recognize ve rbal and nonverbal cues. At times it was necessary for the moderator to ask follow up qu estions, restating, summarizing or asking the group for samples or examples (Stewart a nd Shamdasani, 1990). During questioning the moderator followed the clarificatio n and probing routines stated below:
106 Clarification and Probing Routines 1. Because children will agree with others to avoid st anding out, it will be important to frequently ask if anyone has other ideas or different opinions. 2. Watch for gestures and facial expressions that may reveal something about the accuracy of a comment, or suggest that so meone has a strong feeling about a question being asked. 3. If a talker takes over the conversation, thank th em for sharing and call on another student. 4. Begin with voluntary responses, if some children ar e not participating, call on them. Following the pilot test, the researcher used the d ata gleaned from the ERAS initial surveys to attempt to select focus group pa rticipants from each site. Focus groups were made up of four to six students from each summ er school site. Although the researcher attempted to select students based on th e following criteria: two students whose full scale attitude was in the lowest third, two students whose full scale attitude was average and two students whose full attitude wa s in the highest third made this was not possible in most cases. Additionally, although the researcher attempted to select children with strong verbal skills this was not alw ays possible. Specific information regarding the selection of focus group participants from each site can be found in Chapter Four. Next, the researcher scheduled times to conduct the Focus Group Interviews. Focus Group Interviews were scheduled after the ERA S had been given the second time
107 and before the Stanford 10 was given. The focus gro up interviews took place on Wednesday during the final week of summer school an d were all tape-recorded. Although the researcher and the moderator had allow ed two days for focus group sessions, because the sessions did not last a full hour, all of the focus groups were conducted on one day. The focus groups were held in empty classrooms with the students sitting around a table, in the media center, in a c ircle on the floor in the hallway, or in one case, in a conference room. Stewart and Shamdasani (1990) recommend a table as a way of providing a protective barrier between responden ts which in turn gave less secure members of the group a sense of security. Additiona lly, the circular arrangement of a table provided a maximum opportunity for eye contac t between the moderator and other group members (Stewart and Shamdasani, 1990). In th e focus groups the only noticeable difference in outcomes was that the group in the co nference room required a great deal of redirection. This was because some of the children were fascinated with adjusting the height of the conference room chairs. Although the researcher limited focus groups to one hour per group, the length of the focus groups varied between 30 minutes and one hour. The shorter duration of the focus groups may be because the groups were homogen eous in nature being made up of all struggling readers (Stewart and Shandasan, 1990 ). The lack of involvement could also have been because the participants did not tru st the researcher and moderator enough to open up. The sessions were all tape-recorded and later transcribed by the researcher. Additionally, both the researcher and the group mod erator took notes during each of the focus groups.
108 Classroom Observations. Once the children for the focus groups had been selected, the researcher began classroom observatio ns. The observations took place on Wednesdays and lasted approximately 45 minutes. Dur ing the classroom observations, the researcher gathered field notes about the class room and what the teacher was doing as well as the childrens attitudes during the observa tions. Attitudes were measured using a likert-type scale that emulated the one used in the ERAS survey. (See Appendix L) Also, at least once at each site the researcher had the o pportunity to talk informally with the summer school teachers about the students, summer s chool, and Voyager Passport The results of the classroom observations and meetings with the summer school teachers are collapsed into vignettes and are also found in Chap ter Four. During the initial observations it appeared that th e students attitudes improved as the day went on. Therefore, the researcher varied t he times she completed observations in the classrooms. By varying the times, this meant t hat the classroom teachers did not know when the researcher was coming. It also meant that the researcher was able to observe different parts of the summer school day. D uring the informal classroom observations the researcher used her laptop to type notes on what activities the focus group children were participating in and their focu s, behavior, attitude, and participation during the activity. It was also an opportunity to observe the interaction between the teachers and the children. When the children were not involved in a Voyager Passport lesson the remainder of the day was spent doing literacy activities. Dur ing the classroom observations the researcher observed children in a variety of activi ty structures.
109 These included: 1. Small Group Voyager Passport Lessons: Lessons where the children were engaged in a lesson with the teacher using Voyager Passport materials where the teacher to student ratio was between 1:3 and 1:6. 2. Voyager fluency lessons with timed reads: A one min ute timed read of a story the children have already read. This occurs in Lessons 1, 2, and 5 and is how Voyager monitors fluency growth. 3. Voyager Passport Whole Group Vocabulary Lessons: lessons where the entire class was engaged in a lesson with the teacher usin g Voyager Passport materials where the teacher to student ratio was more than 1: 6. 4. Literacy Centers: A physical area set aside for spe cific learning purposes. The center consists of appropriate materials to enable children to explore and work independently (As individuals, with partners, or in small groups) and behave as active learners. 5. Independent Reading: An instructional approach that provides reading practice for individual students. Texts are student selected, ba sed on the students interests, needs, and self determined purpose and typically wi thin the students appropriate independent reading range. 6. Shared Reading: An instructional approach that mode ls strategies for reading text. During a Shared Reading lesson everyone has access to the text, in the form of enlarged text or multiple copies of the text.
110 7. Workbook: Two consumable student workbooks accompan y the Voyager Passport Kit. Each of the stories are in the workbooks. Add itionally there are lessons and activities as well as questions to answ er based on the story. 8. Teacher Read Aloud: Teacher reads aloud from a piec e of written text which may be in the form of a picture book or chapter book. T he teacher models reading fluency as well as exposing the students to new gen res and vocabulary words that are above their level. Quantitative Data Analysis Upon completion of data collection, the researcher utilized descriptive statistics to determine the mean and standard deviation of the st udents responses to the attitude survey. Data were organized according to academic, recreational and total reading attitude. The mean, standard deviation, skewness an d kurtosis values were all based on the total sample of 91 students from their response s to 20 items. Each of the 20 questions had a possible score of 4 points. After completing descriptive statistics on the data to determine the mean, standard deviation, skewness and kurtosis values, the resear cher conducted a dependent measures t-test. First, it was important to examine the assu mptions underlying the t -test. (Specific information on the assumptions can be found in chap ter Four). A t -test can be used to determine whether the means of two groups are stati stically different from each other. Since the researcher wanted to compare the mean sco res from the pre-test to the post-test, the recreational, academic and total reading attitu de scores from the ERAS pre-test and post-test were analyzed using a dependent measures t -test.
111 Prior to completing three factorial ANOVAs the rese archer first examined the assumptions that underlie factorial ANOVAs. In an e ffort to determine if attitudes differed in the subgroups, the researcher conducted three separate 2x2 factorial ANOVAs with alpha levels set to .05 for each effect to det ermine if gender, site, and the interaction between gender and site were predictors of change s cores. Following the factorial ANOVAs, when it was appropriate, the researcher con ducted follow up Tukey tests. The researcher also included descriptive statistics for the subgroups. To report the findings, the researcher then transferred those data into box plots (over time) and histograms. The researcher used SAS as the statistical software too l. Qualitative Data Analysis Focus Group Analysis. There is not one way to analyze focus group data th at is well researched and agreed on (Carey, 1995). Focus group analysis is the least agreed on process and the least well developed. Further, an a greed on technique does not exist (Carey, 1995; Kidd & Parshall, 2000). It takes inte rpretation and insight to develop the meaning of a focus group discussion (Stewart & Sham dasani, 1990). In an effort to qualitatively examine students att itudes on reading, gleaned from the focus groups, the researcher used content analy sis (Krippendorf, 1980; Stewart & Shamdasani, 1990) to break data into content chunks and to code the content into conceptual categories. Strauss and Corbin (1990) st ress the need for open coding, which requires the researcher to remain open as new relat ionships and categories emerge during data collection. In this study, the researcher deve loped a set of open codes as she labeled the key points made by the interviewees for each qu estion asked during focus groups.
112 When conducting the content analysis, the first ste p was to transcribe the audiotapes of the five focus group sessions. The transcr ipts served as the basis for further analysis. The researcher also used additional obser vational data in the form of notes that were taken during the focus group sessions by the r esearcher and the moderator. This observational data helped the researcher to interpr et the transcripts. After the tapes were transcribed, the researcher re ad through the transcripts for each school site for Question 1. In an effort to ex amine the meaning of the focus group discussions and its implications for research on st ruggling readers attitudes, the results of the focus group discussions were coded using semant ic content analysis (Stewart & Shamdasani, 1990). Content analysis is defined by K rippendorf (1980) as, a research technique for making replicable and valid inference s from data to their context (p. 21). When implementing the first stage, Data Making, the researcher defined the appropriate unit or level of analyses as words and phrases. The temporal designation for creating categories was iterative because, although the rese archer had categories in mind from the survey administered during the quantitative portion of the research, the researcher did not want to limit the temporal designation to just thos e categories. By utilizing an iterative temporal designation the categories were able to be created at various points during the research process (Constas, 1992). Next, the researcher used the cut and paste techniq ue (Stewart & Shamdasani, 1990) to go through the transcripts and identify th e sections that were relevant to the research question. Based on this initial reading, t he researcher determined a classification system for assigning units to categories. This was done beginning with the first question. The researcher went back to the transcribed focus g roup session and took the students
113 responses and assigned each childs response to a w ord or phrase. For instance, the first question was, Tell me about summer school. What pa rt of summer school do you like best? Responses varied and consisted of comments s uch as, I like recess, or I like reading independently. After going through each st udents response, the following codes or chunks emerged: timed reading, independ ent reading, read aloud, pass test, learning, recess, playing with friends, meeting new people, games and teachers. The researcher then took the codes or chucks and reco rded them on a matrix by the childs name and question 1. From the matrix the researcher then looked at words and phrases that could be clustered together into a category. For instance, recess, playing with friends and meeting new people were all coded into the cate gory social. However, it was important for the researcher to und erstand that, The recording or coding of individual units is not content analysis (p. 112) (Stewart & Shamdasani, 1990). Stewart and Shamdasani recommend the use of virtually any analytic tool (p.113) when analyzing focus group data. In this st udy, the researcher used attribution analysis (Janis, 1965) to determine the frequency w ithin which certain objects were mentioned. Next, the researcher converted the frequ ency tables into percentages. Percentages are shown visually through pie graphs. In an effort to establish the reliability of the fi rst stage of data analysis, recording, Krippendorf (1980) recommends that the researcher e xecutes an explicit set of recording instructions which represent the rules for assignin g units to categories. In regards to inter rater reliability, the second rater who was also th e moderator, served as a second coder on 20 percent of the transcripts. The second rater wor ked from a typology that the researcher had already established. The typology included a li st of units. The second rater simply
114 assigned words of phrases to units. The second rate r coded 6 different questions at varying sites. In all the second rater coded 26 sep arate responses. A different question was coded at each site, except Franklin Elementary where two questions were coded. 92 percent of the codes were the same as the researche r. There were only two instances where there was a discrepancy. Both instances invol ved students who gave an answer and then were probed. During the probe, they gave anoth er answer. After collaborating, the researcher decided to accept both responses. Appen dix M offers an example of how the transcripts were coded. Classroom Observations. Classroom observations allowed the researcher the opportunity to observe the children during summer s chool to see what activities they were engaged in as well as what their attitudes wer e during the activities. The researcher relied on a priori categories to code the observati on and interpret the childs attitude. In the case of activity structures during summer schoo l, the categories were based on a mandate from Just Read Florida that specifically stated which approved activities could take place during the summer school day. During the visits the researcher observed students engaged in activity structures which were also the a priori categories. Each of these activity structures are previously discussed in this chapter. Next, the researcher quantified the data by counting how many times the category was observed (Tjora, 2006). Finally a cross case analysis of the findings was c onducted using the data to look for patterns across sites. The researcher then used the information collected from the classroom observations to write a vignette of each of the tar geted classrooms. The information from
115 the vignettes was then used to do a cross case anal yses which can be found in chapter Four. Triangulation of Findings. Finally, in an effort to use several different rese arch methodologies to research students attitudes about reading, the researcher triangulated the findings from the ERAS survey, focus groups and classroom observations. Mixed Data Analysis Due to the mixed nature of the study, the researche r had intended to quantitize the qualitative data into quantitative data to make sta tements about the findings. However, during the course of the study, Dr. Onwuegbuzie who developed the concept of quantitizing and was going to be instrumental in se rving as a mentor during data analysis, moved to another university. This unexpected change caused the researcher to revise her data analysis plan and to analyze quantitative and qualitative data individually before triangulating the findings from the focus groups, t he classroom observations and the ERAS surveys. Evidence of Ethical Considerations and District Per mission Finally, in an effort to protect the participants, the sampling design adheres to the ethical guidelines set forth by the International R eview Board. The researcher completed the necessary IRB application from the University o f South Florida. Part of this application included permission from the school dis trict where the study was conducted (see Appendix N). Included in the proposal were cop ies of the districts permission letter, consent forms for teachers and parents, and assent forms for students. After receiving IRB approval, the researcher first met with the ass istant superintendent. After she gave
116 initial approval for the study, she requested that the researcher modify the way the fidelity checks would be completed. She stated that because the researcher was also a teacher in the district, it would be inappropriate for the researcher to conduct the fidelity checks on her peers. Next, the assistant superinten dent asked the researcher to write a formal letter requesting permission from the distri ct to conduct the study (see Appendix O). As summer school grew closer, the researcher was as ked to meet with the Director of Elementary Education who had specific q uestions about how the study would impact the summer school teachers. The Director of Elementary Education gave permission for the researcher to attend the summer school training day and to train the teachers in the administration of the Elementary Re ading Attitude Survey (McKenna & Kear, 1990). Additionally, he made the decision tha t the survey be given to all of the students on the first day of summer school. During this same meeting, the Director of Elementary Education requested that in order to imp act the least amount of classrooms, the researcher limit participants for the focus gro ups to one class at each summer school site. The implications of this decision are discuss ed in Chapter Five. Organization of Remaining Chapters Chapter 4 summarizes the findings of the study. Bot h quantitative and qualitative findings are reported. Chapter 5 presents a summary of the study, conclusions and implications derived from the research findings, re commendations for practice based on the study conclusions and implications, and recomme ndations for future research.
117 CHAPTER 4 FINDINGS Overview of Chapter The purpose of this mixed method study was to deter mine how summer school using Voyager Passport would impact retained third-grade students attitu des about reading. The chapter begins with an overview of the summer school training. Next, information on the fidelity measure and a summary o f the fidelity data is presented. The remainder of the chapter is organized by resea rch question and begins by addressing the quantitative question, What is the effect of a scripted literacy program on the reading attitudes of elementary school struggling readers? To address the quantitative question, the researche r begins by providing a summary of descriptive statistics from the initial results of the Elementary Reading Attitude Survey (McKenna & Kear, 1990). Following t he descriptive data, the researcher presents the findings of the inferential statistics First, the researcher examines the assumptions as they relate to the dependent measure s ttest. Following an examination of the assumptions, the researcher examines the depend ent measures t -test findings. After presenting the findings of the t -test, the researcher examines the assumptions that underlie the Factorial ANOVA. Next, the researcher presents the results of three 2x2 Factorial ANOVAS. Data in this section are organize d around recreational attitude,
118 academic attitude, and total reading attitude. Next the result of a follow up Tukey test is shared. This section concludes with a summary of th e quantitative findings. To answer the second question, What do elementary school struggling readers perceive to be the effect of a scripted summer scho ol literacy program on their attitudes toward reading? The researcher begins by presenti ng information on the focus group participants in regards to gender, ethnicity, site and prior retentions and ERAS scores. Then the researcher provides short case vignettes w hich provide a snapshot of one teacher and her classroom at each school site. Next a cross case analyses of the vignettes from the five sites presents similarities and diffe rences across each of the classrooms. Next, the researcher provides a summary of the focu s group findings organized by question. Then the researcher provides the themes t hat emerged from the focus groups. Finally, in an effort to use several different rese arch methodologies to research students attitudes about reading, the researcher t riangulates the findings from the ERAS survey, focus groups and classroom observations. Summer School Training Training for summer school teachers took place the Friday before summer school began and lasted four hours. The training was facil itated by the summer school coordinator for the district and three representati ves from Voyager Passport During the four hour training, each teacher received a Teacher s Resource Kit and the materials necessary to implement Voyager Passport in their summer school classroom. Although there were 35 teachers in the initial training, in the end a total of 29 teachers participated in the study. Four of the teachers were released du e to low enrollment. Another teacher was absent for the initial training and never recei ved the materials. A sixth teacher was
119 absent at the end of the study and was unable to ad minister the post assessments. During the training the third-grade summer school teachers received specific guidelines and instructions on how to structure their summer schoo l day (see Appendix P) as well as how to implement Voyager Passport in their summer school classrooms. Additionally, the teachers received tips and suggestions for impl ementing the program with enthusiasm and fidelity. Additional information regarding the teacher training can be found in Chapter Three. Fidelity Checks Fidelity of implementation is the actual presentati on of instruction the way it was intended to be delivered (Gresham et al., 2000). Sp ecifically, it is the adherence to the intervention protocol in comparison with the origin al program design (Mihalic, 2002; Mowbray et al.,2002). Measurement of fidelity is es pecially crucial with studies that seek to provide evidence for the effectiveness of an int ervention (Mowbray et al., 2002). In regards to educational research, if there is a h igh rate of fidelity in the implementation of a program, then the administratio n and staff can rule out this variable in regards to student achievement. Gresham et al., (1990) explored the extent to which integrity was assessed in the literature on learnin g disabilities (LD). This was achieved through an analysis of articles in three LD journal s from January 1995 to August 1999. Results of the analyses revealed that of the 479 ar ticles published in these journals, although 65 articles were focused on an interventio n; only 12 of the articles measured and reported data on treatment integrity. Establishing the fidelity of implementation is crucial to assuring that procedures are implemented with in tegrity.
120 Developing measures of fidelity, validating them, a nd using them can be intensive and costly. However, the other option is to recomme nd programs that are either not effective, or are only effective if implemented in a particular way (Borrelli, Sepinwall, Ernst, Bellg, Szaijkowski, Breger, DeFancesco, Leve sque, Sharp and Ogedegbe, 2005). When establishing a measure to assess the fidelity of the implementation of a program, it is vital that the fidelity measure itse lf is supported through reliability and validity evidence. In order for researchers to have sufficient evidence to support internal validity, there must be a valid measure of the leve ls of validity during the implementation of an intervention in a classroom (Dumas, Lynch, La ughlin, Smith and Prinz, 2001). It is important to follow steps when establishing f idelity criteria. The following steps are involved in establishing fidelity criteri a: (1) Identify critical intervention components and define measurable indicators for the components. (2) Collect the data to measure the indicators. (3) Examine the reliability and validity of the fidelity criteria. When examining fidelity measures, The National Cent er for Learning Disabilities (http://www.ncld.org/content/view/1220/389/ ) recommends that fidelity of implementation measures: (1) Link interventions to improved outcomes (credibility) ; (2) Definitively describe operations, techniques, and c omponents; (3) Clearly define responsibilities of specific persons; (4) Create a data system for measuring operations, techniques, and components; (5) Create a system for feedback and decision making (formative); and (6) Create accountability measure s for non-compliance (Mowbray et al., 2002). A summary of the fidelity data reported by the dist rict can be found in Table 8. The chart contains the observation (which comes dir ectly from the Voyager Fidelity
121 Measure) that was used during summer school. The fi delity measure asks the observer to rate each observation as Clearly Evident, Somewh at Evident or Not Evident. When referring to Mowbray et al., it is important to fol low steps to establish criteria fidelity. Voyager Passport followed the first step by identifying critical in tervention components on their fidelity measure (Classroom Organization, Instruction and Pacing). However, although they defined indicators for the components measuring some of the components might be subjective. For instance, one of the items is, Pre-planning is evident in lesson delivery. This could be difficult to measure depen ding on specific attributes relating to the implementing teacher. Further indicators of how to assess whether or not preplanning is evident are not offered. A complete lis t of the measureable indicators for the components can be found in Table 8. The second step in establishing fidelity criteria involves collecting data to measure the indicators. Voyager Passport provides a way to collect data to measure the indicators on their fid elity measure. However, a vital step, Examine all reliability and validity of the fideli ty criteria has been omitted. Although Voyager Passport is one of 101 supplemental intervention reading programs that have been reviewed by the Florida Cen ter for Reading Research (FCRR), and reviewed under Reading First, Voyager Passport lacks any norming data on their fidelity measure on either the Voyager website, or on the Florida Center for Reading Research website (FCRR). After contacting Voyager Expanded Learning Systems directly the researcher was told that the company had no nor ming data on the fidelity measure. Additionally, the researcher contacted the district where the study was conducted. The researcher was told that the district did not have any norming data on the actual fidelity measure as well.
122 The researcher had planned on revising the district s fidelity measure to include specific information on the teacher (enthusiasm, af fect, and quality) and conducting the fidelity checks herself during summer school. Howev er, the district would not allow the researcher to complete the fidelity checks or revis e the measure. (see Appendix Q). This decision was made by the assistant superintendent. Her reasoning was that since the researcher was also a third-grade teacher in the di strict where the study was conducted, for the researcher to assess her peers might have b een seen as a conflict of interest. Fidelity indicates the extent to which teachers fol low a curriculum that is written. However, fidelity provides no insight into either t he quality of the curriculum or the quality of learning that might result from its use Although a total of 177 fidelity checks were conducted by the summer school coordinator fro m the district office, reading coaches at the summer school sites, and administrat ors at the summer school, because there is no norming data on the fidelity measure th e researcher is unable to assess the data or conclude that the intervention was implemented w ith fidelity.
123 Table 8 Summary of Fidelity Visits Measureable Indicators Clearly Evident Somewhat Evident Not Evident Classroom Organization Voyager materials 171 (96%) 5 ( 3%) (.006%) are accessible. Small group area 149 (84%) 16 (9%) 12 (7%) is designated. Instruction Preplanning is evident 149 (84%) 20 (11%) 8 (.05%) in lesson delivery. Teacher follows daily 155 (86%) 17 (9%) 9 (.05%) lesson plan. Instruction from both 142 (80%) 14 (8%) 20 (11%) modules delivered. Students respond chorally 155 (88%) 8 (.0 5%) 14 (8%) and individually during lesson Reading behaviors and 148 (84%) 13 (7%) 16 (9%) expectations are evident. Pacing Instructional pacing 110 (62%) 19 (11%) 48 (27%) matches suggested times. Are the adventure and lesson numbers indicated to students or posted on board? 112 (63%) 19 (11%) 46 (26%) n=177 ___________________________________________________ _____________________
124 Quantitative Findings The quantitative findings are organized around the first research question: What is the effect of a scripted literacy program on the re ading attitudes of elementary school struggling readers? To answer the question, the rea ding attitudes of 91 third-graders attending summer school were measured using the Ele mentary Reading Attitude Survey (McKenna & Kear, 1990). The survey was administered the first day of summer school as well as during the last week. Data were collecte d from five different summer school sites in the targeted district. This section will begin by presenting the findings from the initial results of the Elementary Reading Attitude Survey as they relate t o descriptive statistics. Following the descriptive data, the researcher presents the findi ngs of the inferential statistics. First, the researcher examines the assumptions as t hey relate to a dependent measures t test. Following an examination of the assumptions, the researcher examines the dependent measure t -test findings. After presenting the findings of th e t -test, the researcher examines the assumptions that underlie t he factorial ANOVA. Next the researcher presents the results of three 2x2 Factor ial ANOVAs. Data in this section is organized around recreational attitude, academic at titude, and total reading attitude. Next, the results of a Tukey test are shared. This sectio n concludes with a summary of quantitative findings. Descriptive Statistics The researcher utilized descriptive statistics to d etermine the mean and standard deviation of the students responses to the attitud e survey. Table 9 organizes the data according to academic, recreational and total readi ng attitude. The mean, standard
125 deviation, skewness and kurtosis values are all bas ed on the total sample of 91 students from their responses to 20 items. Each of the 20 qu estions had a possible score of 4 points. Figure 3 provides a summary of the mean scores on t he ERAS. When mean scores for pre-test academic, recreational and tota l attitude scores were compared to mean scores for post-test academic, recreational and tot al attitude scores, there was little difference in mean scores. Figure 3 confirms these findings. The maximum academic and recreation scores on the ERAS were 40, and the maximum total score on the ERAS was an 80. A score of 50 on the ERAS is considered to be directly in the middle which can be interpreted as an indifferent score. Table 9 offers information on the mean, standard de viation, skewness and kurtosis values for each of the tests. When considering the distribution of scores, all of the scores are negatively skewed. This can be interpreted as t he scores on ERAS were clustered on the right side of the distribution. Additionally, t he kurtosis values of (<1) suggest a platykurtic distribution with the majority of value s occurring the same number of times.
126 Figure 3. Mean Scores ERAS " ! n #n #n n n n n $ $ rnn %nr
127 Table 9 Descriptive Statistics Variable N Mean Standard Deviation Skewness Kurtosis Pre-test academic 91 28.85 6.70 -0.59 -0.30 Post-test academic 91 29.16 6.77 -0.56 -0.18 Pre-test recreational 91 29.24 6.21 -0.44 -0.44 Post-test recreational 91 28.88 6.54 -0.27 -0 .78 Pre-test Total 91 58.09 12.25 -0.55 -0.43 Post-test Total 91 58.05 12.63 -0.33 -0.68 ___________________________________________________ ___________________ Assumptions Underlying the Dependent Measures t-Tes t Next, in an effort to screen data for possible viol ations the researcher examined the following assumptions underlying the dependent measures t -test. Because the data were collected from an independent sample of differ ent scores from the sample, there is no reason to think that the independence assumption was violated. Next, because the scores were normally distributed, there was no need to question normality. The-Shapiro Wilk test (Shapiro & Wilk, 1965) tests the null hyp othesis that a sample came from a normally distributed population. In this study, pro babilities of .97, .95, and .96 for recreational, academic and total reading attitude r espectively provide evidence of failing
128 to reject the null hypothesis for normality. Additi onally, the skewness and kurtosis values appeared normal [pre-test sk = -0.55, ku = -0.43; post-test sk = -0.33, ku = -0.68]. Inferential Statistics Descriptive statistics provided basic data about th e sample and the measures. In an effort to reach conclusions that extended beyond the immediate data alone, the researcher used inferential statistics. First, the researcher conducted a dependent measures t -test. Dependent Measures t-Test A t -test can be used to determine whether the means of two groups are statistically different from each other. The recreational, academ ic and total reading attitude scores from the ERAS pre-test and post-test were analyzed using a dependent measures t -test. Since the researcher wanted to compare the mean sco res from the pre-test to the post-test, this was an appropriate test. This analysis revealed no statistical significance between mean test scores (academic, recreational and total) during the two l evels of time (pre and post). Specifically, when the difference in recreational a ttitude was examined the results indicated that the change was not statistically sig nificant [ t (90) = -0.36, p = .55]. When the difference in academic attitude was examined, o nce again the findings were not statistically significant, [ t (90) = .32, p = .61]. Additionally, when the difference in total attitude was examined, the findings were not statis tically significant, [ t (90) = -.03, p = .98]. Perusal of Table 10 provides a summary of the mean, standard deviation, skewness and kurtosis values from the paired samples t-test.
129 Table 10 Results of Paired Samples t-Test Measure N Mean SD Skewness Kurtosis Difference Recreation 91 -.36 5.83 0.13 0.88 Difference Academic 91 .32 6.00 0.62 1.23 Difference Total 91 -.03 10.59 0.59 1.58 ___________________________________________________ ___________________________________ Factorial ANOVAs Following the dependent measures t -test, in an effort to determine if gender and school site would result in differences in differen ces in recreational, academic and total reading attitude, three 2x2 factorial ANOVAs were c onducted. In order to suggest that differences in gender and school site would result in differences in recreational, academic and total attitude about reading, chance must be ru led out as a plausible explanation for the observed differences in the sample. To assess t he tenability of a chance explanation, three 2x2 factorial ANOVAs were conducted with an a lpha set at .05 for each effect. The degree to which the Type 1 error rates were actuall y controlled to the specified alpha level depended on how adequately the data met the a ssumptions of independence, normality, and equal variances.
130 Assumptions The assumption of independence was met by ensuring that different observations came from different individuals. The ERAS tests wer e administered individually and each student completed his/her test alone. The desc riptive statistics indicated that the assumption of normality was not violated. Next, bec ause the sample size was small in some of the subgroups, and the variances found in t he different subgroups are unequal; it is possible that the assumption of equal variances was violated. Finally, in regards to the normality assumption, the low sample size of some o f the sub groups may have caused the researcher to violate normality. According to t his analysis of the assumptions, since none of the assumptions were violated in a manner t hat would have substantial consequences on the interpretations, it appeared re asonable to conduct the factorial ANOVAs. Results In an effort to determine if attitudes differed in the subgroups, the researcher conducted three separate 2x2 factorial ANOVAs with alpha levels set to .05 for each effect. The results of the 2x2 factorial ANOVAs ar e presented in Tables 11, 12 13, 14, 16 and 17 and illustrated in Figures 5, 6, and 7. The results of the Factorial ANOVAs are organized by recreational attitude, academic attitu de, and total attitude on The Elementary Reading Attitude Survey (McKenna and Kea r, 1990). Recreational Attitude First, recreational reading attitude was examined. A summary table for the ANOVA on recreational attitude is provided in Table 11. When gender was examined, the obtained [ F (1, 90) =1.16, p =.28], was judged not to be statistically signific ant using
131 a predetermined Type 1 error rate of .05. This indi cates that the observed difference in recreational attitude for the male students is not different enough from the observed difference in attitude for the female students to c onclude that recreational attitude differences would differ across gender in the popul ation. The p -value of .28 suggests that it is reasonable to accept the null hypothesis and conclude that there is not a difference in recreational attitude based on gender. Next, when school site was examined, the obtained [ F (4, 90) =1.14, p =.34] was also judged not to be statistically significant usi ng a predetermined Type 1 error rate of .05. This indicates that the observed difference in recreational attitude at one school site was not different enough from the observed differen ce in recreational attitude at another school site to conclude that recreational attitude would differ across school sites. The p value of .34 suggests that it is reasonable to acce pt the null and conclude that there is not a difference in recreational attitude based on scho ol site. Finally, when the interaction between gender and sc hool site was examined, the obtained [ F (4, 90) =1.95, p =.11] was also judged not to be statistically s ignificant using a predetermined Type 1 error rate of .05.This indicates that the observed difference in recreational attitude based on gender and site w as not different enough from the observed differences at another site in regards to gender and site. The p -value of .11 suggests that it is reasonable to accept the null h ypothesis and conclude that there is not a difference in recreational attitude based on gender and site.
132 Table 11 Variable Difference in Recreational Attitude Source SS MS F Value p ___________________________________________________ ___________________________________ Gender 38.98 38.98 1.16 0.28 Site 153.58 38.40 1.14 0.34 Gender*Site 261.69 65.42 1.95 0.11 Error 2721.03 33.59 n=91 ___________________________________________________ ____________________ Table 12 provides specific information in regards t o the sample size, mean, standard deviation, skewness and kurtosis values fo r the changes in recreational attitude based on school site and gender. This figure indica tes high variability amongst the subgroups in regards to sample sizes. Additionally, al though differences in scores are noted, there are no trends in scores based on gender or si te. The histogram shown in Figure 5 provides a visual display of the mean change scores in regards to recreational attitude on the ERAS based on school site and gender.
133 Table 12 Difference in Recreational Attitude by Site and Gen der ___________________________________________________ ___________________________________ Site/Gender N M SD SK KU ___________________________________________________ ___________________________________ Dolphin Male 8 .75 6.76 -0.92 .34 Dolphin Female 9 -2.33 5.59 -0.72 -0.23 Horn Male 13 .15 4.88 -0.25 -0.63 Horn Female 4 5.5 9.68 1.30 1.0 Carter Male 6 -3.17 6.70 .98 1.58 Carter Female 7 .57 4.35 -0.68 .29 Franklin Male 5 -3.2 3.42 -1.65 3.33 Franklin Female 6 .67 4.18 -0.46 .44 Lincoln Male 21 .48 5.87 0.28 0.15 Lincoln Female 12 -2.08 6.05 -0.89 1.45 ___________________________________________________ _____________________
134 Figure 4. Difference in Recreational Attitude in Regards to Gender and School Site " "" ! &'r(r)n*r+r,rr rr rrnn %#,*-%#,-
135 Academic Attitude Next, the researcher examined variable differences in academic attitude. A summary table for the ANOVA on academic attitude is provided in Table 13. When gender was examined, the obtained [ F (1, 90) =1.01, p =.32], was judged not to be statistically significant using a predetermined Typ e 1 error rate of .05. This indicates that the observed difference in academic attitude for th e male students is not different enough from the observed difference in attitude for the fe male students to conclude that academic attitude differences would differ across gender in the population. The p -value of .32 suggests that it is reasonable to accept the null h ypothesis and conclude that there is not a difference in academic attitude based on gender. However, when school site was examined, the obtaine d [ F (4, 90) =2.87, p =.03] was judged to be statistically significant using a predetermined Type 1 error rate of .05. This indicates that the observed difference in atti tude based on school site might be different enough to conclude that academic differen ces would differ across school site in the population. The p-value of .03 suggests that if the null hypothesis was true, the probability of obtaining an F as large or larger th an the one obtained is .03. Since this probability is so small (less than .05), the null h ypothesis is rejected in favor of an alternative hypothesis that suggests at least one p air of population group means differ. When the interaction between gender and school site was examined, the obtained [ F (4, 91) =2.09, p =.09], was judged not to be statistically significa nt using a predetermined Type 1 error rate of .05. This indica tes that the observed difference in academic attitude based on males and females at one school site is not different enough from the difference in academic attitude of males a nd females at another school site to
136 conclude that academic attitude differences would e xist across gender and site in the population. Table 13 Variable Differences in Academic Attitude ___________________________________________________ _____________________ Source SS MS F Value p ___________________________________________________ _____________________ Gender 34.16 34.16 1.01 .32 Site 390.46 97.61 2.87 .03 Gender*Site 283.68 70.92 2.09 .09 Error 2751.38 33.97 n=91 ___________________________________________________ _____________________ Table 14 provides specific information in regards t o the sample size, mean, standard deviation, skewness and kurtosis values fo r the changes in academic attitude based on school site and gender. This table indicat es high variability amongst the subgroups in regards to sample size. Additionally, alt hough differences in scores are noted, there are no trends in scores based on gender or si te. The histogram shown in Figure 6 provides a visual display of the mean change scores in regards to academic attitude on the ERAS based on school site and gender.
137 Table 14 Differences in Academic Attitude in Regards to Gend er and School Site ___________________________________________________ ____________________ Site/Gender N M SD SK KU ___________________________________________________ ____________________ Dolphin Male 8 1. 4.41 -0.65 -0.06 Dolphin Female 9 -.56 4.48 -0.32 -0.38 Horn Male 13 .62 6.50 .29 1.70 Horn Female 4 9.5 8.27 0 -1.66 Carter Male 6 .67 3.01 -0.25 .88 Carter Female 7 -2.17 3.19 -1.44 2.44 Franklin Male 5 -2. 3.46 -1.92 3.67 Franklin Female 6 -2.17 3.19 -1.44 2.44 Lincoln Male 21 .19 8.40 .52 -0.59 Lincoln Female 12 -1.75 3.89 -1.05 .27 ___________________________________________________ ____________________
138 Figure 5. Difference in Academic Attitude in Regards to Gend er and School Site ! &'r(r)n*r+r,rr rr %n *nn
139 Tukey Test To determine more precisely which sites differed in regards to academic attitude from each other by a statistically significant amou nt, a Tukey test of all pair wise comparisons was conducted. The Tukey test can be us ed as a post hoc procedure to determine where the significant differences lie whi le maintaining the overall alpha rate at .05. The mean differences and confidence intervals around these differences are presented in Table 15. The results indicate that al though was variation amongst the sites, the variation could not be pinpointed to an exact v ariation between sites. This could be due to the small sample size at some of the sites.
140 Table 15 Tukeys Studentized Range for Difference in Academi c Attitude ___________________________________________________ _________________________ Site Comparison Difference between Means Confidenc e Limits ___________________________________________________ ____________________________ H-C 1.17 -4.82 7.16 H-D 2.53 -3.05 8.11 H-L 3.22 -1.63 8.08 H-F 4.80 -1.50 11.09 C-D 1.36 -4.63 7.35 C-L 2.05 -3.30 7.38 C-F 3.63 -3.03 10.30 D-L .70 -4.16 5.55 D-F 2.27 -4.03 8.56 L-F 1.58 -4.09 7.24 Total Attitude Finally, the researcher examined variable differenc es in total attitude. A summary table for the ANOVA on total attitude is provided i n Table 16. When gender was examined, the obtained [ F (1, 90) = 1.42, p =.24], was judged not to be statistically significant using a predetermined Type 1 error rate of .05. This indicates that the observed difference in total attitude for the male students is not different enough from the observed difference in attitude for the female stud ents to conclude that total attitude
141 differences would differ across gender in the popul ation. The p -value of .24 suggests that it is reasonable to accept the null hypothesis and conclude that there is not a difference in total attitude based on gender. When school site was examined, in regards to total attitude the obtained [ F (4, 90) =2.23, p =.07] was also judged not to be statistically sign ificant using a predetermined Type 1 error rate of .05. This indica tes that the observed difference in attitude based on school site is not different enou gh to conclude that total attitude differences would exist across school site in the g eneral population. The p -value of .07 suggests that it is reasonable to accept the null h ypothesis and conclude that there is not a difference in total attitude based on school site. When the interaction between gender and school site was examined, the obtained [ F (4, 90) =2.28, p =.07], was judged not to be statistically significa nt using a predetermined Type 1 error rate of .05. This indica tes that the observed difference in total attitude based on males and females at one school s ite is not different enough from the difference in total attitude of males and females a t another school site to conclude that total attitude differences would exist across gende r and site in the population.
142 Table 16 Variable Differences in Total Attitude ___________________________________________________ __________________ Source SS MS F Value p ___________________________________________________ __________________ Gender 152.35 152.35 1.42 .24 Site 955.82 238.96 2.23 .07 Gender*Site 977.81 244.45 2.28 .07 Error 8675.26 107.1
143 Table 17 Difference in Total Attitude in Regards to Gender a nd School Site ___________________________________________________ ___________________________________ Site/Gender N M SD SK KU ___________________________________________________ _____________________ Dolphin Male 8 1.75 10.50 -1.19 1.07 Dolphin Female 9 -2.67 8.81 -0.54 -0.77 Horn Male 13 .69 10.14 .15 2.40 Horn Female 4 15 17.53 .84 -0.79 Carter Male 6 -2.5 9.90 .66 -1.18 Carter Female 7 2.86 3.53 -0.60 -0.43 Franklin Male 5 -2. 3.46 -1.92 3.67 Franklin Female 6 -1.5 4.60 1.52 2.73 Lincoln Male 21 .67 13.32 .36 -0.27 Lincoln Female 12 -3.83 8.17 -.41 -0.88 ___________________________________________________ _____________________
144 Figure 6. Difference in Total Attitude in Regards to Gender and School Site ! &'r(r)n*r+r,rr rrr rrnn %n *nn
145 Table 17 provides specific information in regards t o the sample size, mean, standard deviation, skewness and kurtosis values fo r the changes in total attitude based on school site and gender. This figure indicates high variability amongst the sample sizes of the sub-groups. Additionally, although differences in scores are noted, there are no trends in scores based on gender or site. The histogram sh own in Figure 4 provides a visual display of the mean change scores in regards to rec reational attitude on the ERAS based on school site and gender. Summary of Quantitative Findings In summary, the findings from an investigation on t he fidelity data revealed that although the district conducted 177 fidelity checks on the implementation of Voyager Passport during summer school, because there is no norming data on the fidelity measure the researcher is unable to conclude that the inter vention was implemented with fidelity. An examination of the assumptions underlying a depe ndent measures t -test revealed no violations that would be considered pro blematic. A dependent measures t-test revealed little differences between mean test score s and levels of time. The results of the two-way factorial ANOVAs indicate that changes in r ecreational attitude and total reading attitude following summer school using Voyager Passport as an intervention with third-grade struggling readers were not statistical ly significant. Additionally, the interaction between recreational attitude and schoo l site was not statistically significant. When academic attitude was examined the results ind icated that changes in academic attitude were statistically significant based on th e summer school site. However, the results of the follow-up Tukey test indicated that although there was variation amongst the sites, the variation could not be pinpointed to where the exact significance lies. This
146 could be due to the small sample size at some of th e sites. Thus, overall, the results of the quantitative portion of the study revealed no signi ficant differences in students attitudes following summer school using a scripted literacy p rogram. Qualitative Findings The findings in this section address the following research question: What do elementary struggling readers perceive to be the ef fect of a scripted summer school literacy program on their attitude towards reading? The intent of this question was to target a group of 4-6 students from each classroom at the five summer school sites. The main purpose of the focus groups was to allow the s tudents to discuss and articulate in their own language their perception of how summer s chool using a scripted literacy program impacted their attitudes about reading. In an effort to observe the students attitudes dur ing summer school lessons, field notes were gathered during weekly classroom observa tions at each of the five sites. To complete the triangulation, focus groups were then conducted using these same targeted students the last week of summer school. Although c hildren are at the center of the No Child Left Behind Legislation, because politicians are representing them, often the childrens voices are excluded. This exclusion of the voices of children from the political culture of the public sphere has become c ommonplace (Kulynych, 2001, p. 259). While the survey portion of this study was vital, f ocus groups provided a way to meaningfully involve children in the research proce ss. Thus, the purpose of the focus groups was to get inside the heads (Langford & Mc Donagh, 2003; Johnson & Christensen, 2004) of the struggling third-grade re aders to determine their perceptions on
147 how a scripted literacy program had impacted their attitudes toward reading. Data collected through classroom observations and focus groups provided the researcher with a more in-depth understanding of struggling readers perceptions of their reading attitudes than would have been obtained from just t he ERAS (McKenna & Kear, 1990). This section begins with information on the focus g roup participants. Then the researcher presents classroom vignettes consisting of a narrative description of each of the summer school sites, targeted classrooms and in terviews with the targeted classroom teachers. Next, the findings from the field notes gathered during classroom observations are presented. Finally, the researcher presents the results of the focus groups. Focus Group Participants A nested portion of the quantitative sample was sel ected to participate in the focus groups. The selection criteria were originally base d on students initial attitude score on the ERAS (McKenna & Kear, 1990). The researchers o riginal intent was to intentionally select participants for focus groups whose initial full scale raw score attitude on the ERAS (McKenna & Kear, 1990) represented three disti nct levels (0-40 low, 41-60 average, and 61-80 high). When selecting these stud ents for each of the individual focus groups, the researcher attempted to select students with different levels of attitude scores. However, because the district wanted to have interr uptions in classrooms kept at a minimum, they asked that the focus group participan ts come from just one classroom at each individual site. This was challenging because of the variation in numbers of retuned consent forms at the different sites. The site with the most returned consent forms was Lincoln where 34 students had permission to partici pate in the study. Two sites (Carter and Franklin) had just 12 students with permission to participate in the study. This
148 variability in regards to site made it impossible t o select a classroom at each school site with participants whose scores on the ERAS (McKenna & Kear, 1990) varied according to the selection criteria. Additionally, only 10 of the 91 students who had pa rent consent to participate in the study had initial attitudes in the low range. T his further complicated the selection process, and meant that ultimately, of the 22 child ren involved in focus groups, only two students (10%) had initial low attitudes. Therefore the focus group participants were not always made up of two students with high attitudes, two students with average attitudes and two students with low attitudes as measured by the initial ERAS (McKenna & Kear, 1990). This also meant that not all of the children had strong verbal skills. These complications in recruitment of focus groups partic ipants had an effect on the group dynamics and ultimately influenced the content of t he focus groups. Specific information on each of the targeted focus group participants is provided in Table 20. The literature on focus groups provides varying adv ice in regards to the ideal size for focus groups (Morgan & Krueger, 1997). When con sidering group size, Edmunds (1999) recommends that researchers work with mini g roups of five or six participants. Although the researcher intended to have five to si x students in each of the focus groups, in the end focus groups ranged from three to five p articipants. The small groups reflected the practicalities of recruitment and last minute d rop outs due to absences. Conversely, the focus group made up of just three participants was tiring for all involved and was less of a focus group and more of an interview. In all, although 24 students were targeted and observed, only 22 students from five summer school sites took part in the actual focus groups (see Table 18).
149 Table 18 Number of Focus Group Participants by Site ___________________________________________________ ____________________ Site Teacher Number of Students ___________________________________________________ ____________________ Horn Mr. Owl 5 Dolphin Mrs. White 3 Carter Mrs. Fields 5 Lincoln Mrs. Smith 5 Franklin Mrs. Golden 4 The 22 students selected for focus groups varied i n regards to school site, ethnicity, gender, previous retentions, and initial and final ERAS raw scores (McKenna & Kear, 1990) (See Table 18 and Figures 7 and 8). I n regards to gender 45 percent of the participants were male and 55 percent of the partic ipants were female. When ethnicity was considered, Hispanic students represented 55 pe rcent of the focus group participants. The ethnicity of the remaining participants was Cau casian (18 percent), AfricanAmerican (23 percent), and Korean (4 percent). In r egards to initial attitude as measured by the ERAS, nine percent of students had an initia l attitude that was low (0-40), 36 percent had an initial attitude that was considered average (41-60), and 55 percent had an initial attitude that was considered high (61-80).
150 Table 19 Information on Focus Group Participants (n=22) ___________________________________________________ ___________________________________________________ _________________ Name ID# Site Gender Ethnicity Prior Retentions Initial ERAS Final ERAS Diff ___________________________________________________ ___________________________________________________ __________________ Jill 24 Horn Female Hispanic No 48 (average) 52 +4 Yasmin 29 Horn Female Hispanic No 64 (high) 63 -1 Mark 23 Horn Male Hispanic Yes 71 (high) 74 +3 Alex 22 Horn Male Hispanic No 46 (average) 53 +7 Eriberto 21 Horn Male Hispanic No 47 (average) 49 +2 -----------------------------------------------------------------------------------------------------------------------------------------------------Allison 54 Dolphin Female Caucasian No 53 (aver age) * Daisy 55 Dolphin Female African American Yes 38 (low) 27 -11 Peter 56 Dolphin Male Hispanic No 60 (average) 67 +7 ------------------------------------------------------------------------------------------------------------------------------------------------------Mary 2 Carter Female Hispanic Yes 71 (high) 7 4 +3
151 Table 19 Continued Information on Focus Group Participants (n=22) ___________________________________________________ ___________________________________________________ __________________ Name ID# Site Gender Ethnicity Prior Retentions Initial ERAS Final ERAS Diff. ___________________________________________________ _________________________________________________ Jane 5 Carter Female Hispanic Yes 67 (high) 69 +2 Jesus 4 Carter Male Hispanic Yes 43 (average) 41 -2 Juan 1 Carter Male Hispanic No 67 (high) 66 1 Eric 3 Carter Male Hispanic No 66 (high) 56 10 ------------------------------------------------------------------------------------------------------------------------------------------------------Daphney 38 Franklin Female African American Yes 37 (low) 44 +7 Terrance 39 Franklin Male African American Yes 7 8 (high) 76 -2 Lucy 37 Franklin Female Hispanic No 73 (high) 7 2 -1 Rene 40 Franklin Female Korean No 78 (high) 76 -2
152 Table 19 Continued Information on Focus Group Participants (n=22) ___________________________________________________ ___________________________________________________ __________________ Name ID# Site Gender Ethnicity Prior Retentions Initial ERAS Final ERAS Diff. ___________________________________________________ _________________________________________________ Carl 56 Lincoln Male Caucasian Yes 60 (average) 67 +7 Allan 71 Lincoln Male Caucasian No 65 (high) 75 +10 Sydney 68 Lincoln Female Caucasian No 64 (high) * Shamika 70 Lincoln Female African American Yes 58 (average) 41 -17 Alyssa 72 Lincoln Female African American No 63 (high) 66 +3
Figure 7. Gender of Focus Group Participants Figure 8. Ethnicity of Focus Group Participants %n *nnnr ('r )r #r#nr .nrnr 153 Gender of Focus Group Participants Ethnicity of Focus Group Participants /nrn r -r0 r
154 Classroom Observations Observation was a valuable tool and an alternate so urce of data for triangulation with information gathered through focus groups and ERAS surveys (Adler & Adler, 1994). Although focus groups provided a snapshot of the students attitudes the last week of summer school, throughout summer school the rese archer spent time each week at the five different summer school sites observing target ed students in their summer school classrooms. The classroom observations lasted appro ximately 45 minutes and took place during varied times throughout the day. The purpose of the observations was to gather field notes which would offer support for the focus group findings and the survey. Additionally, these observations allowed the resear cher to see what it was like inside each of the targeted classrooms. Field notes were taken using a laptop computer during weekly classroom observations at each of the summer school sites for a total of four or five site visits per school in addition to the focus group th at was held the last week. When gathering field notes, the researcher used ane cdotal notes to record a description of the adults and children in the class room, activities that were taking place, conversations that were going on, and the attitude of the children during the observation (Bogdan & Biklin, 1998). Next, upon returning home, the researcher read through the anecdotal notes taken that day and added anything t hat was not included. From the anecdotal records, the researcher provided a genera lized description of the observations described in the anecdotal records (Tjora, 2006). T hen the researcher relied on a priori categories to code the observation and interpret th e childs attitude. Next, the researcher quantified the data by counting how many times the category was observed (Tjora, 2006).
155 Finally a cross case analysis of the findings was d one on the data to look for patterns across sites (see Appendix R). Field note Analysis Content analysis allows inferences to be made whic h can then be corroborated using other methods of data collection (Krippendorf f, 1980). When coding the data, since the researcher had already established the categori es of the observations the researcher relied on a priori coding. A priori coding is defin ed as, Coding where the categories are established prior to the analysis (Stemler, 2001). In the case of activity structures during summer school, the categories were based on a manda te from Just Read Florida that specifically stated which approved activities could take place during the summer school day. During the visits the researcher observed stud ents engaged in the following activity structures which were also the a priori categories: small group Voyager Passport lessons; Voyager fluency lessons where the students participated in timed reads; Voyager Passport whole group vocabulary lessons; literacy centers; independent reading; shared reading; workbook and teacher read aloud. Each of t hese activity structures are defined in Chapter 3. When examining the reliability of a priori modeling because it does not depend on established theories for support, it is importan t to ask questions such as whether or not the measure seems like a reasonable way to gain the information the researchers are attempting to obtain. Since the purpose of the clas sroom observations was to offer support for the focus group findings and the survey and allow the researcher to see what it was like inside each of the targeted classrooms, the measure was an appropriate way to provide the researcher with the information.
156 During the observations the researcher also recorde d the students attitude using a Likert-type scale emulating the one used in The Ele mentary Reading Attitude Survey (McKenna & Kear, 1990). The scale ranked the childr ens attitude as enthusiastic, engaged, indifferent, off-task or withdrawn (Append ix L). Case Vignettes Case vignettes can be used to highlight important a spects in teaching. In this study, the researcher organized the findings from t he classroom observations into one case vignette on each of the summer school sites. T hese vignettes included: the student enrollment at the site; the number of teachers at t he site; a description of the school and neighborhood where the school was located; a descri ption of the treatment the researcher received from the staff of the school (office, admi nistration, classroom teacher); a description of the physical environment of the clas sroom; a description of the activities the students were engaged in during observations as well as their attitude during observations; and information about the classroom t eacher including results of any interviews. Horn Elementary Mr. Owl There were 61 third-grade students who attended Ho rn Elementary School during the summer. Complete data were collected from 26 pe rcent of the students. In regards to gender, twelve male students and four female studen ts participated in the study. The students were taught by 6 different teachers. Horn Elementary was located in a rural part of town. During the regular school year, Horn Eleme ntary was a magnet school for Mass Communications Graphic Art and Design. Additionally the school held Title I status due to the high percentage of free and reduced lunch st udents who attended the school (85
157 percent). The school was rebuilt approximately ten years ago with a modern design. The buildings that housed the classrooms, media center, office, and cafeteria were arranged around a large, grassy courtyard. The grounds and s chool were all well kept and maintained with fresh paint and landscaping that ha d been cared for. When the researcher visited the school the office s taff was friendly and helpful. Although the researcher never had the opportunity t o meet with either of the site administrators, they were both supportive of the st udy by e-mail. During the second campus visit when the researcher was trying to sele ct a classroom to use for the study, the secretary helped the researcher to examine the clas s lists and determine which classroom had the most children with permission to participat e in the study. Additionally, the third grade teachers returned all surveys and permission slips to the office in a timely manner. The classroom was spacious and clean. The walls had a fresh coat of paint and the floors were carpeted in new carpet. There were wind ows along one wall, which allowed natural sunlight to stream into the classroom. The temperature of the classroom was comfortable. The desks were arranged in a large U s hape. Additionally, areas of the room were set up for small group teaching areas and cent ers. The bulletin boards had paper on them and were also simply decorated. The classroom the researcher observed in was taught by Mr. Owl, a middle aged Caucasian male. Mr. Owl was a ten-year veteran teac her who had made teaching his second career. His teaching experience came from te aching both the third and fourth grade. This was Mr. Owls third year teaching summe r school. Mr. Owl made the researcher feel comfortable in his classroom. When it was appropriate, Mr. Owl would talk with the researcher about how the students wer e doing.
158 When the researcher had the opportunity to talk wit h Mr. Owl he expressed his interest in struggling students and said, I enjoy working with the students in summer school because they are all on an equal playing fie ld. During the regular school year, the struggling students just get further behind. I beli eve in setting the kids up for success. During an informal interview Mr. Owl expressed that he was very pleased with the results he was seeing in summer school. He explaine d that the teachers at his school had grouped the children homogeneously based on their r eading ability. The teachers determined the students reading ability based on a n initial Voyager Passport assessment that assessed how many words the students could rea d in one minute. The initial assessment was given at the beginning of summer sch ool. The students in Mr. Owls class were the students who had scored the highest on the one minute timed read. Because of this Mr. Owl was not using the word stud y component of Voyager Passport just the Comprehension and Vocabulary components. M r. Owl stated that he had seen a lot of growth in vocabulary and fluency especially in the Hispanic students in his class. I observed five students in Mr. Owls class, two gi rls and three boys. All five of the students were Hispanic. Only one of the five st udents had been retained before. A total of 20 student observations were conducted in Mr. Owls classroom during different parts of the summer school day. Although the times of the observations varied, every time the researcher observed in Mr. Owls classroom he was teaching a Voyager Passport lesson. These lessons ranged from small group to wh ole group lessons. Additionally, the researcher observed Mr. Owl doing timed one-minute reads with the students. During the observations of the five targeted students, the students were either reading independently or engaged in a Voyager Passport groups.
159 In regards to attitude during the observations, dur ing 75 percent of the observations the students attitudes appeared to be enthusiastic or engaged. These students exhibited behaviors such as sitting on the edge of their chair during lessons, begging Mr. Owl if they could go next, following al ong with their finger as they read the text, making eye contact with Mr. Owl and smiling, and raising their hand during lessons. During the visits to Mr. Owls classroom the resear cher observed two students who were off-task. One student kept turning around and watch ing the researcher while Mr. Owl was teaching. The other student was rolling around on t he ground during independent reading time, and went into the bathroom for 10 minutes. Th e researcher did not observe any students in Mr. Owls class who were indifferent or withdrawn during the observations. Dolphin Elementary Mrs. White There were 54 third-grade students who attended Dol phin Elementary School during the summer of 2007. Complete data were colle cted from 34 percent of the students. In regards to gender, eight male students and nine female students participated in the study. The students were taught by five diff erent teachers. Dolphin Elementary was located in a neighborhood known for high crime inci dents in a suburban part of town. The school held Title I status during the regular y ear due to the high percentage of free and reduced lunch students who attended the school (88 percent). During the summer of 2007, Dolphin Elementary was i n the process of being rebuilt next to the current campus. The buildings t hat housed the classrooms, media center, office, and cafeteria were all inside with wide air-conditioned hallways connecting the different parts of the campus. The l ighting in the carpeted hallways was dim and the carpet was worn. There was a musty odor that lingered throughout the
160 school. There were tables and chairs in the center of the hallways for volunteers to work with students. The classrooms were all off of the m ain hallway. Teachers taught with their classroom doors open. Some of the classrooms contained windows. There were many boxes piled up in the hallways. Because a par t of the campus had become a construction site and there were construction worke rs busily working on the new building the entrances to the school kept changing depending on the construction that day. Campus entrances and parking locations varied each week depending on the construction. During each of the visits to the school the office staff on duty was always different. This meant that the researcher had to re introduce herself during each visit and explain the purpose of her visit. The researcher ha d the opportunity to talk with one of the site administrators on the first day of summer scho ol. During the meeting he told the researcher that although the district had asked tha t permission slips for the researchers study be sent home with summer school students prio r to summer school, he had decided not to send do so at his school because he did not think it was necessary. During the second visit to the school, the research er met with the same site administrator again, this time to ask for assistanc e in determining which classroom had the most students who had returned yes informed con sents. The site administrator could not provide the researcher with class lists of the third-grade classrooms. Instead he escorted the researcher down to the first third-gra de classroom, Mrs. Whites room. There only ended up being four students in Mrs. Whi tes Class who had permission to participate in the study. The third grade-teachers at Dolphin Elementary did not return all surveys and permission slips to the office in a timely manner. There was confusion amongst the staff
161 in regards to where to turn the materials in and wh en to administer the surveys. The researcher had to go to the individual classrooms a nd ask the teachers for results of the post survey since in many cases they had not been r eturned to the office. The targeted classroom at Dolphin Elementary School was taught by Mrs. White. Her classroom was small in comparison to some of th e other classrooms at the site. The walls of the classroom contained paint that had bec ome faded and discolored. The carpet on the floor was worn and stained. There were no wi ndows, however there were two doors. One of the doors led outside and had mud sta ins on the frosted glass. The temperature of the classroom was approximately 65 degrees, even though it was in the lower 90s outside. Many of the children wore bulky coats and sweaters inside the classroom. The desks were arranged in rows. The re was a large kidney shaped table at the front of the room and a rectangular shaped tabl e in the back of the room. Although there was a tape recorder in the back of the room, there were no literacy centers set up in the room. The bulletin boards did not have paper on them but there were some posters hanging on the boards as well as on the wall. Addit ionally, there was a dry erase board which contained a word wall. There were many boxes lined up around the outside perimeter of the classroom. Mrs. White was a middle aged Caucasian female. This was Mrs. Whites first year teaching summer school. There were a total of ten s tudents in her class, six girls and four boys. Mrs. White was very busy teaching during all of the classroom visits which made it difficult for the researcher to have an opportunity to talk with her. During one incident the researcher stood next to the kidney shaped tabl e and smiled during a Voyager Passport lesson waiting for Mrs. White to look up, but she never did. Rather, she
162 continued to teach and then announced it was time f or lunch. As the children headed out the door she said, I sure would like to visit with you! In comparison to the other classroom visits, the researcher had the most troub le communicating with Mrs. White. On one occasion while the students were working on phonics worksheets Mrs. White talked with the researcher about how the stud ents progress in summer school. During this visit Mrs. White expressed her frustrat ion that the children were at so many different levels. During the researchers observati ons the students displayed a flat affect. The majority of the talking that took place in the classroom was by Mrs. White. There was not ample wait time for the children to respond to the frequent comments and questions made by Mrs. White. For instance, during one visit she introduced a text by stating, While Im reading, Im making a picture i n my mind. Are you making a movie in your mind? The students just stared at her blan kly. No matter how much Mrs. White smiled and talked with the students, during all of the observations I never observed any of the students in her class smiling, laughing or talk ing. The students only responded to questions if they were called on specifically. The researcher observed four students in Mrs. White s class, three girls and one boy. Two of the students were African American, one student was Hispanic and one student was Caucasian. One of the students was abse nt during focus groups. Two of the four students had been retained before. One of the students was an ELL student with little experience using the English language. A total of 18 student observations were conducted in Mrs. Whites classroom during different parts of the summer school day. Du ring three of the four observations Mrs. White was teaching whole group Voyager Passport lessons. During these lessons
163 the ten students gathered around a kidney shaped ta ble. Because there was not enough room for the children to fit around the table, ther e were actually two rows gathered around the table. During the fourth lesson all of t he students were completing a phonics worksheet with a partner. While the students worked with a partner on the phonics worksheet the students smiled and talked with one a nother. In regards to attitude during student observations during 39 percent of the observations students attitudes were classified as enthusistic and engaged. During these observations the students talked with one another, smiled, read chorally, or raised their hand to volunteer an answer to a question. During 3 4 percent of the observations the students appeared indifferent. They did not comment smile or raise their hand. During 22 percent of the observations students were off task. This was displayed by not following along in their reading book, putting their head dow n during a lesson, crossed arms, no expression, and looking around the room. During 5 p ercent of the observations the students appeared withdrawn. One student, on two se parate observations, physically turned around during the lesson crossed her arms an d stared at the wall with a sad look on her face. Carter Elementary Mrs. Fields There were 65 third grade students who attended Car ter Elementary School during the summer of 2007. Complete data were colle cted from 18 percent of the students. In regards to gender the class was made u p of five males and seven female students. The students were taught by seven differe nt teachers. The school was located in a rural area and was a Title One School during the regular school year due to the high percentage of free and reduced lunch (90 percent).
164 The school was remodeled a few years ago and had ad opted a multicultural theme. The hallways were painted bright colors and each hallway spotlighted a different country. A two story media center was the heart of the school. The classrooms, media center, office and cafeteria were all arranged unde r one roof with different wings all built around the media center. The campus was clean and a ppeared to have new carpet and new paint. There were windows in the classrooms whi ch allowed natural sunlight into the classrooms. The school site was very clean. When the researcher visited the school the office staff was friendly and helpful. The researcher was always greeted by the same secre tary who always asked if she could help. As with Horn Elementary the researcher never had an opportunity to sit with the site administrators; however, the administrators were su pportive via e-mail. Also like Horn Elementary on the second visit the secretary helped the researcher to examine the class lists and determine which classroom had the most ch ildren with permission to participate in the study. Additionally, the secretary provided the researcher with a map and explained how to get to the spotlighted teachers classroom. The teachers at Carter Elementary were very cooperative and turned surveys and permission slips into the office in a timely manner. Each hallway in the school was dedicated to a diffe rent country. Mrs. Fields classroom was located in The Mexican themed Hallw ay. On my first visit to the classroom, I could smell buttered popcorn before I entered her doorway. The classroom was freshly painted and had new carpet. The bulleti n boards had nothing on them and there were only a few charts hanging up around the room. The lighting was dim in the classroom because only half of the lights were turn ed on. One of the corners of the
165 classroom was filled with packed cardboard boxes. T he desks were arranged in two groups with all of the boys in one group and all of the girls in another group. Mrs. Fields was a middle aged Hispanic female. She spoke often to the researcher and always welcomed the researcher to the room duri ng each visit. Mrs. Fields frequently served food to her students. Additionally, she alwa ys offered candy, pretzels, popcorn or whatever treat the class was enjoying to the resear cher. Mrs. Fields taught fifth grade, also at Carter Elementary School, during the school year. She had been teaching for twelve years and had taught summer school previousl y. When the researcher had an opportunity to informall y talk with Mrs. Fields she told me that although the students enjoyed the Voyager Passport lessons, she felt it was very rote to do Voyager Passport lessons all day long, therefore she felt it was im portant to supplement with other literacy activities. Howev er, during the observations the researcher did not observe any supplemental literac y activities. Additionally, Mrs. Fields told the researcher that giving the children freque nt treats and snacks helped to keep them going. Every time the researcher visited the children were eating some type of snack. The researcher observed five students in Mrs. Field s classroom, two girls and three boys. All five students were Hispanic. Three of the five students had been retained in first grade. A total of 5 observations and 30 st udent observations were done by the researcher in Mrs. Fields Classroom. The researcher observed Voyager Passport taught whole group with all of the students in their seats as well as Voyager Passport taught small group with some of the students gathered arou nd a kidney shaped table. Additionally, the researcher observed the students partner reading.
166 In regards to attitude, the results indicated that the students were enthusiastic or engaged during 83 percent of the observations. This was evidenced by students raising their hands, smiling, being on task during games an d offering answers to questions. None of the targeted students were withdrawn during any of the observations. An additional 6 percent of the students were off task during indepe ndent reading. One played with flashcards during a whole group lesson. The other s tudent spent a long time in the bathroom during a partner read. An additional 6 per cent of students were indifferent, one during Voyager Whole Group and one during partner reading. Lincoln Elementary Mrs. Smith There were 80 students who attended Lincoln Element ary School. Complete data were collected from 43 percent of the students. Lin coln Elementary school had more returned yes informed consent papers than any other site. In regards to gender there were 23 male students and 11 female students. Lincoln El ementary School was located in a suburban neighborhood. The school was recently torn down and rebuilt. The new modern facility was very spacious. The entire building was two stories with high ceilings and large classrooms. Everything in the building was ne w including the furniture. The school was the only site that did not have Title One statu s during the school year. The office staff was very attentive to the research ers needs during the six weeks of summer school. Whenever the researcher e-mailed the site administrator she would forward the e-mail to her secretary who would take care of any requests. When the researcher arrived at the school for the second vis it the site administrator was waiting for her. She had already asked her secretary to have cl ass lists ready. This made it very easy for the researcher to select a class to work with. The selected classroom had gathered 90
167 percent of the informed consents back. The site adm inistrator then walked the researcher upstairs to Mrs. Smiths classroom. When the researcher arrived in Mrs. Smiths classro om the first thing she noticed was that the classroom was decorated for summer sch ool. There were attractive bulletin boards, a word wall, and colorful posters posted ar ound the room. Unlike the other spotlighted classrooms, Mrs. Smith had adopted many of the classroom routines from the regular school year in her summer school classroom. For instance, she had helpers each day and a Helper bulletin board. Additionally, all of the children went to recess each day. The children completed a Jumpstart assignment rel ating to spelling patterns as they arrived each day. Each day Mrs. Smith would post th e days agenda at the front of the board. During the regular school year Mrs. Smith wa s a third grade teacher in an upper middle class neighborhood. She had taught summer sc hool many times before. Mrs. Smith had been teaching for 33 years and had e xperience teaching a variety of grades. Her knowledge of curriculum was evidence d by the way she asked higher order thought provoking questions to the students d uring lessons. The responses the students gave to questions offered further support for what the children were learning in school. She was a very thorough teacher which was e videnced by the use of the entire Voyager Passport program with all of her students, even the optiona l Word Study component and Progress Monitoring. Mrs. Smith always spoke to the researcher when she was observing and offered her views and opinions on summer school freely duri ng each of the visits. When the researcher had the opportunity to talk with Mrs. Sm ith about summer school she said that the students enthusiasm and attitudes increased th roughout the day and that they were
168 very quiet and withdrawn in the morning, but by the end of the day they were focused and energetic. When the researcher asked Mrs. Smith how Voyager Passport was going, she began to talk about the different components of the program. She first explained to the researcher that she was using the progress monitori ng component with all of her students even though it was not required in summer school. T he progress monitoring component allows a teacher to measure a childs fluency growt h every five lessons. She was also using the optional word study or phonics program wi th all of her students, even though Voyager recommended that this piece only be used with stud ents who scored below a 44 words per minute on a grade level reading passage. Five students were observed in Mrs. Smiths room, two boys and three girls. Three students were Caucasian and two students were African American. Two of the students had been retained in first grade. The rese archer observed in Mrs. Smiths classroom 5 times as well as conducting one focus g roup. During the observations the researcher saw a variety of activities such as Voyager small group, Voyager Timed read, Voyager whole group, centers, workbook activities and part ner reading. During each of the observations Voyager Passport lessons were taking place with either the whole cl ass or a small group. Therefore, the researcher was abl e to observe both Voyager Passport lessons as well as what the other children were doi ng during small group Voyager Passport lessons. In regards to attitude during 84 percent of the observations students appeared enthusistic or engaged. This was higher th an any of the other sites in regards to attitude. An additional 16 percent of the observati ons were indifferent or off-task. None of the observations revealed students who were with drawn.
169 Franklin Elementary Mrs. Golden There were a total of 76 students taught by six tea chers who attended summer school at Franklin Elementary School. Complete data were collected from 16 percent of the summer school students. In regards to gender th ere were five male students and seven female students. The school was located in a suburban area of town. Like many of the other schools included in the study, the school was rebui lt approximately ten years ago. The campus was mainly outdoors, with separate wings for the office, Media Center, and classrooms. The school was considered a Title One s chool during the school year due to the number of students who qualified for free and r educed lunch (78 percent). When the researcher visited the school the office s taff was friendly and helpful. During the second visit when the researcher was try ing to select a classroom to use in the study, the other site administrator was very helpfu l. He went through all of the class lists and helped the researcher to figure out what class to use. Then the site administrator walked the researcher up to the classroom and intro duced her to the classroom teacher. Although the third grade teachers returned all surv eys and permission slips to the office in a timely manner, the percentage of returned info rmed consents was only 12 percent making it one of the lowest schools in regards to s tudent participation in the study. The classroom was spacious, clean and decorated wit h posters and charts. The bulletin boards had paper on them and had been deco rated for summer. The room had new carpet and fresh paint and was spacious. There were ample windows and the lighting was optimal. The classroom teacher was friendly to the researcher and available to talk with her during each visit. The desks were arranged in a u-shape. Although Mrs. Golden
170 had only been teaching for four years, she had stro ng classroom management. This was evidenced through consequences the researcher obser ved for students not abiding by the classroom rules. Mrs. Golden had a different way of breaking up the summer school day then the other targeted classrooms. The day began with 2 sma ll group Voyager lessons. During the Voyager Passport lessons the children who were not engaged in a Voyager lesson visited literacy centers. Following the Voyager lesson Mrs. Golden team taught with the teacher next door. During this time the two classes listene d to a teacher read aloud as well as participating in shared reading lessons using suppl emental reading materials. Following lunch the class split back into two groups and had another round of Voyager lessons before going home. Mrs. Golden offered her feelings about Voyager Passport and how it was being used as a summer school intervention. She told the researcher that many of the children in her summer school class had already been using Voyager Passport as an intervention during the school year. Further, some of the childr en had attended summer school last year and had used the same Voyager Passport materials/kit during that time as well. Because of this Mrs. Golden felt that the children were tired of using it. She shared with the researcher her view that she really saw a change in attitude during Voyager Passport lessons. When commenting on doing Voyager Passport lessons during the afternoon she said, From 12:30 to 1:30 we do a Voyager lesson and its like pulling teeth. Additionally, Mrs. Golden felt that the sum mer school day was much too long. She felt that it should be over at 12:30 instead of 1:30.
171 Mrs. Golden said that in addition to Voyager she had to supplement with additional literacy activities. She said the studen ts really enjoyed writing supplements consisting of descriptive writing as well as lesson s using Venn Diagrams to compare and contrast different books. During my observations at Franklin Elementary School I often observed the ELL students received push in servic es from an ELL teacher. Additionally, the reading coach at the site pulled a small group of students out for remediation. This was the only site where the resea rcher encountered the reading coach or ELL teacher interacting with the children she wa s observing. Four students were observed in Mrs. Goldens cultur ally diverse class. The researcher observed one African American male and o ne African American female, a Korean female, and a Hispanic female. A total of fo ur classroom visits and 20 student observations were done in addition to the focus gro up. During those five observations the researcher saw activities such as Voyager Passport small group, Centers, Shared Reading and Read Aloud. Mrs. Franklin classroom was the on ly class where the researcher observed a shared reading lesson as well as a read aloud lesson. The observations from Franklin Elementary revealed that the majority of t he lessons observed were small group Voyager Passport lessons. Additionally, it was the only class where the researcher observed team teaching. In regards to attitude during the observed lessons, none of the students displayed enthusiastic attitudes during any of the observatio ns. However, 70 percent displayed engaged attitudes, 25 percent of the student observ ations were off task, and an additional 5 percent were withdrawn. The off task student obse rvations were not limited to one
172 teaching method, but rather were spread out amongst Voyager small group, centers, shared reading and read aloud.
173 Table 20 Student Observations During Summer School ___________________________________________________ ___________________________________________________ _____ Activity N Enthusiastic Engaged Indifferent Off Task Withdrawn ___________________________________________________ ___________________________________________________ ___ Voyager Small Group 37 (32%) 7 (19%) 17 (46%) 7 (19%) 5 (14%) 1 (2%) Voyager Timed Read 10 (9%) 6 (60%) 4 (40%) 0 0 0 Voyager Whole Group 36 (32%) 10 (28%) 11(31%) 7 (19 %) 3 (8%) 5(14%) Centers 5 (4%) 0 4 (80%) 0 1 (20%) 0 Independent Reading 5 (4%) 1 (20%) 1 (20%) 0 3 (60%) 0 Shared Reading 1 (.9%) 0 0 0 1 (100% ) 0 Partner Reading 8 (7%) 2 (25%) 3 (37%) 1 (13%) 1 (13%) 1(13%) Teacher Read Aloud 4 (4%) 0 3 (75%) 0 1 (25%) 0 Workbook 8 (7%) 5 (62%) 2 (25%) 0 1 (13%) 0
174 Cross Case Analysis The purpose of the classroom observations and the r esulting classroom vignettes was to gather field notes which would offer support for the focus group findings and the ERAS survey. Additionally, these observations allow ed the researcher to see what it was like inside each of the targeted classrooms. The fi eld notes provide a description of the different activities that the children participated in during summer school and their resulting attitudes. During this analysis, patterns emerged across the sites. Given the results of the five case vignettes, this section br iefly highlights the key similarities and differences in the patterns that emerged across eac h of the sites. Pattern 1: Teachers Thoughts on Voyager Passport One pattern that emerged involved the classroom tea chers thoughts on Voyager Passport Four of the five teachers interviewed felt that Voyager Passport was unable to stand alone in summer school and therefore they had to supplement the program with other materials. One of the teachers said, I am ha ving to supplement (the program) with additional literacy activities. The students really enjoy writing supplements consisting of descriptive writing as well as lessons using Venn D iagrams to compare and contrast. Another teacher offered support for supplementing Voyager by saying, The students enjoy the stories in the Voyager books, but its just too rote to do it all day lon g. I think its important to supplement with other literacy ac tivities. Yet another teacher felt that Voyager was not meeting the needs of her lowest students, My students are at so many different levels. For the kids that are struggling I had to pull some phonics work sheets. Many of the children in my class seem to lack phoni cs. A fourth teacher said, I am
175 using the entire program, even the optional compone nts and I still have to supplement. However, a fifth teacher felt that Voyager Passport was meeting the needs of his kids. When asked about Voyager Passport his response was, Even though some of the kids had Voyager during the school year, its not an issue in summe r school. They are still attending class and still seem interested. This pa ttern in regards to supplemental materials being used during summer school may impac t the study in regards to fidelity. If other methods were used in addition to Voyager Passport this may have played an important part in why the students attitudes about reading did not change. Pattern 2: Teacher Characteristics Another pattern that emerged throughout the classro om visits was in regards to the classroom teacher. All five teachers displayed strong classroom management skills. This was evidenced by the lack of behavior issues t hat occurred while the researcher was observing. Additionally, all five teachers were kin d and caring to their students. This was evidenced by the way they spoke to the children and by the tangible and intangible rewards they offered the children (smiles, hugs, ki nd words, candy, popcorn, etc.). Four of the five teachers appeared to be connected to th eir students. This was evidenced by the way they looked at their students when they spoke, answered questions, and when necessary redirected the students. For instance, du ring one observation the teacher was teaching a small group Voyager Passport lesson while another group of students were working on a vocabulary lesson at a literacy center in the back of the room. One of the childrens voices could be heard above all the othe rs. Without getting up from her group the teacher said, Daphney are you playing ring lea der back there? All I am hearing is
176 your voice. This is your last warning. I like that you are working together but you are way too loud. Only one of the five teachers observed seemed to be disconnected to her students. This teacher was highly energetic and spoke to her children in a fast lively manner. When they didnt answer her questions, she just kept tal king. For example, during one lesson the researcher observed the students were all sitti ng at a table around the teacher who was saying, While Im reading, Im making a movie in m y mind! Are you making a movie in your mind? As the students stared at her blankly s he kept talking. Now Im making a connection. Is anyone else making a connection? Th is pattern may have impacted the findings as well. If the children wanted to please their classroom teacher, this may have influenced the way they responded to the ERAS. Pattern 3: Activities During Observations During the summer a total of 113 observations were conducted. The researcher coded the activity structure that took place during each observation. Due to the a priori design, the activity structures were in place prior to the observations. During the observations students participated in: 37 Voyager small group lessons, 10 Voyager Timed Reads, 36 Voyager whole group lessons, 5 literacy center activities, 5 independent reads, 1 shared reading lesson, 7 partner reading a ctivities, 4 teacher read alouds, and 8 workbook pages. An interesting finding emerged from this pattern that may have impacted the study results. Voyager was being taught whole group during 32% of the classroom visits. Voyager is not intended to be taught whole group. Rather, Voyager Passport lessons are designed for use with groups of no mor e than 5 students. This information was given to teachers during the summer school training (Appendix O). This
177 fidelity issue may have impacted the fidelity of th e implementation which might have impacted the results. Pattern 4: Differences in Teaching Methods Another pattern that emerged during the cross case analyses involved the use of varying materials and teaching methods during summe r school. One of the teachers I observed was using a team teaching model where duri ng the morning she taught her class Voyager lessons but in the afternoon the two teachers tea m taught. I observed their team teaching model on one occasion. All of the stu dents were gathered together for a shared reading lesson which was taught by one of th e teachers to the entire group. During the lesson the other teacher and the ESOL summer sc hool resource teacher were observing. At another site, on the first day of sum mer school the children were put into homogeneous classes based on their fluency level as determined by how many words they could read in one minute on a grade level pass age. Although the researcher was unaware that this ability grouping structure was in place until mid-summer, the class the researcher observed at that site was made up of the highest students. This may have impacted the focus group findings since the student s in that particular focus group were likely to have higher reading attitudes. Pattern 5: Students Attitudes during Observations Attitudes were rated using a Likert-type scale that emulated the scale used in the Elementary Reading Attitude Survey (McKenna & Kear, 1990) and rated attitudes as enthusiastic, engaged, indifferent, withdrawn, or o ff-task (See Appendix R). When attitude was examined across each site the findings show that during 67 percent of the classroom observations, students were engaged or en thusiastic, 13 percent of the students
178 were indifferent during the observed activity. The final 20 percent were withdrawn (15 percent) or off task (5 percent) (See Figure 9). Figure 9. Summary of Attitudes of Classroom Observations of Targeted Students Focus Groups In an effort to achieve triangulation in the study, the final part of data collection involved direct interaction with selected strugglin g third-grade readers. The use of focus groups and classroom observations offered a more in -depth understanding of students attitudes about reading than could have been obtain ed through a survey alone (Barbour & Kitzinger, 1999; Billson, 1994; Edmunds, 1999; Lang ford & McDonagh, 2003; and Morgan, 1988). The participants in the focus groups were the same students the researcher had been collecting field notes on durin g weekly observations of their classrooms during the six weeks of summer school. " ! nr12n3r -r -rn 4rnnr 1+ 56r
179 Focus Groups were held the last week of summer scho ol at each of the five sites. The researcher served as an observer and an outside moderator conducted the focus groups. The outside moderator was a fellow female g raduate student who had been trained in qualitative methods and was accustomed t o working with struggling readers. Stewart and Shamdasani (1990) advise that young chi ldren are often more comfortable with a female moderator. In order to increase data similarity across sites, the moderator used a protocol (See Appendix K). The protocol prov ided a framework for the discussion. The nature of the discussion was to find out what t he students attitudes about reading were following summer school (Langford & McDonagh, 2003; Johnson & Christensen, 2004). This was accomplished through asking a serie s of six questions. These included: 1. Tell me about summer school. Which part of summer s chool do you like best? 2. Do you like reading? Why or why not? 3. What types of books do you like to read the most? 4. Does anyone read to you at home? 5. Have you noticed any changes in your feelings about reading this summer? 6. Tell me about Voyager Passport Focus groups were held in an empty classroom, confe rence room, media center, or empty hallway at the different research sites. Participants sat around a table, or in one case, on the floor in a circle. Focus group session s were tape recorded by the researcher. The length of the focus groups varied, depending on how verbal the participants were.
180 The shortest focus group lasted 30 minutes and the longest lasted 60 minutes. The mean time was 44 minutes. Following introductions, groun d rules and an initial warm-up which was a discussion on pets, the moderator began askin g questions. Data were recorded by the researcher using a small tape recorder. Additio nally, because recording equipment only records a limited amount of all behavior (Stew art & Shamdasani, 1990) it was important to take notes and to document behavioral data as well as verbal responses. Both the researcher and the moderator took notes during the sessions. Each focus group took on a personality of its own. Some of the groups naturally and sequentially answered one person at a time one question at a time (Dolphin n=3 and Horn n=5) while other groups (Lincoln n=5 and Frank lin n=4) offered a great deal of thought provoking comments on each question. These comments often led to further discussions. Additionally, the students at Lincoln (n=5) and Franklin (n=4) piggy backed off one anothers responses. The questions merely served as starting points for a much deeper discussion when it came to the students from Lincoln and Franklin. The students at Lincoln (n=5) and Franklin (n=4) di splayed group cohesiveness (Stewart & Shamdasani, 1990). This was evidenced by the way they were influenced by each others responses. Research by Shaw and Shaw ( 1962) examined different patterns of interaction between high and low cohesive groups Findings revealed that that high cohesive groups were more, cooperative, friendly, and praise worthy of each others accomplishments (Shaw & Shaw, 1962; Stewart & Sham dassani, 1990). The group cohesiveness shown by the focus groups at Lincoln a nd Franklin was in direct contrast to the focus group from Dolphin (n=3).
181 The moderator had to probe the students at Dolphin in order to get them to respond to the questions. This included probing sta tements from the Focus Group protocol (Appendix S) such as, Tell me more, I don t quite understand, can you explain what you mean? The students from Dolphin did not r espond to each others comments. They displayed a flat affect during the focus group Therefore, it is not surprising that the session at Dolphin was the shortest session. Althou gh these findings are upsetting, they are not surprising given the small number of partic ipants in this focus group. Analysis of Focus Group Data Following data collection, focus group data were tr anscribed by the researcher. These transcripts then formed the basis for further analysis. In regards to focus group analysis the process of analysis is the least agree d on and the least well developed. Further, an agreed on technique does not exist (Car ey, 1995). It takes interpretation and insight to develop the meaning of a focus group dis cussion (Stewart & Shamdasani, 1990). In an effort to examine the meaning of the f ocus group discussions and its implications for research on struggling readers at titudes, the results of the focus group discussions were coded using semantic content analy sis (Stewart & Shamdasani, 1990). Content analysis is defined by Krippendorf (1980) a s, a research technique for making replicable and valid inferences from data to their context (p. 21). The codes were words or brief phrases. However, it is important to under stand that, The recording or coding of individual units is not content analysis (p. 112) (Stewart & Shamdasani, 1990). Stewart and Shamdasani recommend the use of virtually any analytic tool (p.113 when analyzing focus group data. In this study, the rese archer used attribution analysis (Janis, 1965) to determine the frequency within which certa in objects were mentioned. Next, the
182 researcher converted the frequency tables into perc entages. Percentages are shown visually through pie graphs. Inter rater reliability was accomplished by having a second rater code 20 percent of the answers to the questions to check to see if the codes were the same as those assigned by the researcher. There were two instance s where there was a discrepancy. Specifically, both discrepancies involved students who gave an answer and then were probed. During the probe, they gave another answer. After collaborating, the researcher decided to accept both responses. The results of th e semantic content analysis are presented below and organized by question.
Figure 10. What part of summer school do you like 4rn'nrnr nr "!# 183 What part of summer school do you like best? 7+r'n n# $nn ,nrr 4rn'nrnr nr /n n 8 n'rn "!# $%r&''( n $nn "!#
184 The focus groups began with the moderator asking th e first question, What part of summer school do you like best? The three most frequent responses were timed reading (19 percent), independent reading (19 perce nt), and social aspects (19 percent). Additional students selected teacher read aloud, le arning, book on tape, pass test, Garfield test, and No response (5 percent) (see Figure 10). Some of the students offered academic responses. Fo ur students talked about timed reading as their favorite part of summer scho ol. When asked about her favorite part of summer school Mary replied, The timed reading. You can read how many words in a minute. You know to beat your time. Allan support ed Marys response when he said, My favorite part is the part where you get timed r eading! Daphney selected teacher read loud as her favorite part of summer school, M y teacher read us a really good book about summer camp out loud. I liked that. Other students answers revolved around learning. F or instance, Rene said, The games and teachers are all nice, you get to learn a lot, we get to learn reading. Karyssa said, You can learn more. Shamika offered a matur e response when she said, Summer school is not just about recess its the fact that you are learning how to read, like last year I had trouble learning how to read so summer school gives me the opportunity to learn how to read. Other students enjoyed the social aspects summer sc hool had to offer. For instance Daphney said, I enjoy reading and I still go to recess. I can see my friends still and I like reading to my friends. Lucy interjected and said, Yeah, I get to meet new people and I can find new books and I can go to the library and get them. I like to go the
185 public library. I like the mall library too (She wa s referring to the bookstore) and you get to buy the books there. Other students answers reflected the real reason t hey were at summer school, to pass the Stanford 10 test. An interesting conversat ion about how focused the children were on the Stanford 10 test evolved when this ques tion was asked to the students at Lincoln Elementary. Carl stated, Well my friends s aid that summer school was prison but I say its not because you get to learn and you get to pass the test on July 13th. Sydney added, You know whats funny is that is tha t the test is on July 13th and a lot of people say thats a bad luck day. Shamika then sai d, Friday the 13th is only bad luck in the nighttime. We are taking the test in the mornin g. Then they all said, Yeah, it will be OK in the morning. This conversation reveals the p ressure that some of the children were feeling to pass the alternative test at the en d of summer school. The students all knew that the only way they could be promoted to fo urth grade was to pass the test.
Figure 11 Do you like reading? Why or why not? #r$""r$ 9n6r nn2r 186 Do you like reading? Why or why not? ,nrr:/n rr 4rnnr7+ .nn'n'r0 #r$""r$ %r&''( 9n6r nn2r ,20 4rnnr7+ .nn'n'r0 #r$""r$
187 The second question asked students Do you like rea ding? Why or why not? Twenty students responded Yes, they liked reading. Only two students stated No, they did not like reading. However, although an initial yes/no answer did not appear to be hard for the children, many of them had a very hard time elaborating as to why they like or did not like reading. In fact, of the students w ho responded yes, 13 provided no elaboration. Four students stated learning and getting informati on as the reason they liked to read. Rene said, I like to read because you can le arn about stuff and you can also learn about people and about things. Carl said, It give s you more information and you can learn better. Other students cited books as he reason they liked to read. Two students stated that books kept them company when they were alone. One of them, Juan elaborated, Reading keeps me company because its the only thi ng I can do at home because I have no brothers or sisters. Another student, Rene stat ed that she liked to read at night because she had a hard time sleeping. She said, I get bored because I dont sleep very much and I can read. Shamika said, I like reading because now I can relate to books. Last year I read the same book over and over and ov er and I got caught just in that one book. Some students said that although they liked reading sometimes it kept them from other activities. Daphney said, I love reading but sometimes I just want to watch TV and my Mom says Daphney did you read? Turn that TV off and go read now! Lucy said, I like to read but sometimes I just dont want to rea d. Like I want to go outside and play with my friends.
188 The two students who responded no when asked if the y liked to read had differing reasons as to why they did not like to read. Terran ce relied, Well, I like reading a little. Its just-I dont like to read because I run out of breath. When asked if he like to read Jesus replied, Not really because it depends on wh at Im reading. Like, if its boring. Here we read interesting books, but are home, well they are boring.
Figure 12. What types of books do you like to read? 5+2+:$n 'n' 8r *r 8 n'rn ;0n "!##$ 189 What types of books do you like to read? n7+ %
190 The third question asked students what types of boo ks they liked to read. This time 21 out of 22 students provided a response. Som e of the responses offered specific authors and series that the students enjoyed readin g. Realistic fiction was a favorite genre among the students. 22% of the students selected r ealistic fiction and continuously mentioned Junie B. Jones as a favorite series. Lucy said, I like reading Pee Wee books and I love Harry Potter but my brother helps me rea d that because its hard. And I.. I like reading just for the books. Six students stated th ey liked mystery books. Many of those six specifically talked about the Magic Tree House series. Daphney said, My teacher got me into reading The Magic Tree House Books and I al so read Junie B. Jones and Judy Mooney. So I like to read all kinds of books but I have to know a lot about the book first before Ill start reading it. Four students loved picture books and specifically mentioned Dr. Seuss books and other rhyming books. Pedro had a hard time verbaliz ing his favorite book, but with a great deal of probing was able to describe Green Eggs and Ham Eric also said, I love Dr. Seuss books. Green Eggs and Ham is my favorite. You can practice reading it really fast. Other students stated non-fiction books such as sci ence books were their favorites. Juan said, I like dinosaur books. Some of the stu dents commented about books with characters from TV such as Hanna Montana and Sponge Bob Square Pants. Still other students loved magazines. Alyssa stated that she liked Weekly Reader magazines. When asked what her favorite books were Sydney relied, Funny and magazines. I like magazines more than books. I like a magazine called M. It stands for Music, Movies and More. And I also read J14. Its a magazine for all ages. Shamika
191 added to what Sydney was saying and said, I like m agazines better than books too. I like Nick (Nickelodeon) magazines. Jesus agreed with th e girls when he said, I like to read magazines, like the ones with BMX magazines. Once again testing emerged in this question when Sh amika stated that she loved the workbooks she did with her Mom to get ready for the test. When asked what types of books he liked Terrance replied, Voyager . Rene also mentioned Voyager books. She said, I like the Voyager books also because we learn the words. And we lear n a lot of things in Voyager books and thats all.
Figure 13. Does anyone read to you at home? =rn 8 rr$%r&''( 192 Does anyone read to you at home? n 7n % rr$%r&''( 7n rr$%r&''(
193 When asked if anyone reads to you at home, 15 of th e students replied yes and cited Mom and sister most often. Brother and uncle were also mentioned. Sydney replied, My Mom reads me bedtime stories. Allan said, Im reading a book with my Mom called, The Dragon Slayer Academy . Daphney said, Yes, my Mom used to read with me when I was a baby. But now she still reads with me but not books more homework stuff. Rene said, I read to myself and my sister, shes 19 she helps me with the words. My parents can speak English and Korean. Thirty-two percent of the students stated that no o ne reads with them at home. No one sighted their father as reading to them.
Figure 14. Have you noticed any changes in your fe >4?2nnr6> >$nnn'n n> 194 Have you noticed any changes in your fe elings about reading ? >4+n> >4?2nnr6> >$nnn'n n2+ nn )*rr rrr r$%r&''( ? >4+n>
195 The fifth question was the most important question asked during Focus Groups and perhaps the most difficult question for the chi ldren to answer. The children were asked, Have you noticed any changes in your feelin gs about reading this summer? When students were asked if they had noticed any ch anges in their feelings about reading this summer 17 students responded yes, four student s responded no and one student offered no response. However, what was really inter esting were the comments when the students made when asked about their changes in rea ding following summer school. Although the children had no trouble describing the ir feelings about reading, they had a difficult time determining if their feelings had ch anged during summer school. Overall their attitudes about reading during focus groups w ere very positive. In fact, 50 percent of the students responded with I like it. An additional 27 percent of students felt that thei r reading had improved as the result of summer school. Daphney said, Well I like d reading before but now I like it even more because of the books. Like this is my sec ond year here, so I like know these books and Im really good at them (Daphney had att ended summer school last year and was retained in third grade. She had used Voyager Passport last summer). Lucy, Well I felt really good about it. I liked reading; I just didnt like reading all day. Well it has changed a little over the past four or six weeks. I ts going good but well theres still some things to work on, like my reading is faster but I still need to learn more. I know my sounds but I still need to work on the really long sentences. Terrance added, Yeah, those long sentences are hard. Alyssa said, I used to not like to read, but now I know what to do. Shamika added Yes, my reading has got ten better. Im getting more physical and I know the who, what, when, and where now. I should be at a fifth grade
196 level, but Im not. I dont want to be at a kinderg arten level. Sydney said, I used to not even like to read magazines, but since Ive been he re Ive been making my Mom get me magazines. I didnt like to read before, but just a little, but now my Mom lets me read about High School Musical and hairstyles. Allan spoke about the timed reads, It (reading att itude) changed. During May I didnt like reading that much because it was timed but she (classroom teacher during school year) doesnt tell you the whole book (readi ng passage read as a timed read) and you couldnt read the whole book. But here my teach er reads the whole story to me first and then I get timed. Allan was referring to the f act that during Voyager lessons he received during the school year he never got to hea r how the passage ended because he couldnt read fast enough. In summer school even if he didnt get to the end, his teacher had already read the passage aloud, so he knew what was going to happen. Sydney had a similar experience during the school y ear at a different school. She piggy backed off Allans answer and said, My teach er did the same thing! But in summer school when shell time us for one minute an d if you dont finish it you can go up to her and shell tell you the rest. Today me an d Allan read like 201 words in one minute. Its almost the whole book! An additional 18 percent credited the summer school teacher while 5 percent enjoyed the book selection at summer school. Rene s aid, Yeah because when I was not in summer school I liked it, but I didnt really re ad that much. And then we went to summer school and my teacher helped me read a lot o f new words and a lot of new books and I like it more. Allan said, My teacher helped me to get better. Sydney offered an honest answer when she said, I like it but I just am happy when Ill be done because I
197 like summer. I like to sleep in. In my class my tea cher helped me. Before I would be like I dont want to read but now since she told us what to do and where to find the words I like it. Jesus said, Before I went to summer scho ol I didnt like to read, because I had to take the test where I have to match the questions t hen I dont like to read. Thats what I have to do at my home school; you know read the que stions on the computer.
198 Figure 15. Tell us about Voyager Passport @nr ,r n .nn'2Anr 0n $nnr (n'0'n n ;205 *nnn 2n nr nr 3nr3n nAnr An 4rnnr n /nn0 6nn 8n'rn nn %r&''(
199 The final question was asked to see what the childr en thought about Voyager Passport When the children were asked to tell us about Voyager Passport the responses were varied. However, what was very interesting wer e the comments the students made during the focus groups especially in regards to Voyager Passport The top three coded responses were timed reading, helps you pass the te st, and vocabulary words. When asked to tell us about Voyager Passport some students made factual comments describing the program such as, To be a b etter reader from reading and reading and over and over you know reading! You a nswer questions. Its reading long stories. Its vocabulary words. Its prefi xes and suffixes The guide tells you where to go. Its timed reading. Its fluency. You read books in the Voyager and theres a really big box and it has A, B and C. C i s where you write the words. A is like sounding out words. B is to help you spell words ba ckwards and forward Voyager is vocabulary words. You should learn the vocabulary w ords because they might be in the story. Voyager Passport its like a book that tells you where to go. Oh t hats the guide, oops. Its a story that you read and it helps you w ith the words and the suffixes and the pre-suffixes. When you read do targeted word study you can learn to spell the words and you can learn what the words mean. Once again testing emerged as a theme. It helps yo u pass the test. And It helps you get ready for the test and it lets you practice so youll know what to do on the test and youll pass it. You might want to read a lot before you take the reading test and you need to read the whole book in a minute to pass the test. You need to read 60 words a minute or you are not going to pass the FCAT test.
200 Voyager is something that you read to learn how to so that when you take the test the FCAT test you can readsome of the Voyager words on it. Other students used emotionally laden words to des cribe the program such as Maria who said, Im good at it. Jane who responde d, Voyager, well it makes you more comfortable reading like if you had company. Eribe rto described Voyager Passport as, Its to be happy. To pass--you know the test. The children all chuckled as Jesus responded, Its reading things over, and over, and over, and over. Juan sat up tall and raised his hands over his head and said, It helps you to be a faster reader, you know like the king of the world! Carl, Im really good at Voyager Passport because I have done it before. Allan however, had the most original an swer, You get to keep the whole box of books at the end of the year. (The Voyager materials come in student boxes. A box consists of all the workbooks needed for each child At the end of summer school the students got to keep the box and take their workboo ks home.) Summary of Focus Group Findings. The purpose of the focus groups was to gain a more in-depth understanding of childrens attitud es about reading. Following the group interviews, the researcher transcribed the tapes fr om the focus groups. Then the researcher read through both the transcripts and an ecdotal notes that were taken by the moderator and the researcher during the interviews. Next, using the framework provided by semantic content analysis, the researcher procee ded to code the responses to the questions into words or phrases. Finally, (Stewart & Shamdasani, 1990) the researcher determined the frequency that objects were mentione d and used descriptive statistics to report the findings.
201 When the responses to each of the questions were an alyzed the following patterns emerged. 1. In regards to reading attitude, focus group finding s revealed that following summer school using a scripted literacy program, 91 percent of the third-graders students liked to read. However, 59 percent of the students could not elaborate as to why they liked to read. Amongst those who did co mment on why they liked to read learning was the most recurring theme. 2. When students were asked about their favorite part of summer school, academic responses were most frequent (read aloud, timed rea ds, independent reading and learning). Additionally, 19% of the students commen ted on the social aspects of summer school. 3. Third-grade summer school students like to read a v ariety of genres. 4. Someone reads at home to 68% of the third-graders. When asked who reads to you, students responded Mom, sister, brother and un cle. No one said their Dad read to them. 5. When asked about whether or not their feelings abou t reading had changed during summer school, although 79 percent said yes the mos t common response was I like it. This finding again provided evidence that third-grade summer school students liked reading. When probed during focus gr oups about whether or not their attitude had changed, the children had a hard time differentiating whether or not their attitude had changed, they just knew they liked to read. 6. When students were asked about Voyager Passport emerging themes ranged from factual themes (timed reading, prefixes and suffixe s, vocabulary words, etc. ) to a
202 myriad of emotionally laden themes (feel more comfo rtable reading, it helps you pass the test, Im good at it, etc.). The top three codes that emerged were timed reading, helps you pass the test and vocabulary wor ds. 7. An additional theme that emerged during focus group questions revolved around testing. Students talked openly about the FCAT and The Stanford 10 Test. They spoke of the importance of passing the alternative test on the last day of summer school in order to go to fourth grade. Triangulation of Findings During this mixed-method study, various methods of data collection provided different advantages and disadvantages. Triangulati on is the use of several different research methodologies to research the same phenome non. In this study, triangulation was achieved through quantitative data from The Ele mentary Reading Attitude Survey (McKenna & Kear, 1990) and qualitative data gathere d from both Focus Groups and field notes collected during classroom observations. Cohe n and Manion (1986) define triangulation as an attempt to map out, or explain more fully, the richness and complexity of human behavior by studying it from mo re than one standpoint (p. 254). Finding #1: Although the reading attitudes of some third-grade struggling readers varied in summer school, a scripted summer school literacy program did not appear to have an effect on the reading attitudes of all third-grader s. The purpose of this mixed-method study was to deter mine if a scripted summer school literacy program would impact third-graders attitudes about reading. In this
203 mixed-method study survey data were used to determi ne students academic, recreational and total attitude about reading both before and af ter summer school. When total reading attitude from the ERAS survey (McKenna and Kear, 1 990) was examined, third-grade students had mean attitudes on both the pre-test ( M = 58.09) and the post-test (58.05) that would be considered average and were slightly above an indifferent attitude score. A score of 50 is considered indicative of a child who has an indifferent attitude about reading whereas a score of 0 would be the lowest sc ore and a score of 80 would be the highest score. Although the change in attitudes var ied across the different school sites, these results confirm that although some of the stu dents attitudes changed, an overall conclusion cannot be made stating that the students attitudes changed as a result of summer school using a scripted reading program. The se results are supported by the results of an independent measures t -test that showed no significant findings in regard s to the change in students recreational [ t (90) = -.036, p = .55], academic [ t (90) = .32, p = .61], and total reading attitude [ t (90) = -.03, p = .98] following summer school. When the focus group findings were analyzed for a c hange in reading attitudes, once again although some of the students felt their attitudes had changed, the majority of the students could not elaborate as to whether or n ot their attitude had changed. When students were asked during focus groups whether or not their feelings about reading had changed during summer school, although 79 percent s aid yes, the most common response was I like it. When probed during focus groups ab out whether or not their attitude had changed, many of the children had a hard time diffe rentiating whether or not their attitude
204 had changed, they just knew they liked to read. How ever, the focus group responses varied based on the school site. Students from two of the school sites (Franklin and Lincoln) expressed that their reading had improved following summer school. Stude nts at Lincoln expressed that their reading had changed following summer school. This w as evidenced by comments made during focus groups. Sydney said, Yes, I like it, before I didnt. I didnt like to read at all before summer school but now I like to read a l ittle bit and now I like to read magazines, with High School Musical and hair stuff Shamikas comment offers further support, Yes, before I was like reading is boring now I know who, what, when, where, and how and I think thats why its changed me. Al lan agreed, It changed, in thirdgrade I didnt like reading all that much. Carl ad ded additional support with this statement, Yes, I always get a 100 on reading. Now I get four right and I used to only get two right. Im really good at Voyager Passport because I did it last year too. The students at Horn also provided positive responses t hat indeed their reading had changed. Mark said, Yes, I read much faster now. I used to read like 60 words per minute, but now I read to the end of the story before the timer goes off. Alex said, I like to read more now. Further support for this finding on the inconsisten cy in the change in reading attitudes is indicated in the results of the classr oom observations which revealed that although during 67 percent of the classroom observa tions the third-grade students appeared engaged or enthusiastic, during 33 percent of the observations the students appeared indifferent or off-task (see Table 20).
205 Finding # 2: Many of the third-graders liked the academic aspect s of summer school When the results of the ERAS were analyzed using a Factorial ANOVA, the initial results revealed a statistically significan t difference in relationship to the difference in academic reading attitude following summer schoo l based on school site [ F (4, 90) = 2.87, p = .03]. However, when a follow up Tukey test was c onducted it was concluded that although there was a difference in the academi c attitudes of third-graders based on school site, the difference could not be pinpointed to a particular site. The focus group findings support this finding that many of th e third-graders liked the academic aspects of summer school. During focus groups when students were asked what their favorite part of summer school was, the majority of answers were academic reasons: read aloud (3), timed read (4), independen t reading (4), book on tape (1), and learning (2) were some of the responses. Four of th e students focused their discussion on timed reading as their favorite part of summer scho ol. When asked about her favorite part of summer school Mary said, The timed reading. You can read how many words in a minute. You know, to beat your time. Juan agreed w ith Mary when he said, Fluency, the one minute read..to see if I could reach the en d before one minute. Other students enjoyed independent reading. Eric said, Reading, b ecause you can read silently. Alex also said, Silent Reading. Alyssa said she liked summer school, Because you learn more. Other students liked it when the teacher rea d aloud. When asked about her favorite part of summer school, Daisy said, When t he teacher reads aloud to me. Yasmin agreed when she said, I like it when Mr. Rowley reads aloud to me.
206 These findings offer further support for the result s of the Factorial ANOVA which showed there was a significant relationship between academic reading and school site. However, when the classroom observations were analy zed, although all of the observations were conducted during academic activit ies, some of the students appeared indifferent, off task or withdrawn during the actua l classroom observations (see Table 20). When the academic activities were examined, t he students attitudes varied based on activity and site. Dolphin Elementary had the highe st number of student observations (11) where students were withdrawn, off-task or ind ifferent. All of those observations were done during whole group Voyager Passport lessons. Both Carter and Lincoln had zero withdrawn observations, two off-task observati ons, and two indifferent observations. At both sites the off task observations occurred du ring whole group Voyager Passport lessons. The indifferent observations were conducte d during Voyager whole group lessons, partner reading and workbook observations. Finding #3: Third-grade students attending summer school were f ocused on standardized testing. Another finding from the study concerned the stude nts focus on standardized testing. This finding was especially evident during focus groups. An interesting conversation about how focused the children were on the Stanford 10 test evolved when the students at Lincoln Elementary were asked What part of summer school do you like best ? Carl stated, Well my friends said that sum mer school was prison but I say its not because you get to learn and you get to pass the te st on July 13th. Sydney added, You know whats funny is that is that the test is on Ju ly 13th and a lot of people say thats a bad luck day. Shamika then said, Friday the 13th is only bad luck in the nighttime. We
207 are taking the test in the morning. Then they all said, Yeah, it will be OK in the morning. This conversation revealed the pressure t hat some of the children were feeling to pass the alternative test at the end of summer s chool. The students all knew that the only way they could be promoted to fourth grade was to pass the test. Testing also emerged during focus groups when the students were asked, What types of books do you like to read? This time Shamika talked about h ow much she enjoyed the workbooks she did with her Mom to get ready for the test. Whe n students were asked to tell us about Voyager Passport once again testing emerged as a theme. The followi ng comments offer support for this finding: It helps you pass the test. Its to be happy to pass the reading test It helps you get ready for the test and it lets yo u practice so youll know what to do on the test and youll pass it. You might want to read a lot before you take the r eading test and you need to read the whole book in a minute or you are not goin g to pass the FCAT test. Voyager is something you read to learn how to (read) so wh en you take the FCAT test you can read some of the Voyager words on it. Voyager helps you to practice to know what to do on the te st so you can pass it and know it all The focus on the test was also evident during class room observations. During the classroom observations the teachers were focused on the Stanford Test and talked openly about the test. For instance, during one whole grou p lesson Mrs. Fields said, Boys and
208 girls I cannot believe how your reading fluency has improved. You are going to be so ready to take the Stanford! During another observa tion Mr. Rowlie said to Yasmin, If you read like that when you take the Stanford Test you will blow them away! When observing a lesson in Mrs. Smiths classroom she wa s explaining how important reading speed was when taking the Stanford Test. If you ar ent reading 100 words a minute, when you go to take the Stanford Test, you wont ge t through the stories. Finally, when the results of the ERAS survey were a nalyzed, one of the academic questions asked the students, How to you feel when you take a reading test? When this question was examined by itself the results of the pre-test revealed a mean score of [ M = 2.42, SD = 1.19] for the pre-test and [ M =3.49, SD = 6.63] for the post-test. However, on a cautionary note, this change score represents a sma ll change in just one question out of 20 and is not necessarily statistically significant. T hus, when examining this question in isolation, although it appears that the students a ttitudes toward taking a reading test changed as a result of summer school using scripted literacy, it warrants further tests to determine if this finding is statistically signific ant. McKennas Model of Reading Attitude When considering attitudinal change it is important to refer back to McKennas (1994) model of reading attitude. McKennas model e xamines three principal factors influencing attitudinal change: (a) beliefs about t he outcomes of reading in light of the judged desirability of those outcomes; (b) beliefs about the expectations of others in light of ones motivation to conform to those expectation s; and (c) the outcomes of specific incidents of reading (McKenna et al., 1995). Simply stated, the McKenna model supports
209 the notion that, An individuals attitude toward r eading will develop over time principally as the result of three factors: normati ve beliefs, beliefs about the outcomes of reading and specific reading experiences (p. 939). Normative Beliefs Themes obtained from focus groups support McKennas model (1994) of reading attitude. McKennas Model specifies that normative beliefs play an important role in the development of attitudes. Specifically, the model p redicts, If a childs cultural environment encourages, models, and reinforces read ing, more positive attitudes should result (p. 941). The themes that emerged from the focus groups indicated that the children did have positive feelings about reading. In fact at the end of summer school 91percent of the students stated that they liked to read. When students were asked about their favorite parts of summer school academic resp onses were most common. Additionally, when the results of the focus groups, ERAS, and classroom observations are all combined it is also of interes t that overall the third-graders attitudes were average to begin with ( M = 58.09). This finding is confirmed by the results of the classroom observations which revealed that d uring 67 percent of the classroom observations, the thirdgrade students appeared eng aged or enthusiastic. However, of the 91 students who had permission to participate in th e study, only 10 had an initial low attitude. It is important to note that 73percent of the third-graders attending summer school did not have permission to participate in th e study. Research supports a lack of parental involvement with many struggling readers ( Jimerson & Kaufman, 2003). Thus, it is possible that the students who participated in t he study came from families where there
210 was parental involvement. Further support for this finding is shown from the results of the focus groups that revealed that someone read at home to 68 percent of the children. Third-graders liked the academic aspects of summer school. This finding was supported by the results a Factorial ANOVA that exa mined the relationship between academic attitude and school site. The initial resu lts revealed a statistically significant difference in relationship to the difference in aca demic reading attitude following summer school based on school site. However, when a follow up Tukey test was conducted it was concluded that although there was a difference in the academic attitudes of third graders based on school site, the differen ce could not be pinpointed to a particular site. Instructional Approach Although McKenna (1994) postulates that certain ins tructional approaches might harbor successful experiences, which would in turn lead to more positive beliefs about the outcomes of reading and contribute to attitude indirectly; previous research fails to support the notion that instructional approaches ca n influence reading attitude. McKenna, Stratton, Grindler and Jenkins (1995) reported no d ifference in the attitudes of one to five students taught using a whole language approach as compared with their peers who received reading instruction from a basal reader. Although the findings from the qualitative portion of the study did not show an in crease in mean scores from pre-test to post-test, the themes that emerged from the qualita tive portion of the study revealed that the students had a positive experience in summer sc hool. During Focus Groups the students were asked, Tell me about Voyager Passport . Although none of the themes
211 that emerged from this question describing Voyager Passport were negative, some evoked factual themes ( Voyager Passport is vocabulary words.) while others evoked emotionally laden themes (It makes you more comfort able reading.) Gender In regards to gender, the findings do not support previous research. Although previous studies have found that male students gene rally possess a more negative attitude towards reading than female students, this study fo und no significance in gender in regards to struggling readers attitudes about read ing (McKenna et al., 1995). Perhaps if a larger sample was used these findings would have be en different. Testing An additional theme that emerged from the focus gro up findings was that the students believed they were prepared to take the al ternative assessment at the end of summer school. This was evidenced by comments the s tudents made during focus groups. For instance when students were asked about Voyager Passport some of the responses were, It helps you get ready for the test and it l ets you practice so youll know what to do on the test and youll pass it. Age and Reading Attitude McKennas model also predicts that as children get older and they have more options available to them during their leisure time that their attitude towards reading will worsen (Anderson, Tollefson & Gilbert, 1985; Martin 1984). Some of the focus group responses offered support this theory. For instance when students were asked if their attitude about reading had changed during summer sc hool, two students offered their
212 perspectives. Daphney said, I love reading, but so metimes I just want to watch T.V. and my Mom says Daphney did you read? Turn off that TV and go read now! Lucy said, I like to read but sometimes I just dont want to rea d. Like I want to go outside and play with my friends.Sydney said, I like it but just a m happy when Ill be done because I like summer. I like to sleep in. These statements made by third-graders also offer support for research that shows struggling readers attitudes about reading decline as they get older (Ishikawa, 1985; Ross & Fletcher, 1989). Research supports that as students get older they become more involved in extracurricular activities and social obligations and show less interest in reading during free time (McK enna et al., 1995). Organization of Remaining Chapter The final chapter contains five sections. The first section provides a summary of the study. The second section describes limitations implications and conclusions derived from the research findings. The third section discu sses the recommendations for practice based on the study limitations, implications and co nclusions. The fourth section offers recommendations for future research. The fifth sect ion offers closing thoughts.
213 CHAPTER 5 SUMMARY, CONCLUSIONS, RECOMMENDATIONS AND IMPLICATI ONS Overview of Chapter The purpose of this mixed-method study was to deter mine if summer school using a scripted literacy program (Voyager Passport ) would impact retained third-grade struggling readers attitudes about reading. This c hapter contains five sections. The first section provides a summary of the study. The second section describes limitations, implications and conclusions derived from the resea rch findings. The third section discusses the recommendations for practice based on the studys limitations, implications, and conclusions. The fourth section offers recommen dations for future research. The fifth section offers closing thoughts. Summary of the Study Retained third-grade students in the state of Flor ida are encouraged to attend a district summer reading camp where they receive int ensive reading instruction using a research-based reading intervention program. Follow ing six weeks of summer school the students are given an alternative reading test. If they pass the test, they receive a good cause exemption and are promoted to fourth grade. In the district where this study was conducted the research-based intervention used duri ng the summer of 2007 was Voyager Passport a program that utilizes a scripted literacy format This mixed-method study explored and examined the effect of a scripted lite racy program on the reading attitudes of third-grade struggling readers.
214 Research Questions The intent of this study was to answer the followin g research questions: 1. What is the effect of a scripted literacy program o n the reading attitudes of elementary school struggling readers? 2. What do elementary school struggling readers percei ve to be the effect of a scripted summer school literacy program on their attitudes toward reading? Quantitative Findings The study lasted six weeks and was conducted at fi ve different school sites in a school district on the west coast of Florida where summer school was held during the summer of 2007. Because students were selected base d on their participation in summer school, the sample was a convenience sample. During the quantitative portion of the study students (n=91) were given the Elementary Rea ding Attitude Survey (ERAS) (McKenna & Kear, 1990). Complete data were collecte d from 91 students at five different summer school sites. The ERAS survey, whi ch was administered by the classroom teacher the first day of summer school an d the last week of summer school, provided raw scores for academic, recreational, an d total reading attitude. Following data collection the results of the ERAS s urveys were analyzed using descriptive statistics and a dependent measures t -test as well. Additionally, in an effort to assess the tenability of a chance explanation, the researcher conducted three 2x2 factorial ANOVAs which examined gender, school site, and the interaction between gender and school site.
215 Descriptive Statistics First, the researcher analyzed the distribution of attitude scores separately for recreational, academic and total reading attitude u sing descriptive statistics. The difference in mean scores did not show enough varia tion to be of practical importance. In regards to total attitude scores, the students had a mean score of 58.09 on the pre-test and a mean score of 58.05 on the posttest. A total scor e of 50 would reflect a score that was directly in the middle and could be interpreted as an indifferent score, thus scores of 58. 09 and 58.05 would represent average attitudes. Dependent Measures t-test When a dependent measures t -test was conducted to examine the difference in academic, recreational and total attitude scores fr om pre-test to post-test, the findings were not statistically significant in regards to re creational attitude [ t (90) = -.036, p = .55], academic attitude [t (90) = .32, p = .61], and total attitude [t (90) = -.03, p = .98]. Factorial ANOVAs Next, in an effort to determine if attitudes differ ed amongst gender and school site, the researcher conducted three separate 2x2 F actorial ANOVAs (recreational attitude, academic attitude and total reading attit ude) with alphas set at .05 for each effect. The results of the Factorial ANOVAs indicat ed that the changes in recreational attitude and total reading attitude were not statis tically significant with any of the subgroups. The only statistically significant findi ngs related to academic attitude and school site [ F (4, 90) = 2.87, p = .03]. However, a Tukey follow up test revealed t hat
216 although there was a difference in academic attitud e between the school sites, the variation could not be pinpointed to particular sit es. Qualitative Findings McQuire (1989) speculated that the tri-component vi ew of attitude could be measured more effectively through the use of open-e nded responses. Therefore, questions asked during focus groups were open ended. These op en ended responses revealed much more information about struggling readers attitude s about reading than the quantitative findings. To answer the qualitative question, a nes ted sample of the quantitative population was selected (n=22). To complete triangu lation, the qualitative portion of the study relied on both field notes gathered during cl assroom observations and focus groups. Five focus groups were conducted, one at each summe r school site the last week of summer school. The purpose of the focus groups was to get inside the heads (Langford & McDonagh, 2003; Johnson & Christensen, 2004) of s truggling readers to determine what elementary school struggling readers perceived to be the effect of a scripted summer school literacy program on their attitudes toward r eading. During focus groups the students were asked a series of six questions about reading, summer school and Voyager Following data collection, focus group findings wer e transcribed by the researcher. These transcripts then formed the basis for further analysis When the responses to each of the questions were an alyzed the following patterns emerged. In regards to reading attitude, focus grou p findings revealed that following summer school using a scripted literacy program, 91 percent of the third-grade students liked to read. However, 59 percent of the students could not elaborate as to why they liked to read. Amongst those who did comment on why they liked to read, learning was
217 the most recurring theme. When students were asked about their favorite part of summer school, academic responses were most frequent. Thir d-grade summer school students like to read a variety of genres. Someone reads at home to 68 percent of the third-graders. When asked about whether or not their feelings abou t reading had changed during summer school, although 79 percent said yes, the ch ildren had a hard time differentiating whether or not their attitude had changed; they jus t knew they liked to read. When students were asked about Voyager Passport, emerging themes ranged from factual themes to a myriad of emotionally laden the mes. The top three codes that emerged were timed reading, helps you pass the test and vocabulary words. An additional theme that emerged during focus group questions rev olved around testing. Students talked openly about the Florida Comprehensive Achie vement Test (FCAT) and The Stanford 10 Test. They spoke of the importance of p assing the alternative test on the last day of summer school in order to go to fourth grade In addition to the five focus groups, the researche r conducted classroom observations (n=113) with each of the focus group p articipants. The intent of the classroom observations was to offer support for the focus group findings. During the 22 visits 113 student observations were conducted. Res ults of the student observations supported the focus group findings and revealed tha t 67 percent of the students were engaged or enthusiastic during classroom observatio ns. An indifferent attitude was observed 13 percent of the time and students were o ff task or withdrawn 20 percent of the time. In regards to activities observed during cla ssroom observations, the researcher observed children engaged in: 37 Voyager small group lessons, 10 Voyager timed-reads,
218 36 Voyager whole group lessons, 5 literacy center activities, 5 independent reading, 1 shared reading lesson, 7 partner reads, 4 teacher r ead alouds, and 8 workbook pages. Limitations Conclusions and Implications There were numerous limitations that arose during t his study. The first limitation arose because the researcher was actually a third-g rade teacher in the targeted district; numerous problems arose as a result. First, because the district saw it as a conflict of interest for the researcher to conduct the fidelity checks herself, the fidelity checks were conducted by the assistant principals, reading coac hes, and district office personnel. Additionally, the researcher was unable to modify t he fidelity measure. Next, prior to summer school during a meeting with the Director of Elementary Education, a request was made by the district that in order to impact th e least amount of classrooms, for the researcher to limit participants for the focus grou ps to one class at each summer school site. There were direct implications from this deci sion. The researcher had intended on using a nested portion from the quantitative popula tion in the qualitative portion of the study. Specifically, the researcher had intended on focus group participants consisting of two students with high initial attitudes (61-80), t wo students with average initial attitudes (41-60), and two students with low attitudes (0-40) An additional complication arose because there were only 10 students with an initial low attitude with permission to participate in the study. When the researcher was l imited to only using one classroom at each site, it became impossible to find a classroom that met the initial selection criteria. Further, by limiting the focus group participants t o just one classroom at each site the study may not have captured an accurate representat ion of the summer school students.
219 Another limitation involved the participants themse lves. During the Spring of 2007 approximately 547 children in the targeted dis trict qualified to attend summer school based on their FCAT score. It is interesting to note, that among those 547 students, only 61 percent of the students (336) actually atte nded summer school on the first day. Additionally, of the initial 336 students, only 115 students had permission to participate in the study. Furthermore, complete data were colle cted on just 91 students (27 %) attending summer school. Therefore, in the end the actual students who participated in the study were students whose families signed the n ecessary paperwork to enroll them in summer school, made sure that they attended summer school, and made sure they were present for the alternative assessment the last day of summer school. This meant that the 91 participants of this study c ame from families who were more involved in their education and may not have b een an accurate representation of the population of struggling readers. This factor like ly contributed to the findings and helps to explain the fact that overall the third-graders attitudes did not change following summer school. An additional limitation involved the researchers o wn view of scripted literacy programs. Throughout this study the researcher used reflexivity to monitor any biases toward scripted literacy. Reflexivity was one way t he researcher was able to explore the ways in which her involvement with literacy influen ced her research.Because the researcher was also a third-grade teacher in the ta rgeted district, she was mandated to use Voyager Passport as a remediation tool with her own struggling read ers. As the researcher analyzed the findings of the study, and reflected on her own knowledge of
220 what methods have been proven effective when workin g with struggling readers, it became difficult for the researcher to continue to use the intervention tool. Additionally, another limitation occurred when the researcher was reporting the findings of the fidelity measure. Because the resea rcher worked in the district where the study was conducted, at times it was difficult for the researcher to provide the reader with all of the findings, which were often controversial This was especially true in the lack of fidelity in the implementation of Voyager Passport that the researcher reports on during summer school. Recommendations for Practice The results of this study, coupled with the underst andings provided in existing research, leads to some recommendations for teacher s and district administrators. As discussed in the review of literature, although und er Reading First scripted literacy programs are being used as interventions to assist the struggling reader, not everyone supports scripted literacy in the classroom. Richar d Allington (2002) summarizes his view on scripted literacy with this statement, A v eritable trove of scientific research tells us that effective teaching is not standardized and cannot be scripted (p. 28). A students reading attitude plays a key role in wh ether or not he or she becomes a competent reader (Anderson, Hiebert, Scott, & Wil kinson, 1985; Mathewson, 1994; McKenna, 1994). Children who like reading tend to r ead more, and this develops their reading ability. There are two main goals involved in the teaching or reading: instill in students the necessary skills to read effectively a nd to develop a sense of enjoyment toward reading (Sainsbury, 2004). Ultimately, this study hoped to answer the question, Does our instructional method teach reading at the expense of enjoyment?
221 Based on the qualitative findings, when McKenna s model of reading attitudes is applied to the study, upon first glance one woul d expect attitudinal change. However, the quantitative portion of the study did not show a change in the reading attitude of third-grade struggling readers as a result of a scr ipted literacy program ( Voyager Passport ). Rather, the results of the quantitative portion of the study revealed that thirdgrade students attitudes about reading following s ummer school using a scripted literacy program remained average. There are possible explanations as to why there wer e no significant findings in regards to attitudinal change. The first explanatio n concerns the fidelity of the implementation. Fidelity of Implementation One possible reason that the students attitudes di d not change is related to the fidelity of implementation. In regards to education al research, if there is a high rate of fidelity in the implementation of a program, then t he administration and staff can rule out this variable in regards to student achievement. Fi delity of implementation is the actual presentation of instruction the way it was intended to be delivered (Gresham et al., 2000). Specifically, it is the adherence to the interventi on protocol in comparison with the original program design (Mihalic, 2002; Mowbray et al., 2002). A number of factors from this study question the fidelity of the implem entation of Voyager Passport Lack of norming data. Although Voyager Passport is one of 101 supplemental intervention reading programs that have been review ed by the Florida Center for Reading Research (FCRR), and reviewed under Reading First, Voyager Passport lacks any norming data on their fidelity measure on either th e Voyager website, or on the Florida
222 Center for Reading Research website (FCRR). After c ontacting Voyager Expanded Learning Systems directly the researcher was told that the company had no norming data on the fidelity measure. Additionally, the research er contacted the district where the study was conducted. The researcher was told that the dis trict did not have any norming data on the fidelity measure as well. Because there is no n orming data on the fidelity measure, it is impossible to know if the fidelity measure actua lly measured what it was designed to assess. Group size. When the researcher observed in the summer school classrooms, during 32 percent of the observations summer school teachers were implementing Voyager Passport whole group. Voyager is not intended to be taught whole group. Rather, Voyager Passport lessons are designed for use with groups of no mor e than 5 students. This information was given to teachers du ring the summer school training (Appendix O). This finding may have impacted the fi delity of the implementation. Prior exposure to program. Another caution in the fidelity of the implementat ion concerns the way Voyager Passport was implemented with at risk third-grade strugglin g readers. In the targeted district Voyager Passport is currently being used as a supplemental reading program with students who are below grade level in reading. Some of the students who attended summer school had alre ady used the program during the regular school year, for others it was their first encounter with Voyager Passport When used during the school year students receive 4 less ons a week in addition to the 90 minute daily reading block. Some of the students who recei ved Voyager Passport remediation during the school year at their home school receive d the same exact instruction during summer school using the same curriculum. Additional ly, there were some students who
223 had attended summer school last year and used Voyager Passport were retained in third grade and received Voyager Passport as an intervention during the school year and were receiving Voyager Passport for a third time. Sample Because only 27 percent of the summer school studen ts participated in the study, and given the fact that only 10 percent of them had an initial low attitude as measured by the ERAS (McKenna & Kear, 1990), its possible that the struggling readers who did not have permission to participate in the study were th e students with low reading attitudes. This means that the sample was not representative o f the population. Length of Intervention An additional reason why the students attitudes di d not change may relate to the length of intervention. Specifically, it may be tha t the length of the intervention (six weeks) was not enough time to measure change. Supplementing of Curriculum In addition, during interviews with classroom teac hers the researcher learned that four out of five teachers spotlighted during classr oom observations did not feel Voyager Passport could stand alone and were supplementing the progr am. Measurement of fidelity is especially crucial with studies that seek to provide evidence for the effectiveness of an intervention ( Mowbray et al., 2002). Additionally, in regards to educational research, if there is a high rate of fidelity in the implementation of a program, then the administration and staff can ru le out this variable in regards to student achievement. Developing measures of fideli ty, validating them, and using them can be intensive and costly. However, the other opt ion is to recommend programs that are
224 either not effective, or are only effective if impl emented in a particular way (Borrelli, et al., 2005). Although Voyager Passport claims to exceed research-based recommendations recently, the Federal What Works Clearinghouse revi ewed commercial reading programs (2007). The review found that none of the commercia l reading programs on the market had sufficiently rigorous studies to be included in the review by the Clearinghouse. Further, when Voyager Universal Literacy System was reviewed, results revealed that Voyager Universal Literacy System had a positive effect on alphabetics, and negative effects on reading comprehension. Implications National Level Implications from this study can be made at the n ational, state and local level. First, under No Child Left Behind Legi slation, the federal government has made a commitment that every child will be on grade level in reading and math by the year 2014.The purpose of Reading First is, to ensu re that all children in America learn to read well by the end of third-grade (Guidance for the Reading First program, 2002). At the national level reading programs are reviewed un der Reading First to determine if they are based on scientifically based reading research. The Reading First initiative provides states, districts, and schools with funding to impl ement instruction based on scientifically based reading research in grades K-3. Voyager Passport is one of the programs that Reading First promotes as being based on scientifically based reading research. Yet when the What Works Clearinghouse (2007) reviewed Voyager Universal Learning Systems the program was found to have potentially positive effects on alphabetic and pote ntially negative effects on
225 comprehension. If politicians are going to continue to make decisions about the remediation tools used by classroom teachers to ass ist struggling readers, there is a dire need for more independent research on these program s. Local Level. During the 2007-2008 school year the district expan ded the use of Voyager Passport as an intervention to all schools for use with fir st through fifth grade struggling readers. If the district continues to us e Voyager Passport as an intervention during the regular school year and during summer sc hool it would be more beneficial for the students to use two different curriculums. Voyager Passport now offers a summer school curriculum. Additionally, although Voyager Passport is one of 101 supplemental intervention reading programs that have been reviewed by the Flo rida Center for Reading Research (FCRR), and reviewed under Reading First, Voyager Passport lacks any norming data on their fidelity measure. Because there is no norming data on the fidelity measure, it is impossible to know if the fidelity measure actually measures what it was designed to assess. If the district continues to use Voyager Passport it will be important for the district to examine the fidelity of the implementat ion during summer school 2006 in regards to supplemental materials and class size du ring Voyager small group lessons. Additionally, it may be beneficial for the district to gather norming data on the fidelity measure. Recommendations for Future Research As much as this study answers some questions about third-grade struggling readers attitudes about reading, it also leads to new questions. Previous research shows that reading attitude plays an important role on bo th the level of ability attained by a
226 given student and through its influence on reading engagement and practice (McKenna et al., 1995; McKenna & Kear, 1990). Likewise a poor r eading attitude may contribute to aliteracy, a condition when fluent readers choose n ot to read when other options exist (McKenna et al., 1995). One recommendation concerns the need for further re search on the number of students who qualified but did not attend summer sc hool. Under Reading First summer school is optional and provided free of charge to r etained third-graders, yet of the 547 students who qualified to attend summer school, onl y 336 elected to attend summer school. An additional 50 students did not complete summer school. Thus, only 285 students actually completed summer school and took the alternative test the last day. Perhaps there is a need for research on ways to inc rease summer school attendance. This finding on the challenge of encouraging and su staining students participation in summer school supports previous fi ndings on summer school (Borman & Dowling, 2006). During the implementatio n of The Teach Baltimore program approximately 50 percent of the students as signed to the program did not attend the program consistently and with enough regularity to make a difference (Borman & Dowling, 2006). Findings from a meta-analysis (Coo per et al., 1996) on 93 studies of summer school achievement revealed that children wh o were middle class benefitted from summer school more than those who were disadva ntaged. This may be due to the fact that poor attendance in summer school may be a symptom of an uninvolved family. This offers furthe r support for the faucet theory, which has shown that the summer slide is particular ly harmful to students from low socio-economic status (Cooper et al., 1996; Heynes, 1978; Downey et al.; 2004). Not
227 only is there an achievement gap for students as a result of summer school, but this gap tends to be greater among the have-nots than amon g the haves. Although this research study supports these findings, it is also points to the need for further research that explores how to encourage and sustain students attendance in summer school. Research on struggling readers has shown that retai ned students often come from homes with that lack parental involvement. Because 73 percent of the students did not have permission to participate in the study, furthe r research is needed that explores why so many children lacked permission to participate i n the study. During the Pilot Study on the ERAS (McKenna & Kear 1990) with a group of third-grade retained students, the return rate for permission to participate in the study was much higher. A total of 26 students enrolled in two third grade retention classes were asked to participate in the study. During the actua l Pilot Study, 62 percent of the students had permission to participate in the study. Complet e data were collected from 58 percent of the participants. The fact that the researcher w orked at the school where the Pilot Study was conducted and knew many of the students m ay have contributed to the higher return rate. During the actual study if the parents of the summer school students had known the researcher they may have been more likely to allow their children to participate in the study. There clearly is a need for additional studies that explore the impact different instructional methods have on childrens attitudes about reading. These studies could offer support for alternatives to scripted literacy programs as ways to assist struggling readers. This is especially important when working with struggling third-grade readers since findings have shown that as children grow old er their attitude about reading
declines. This decline in the attitudes of children as they age supports the importance of early intervention and assisting struggling readers at an early age (McKenna et al., 1995). During this mixed the voice of the struggling reader. There is a ne ed for further research that captures the voice of the struggling reader (Langford & McDonag h, 2003; During this study comme about Voyager Passport but about the pressure they felt in regards to the FCAT and Stanford 10 Test. There is a need for future resear feelings and attitudes about study group cohesiveness appeared to play a big rol e in students responses during focus groups, there is an additional need for future rese arch on cohesiveness during focus groups with children. Previous studies have shown that focus groups with high levels of cohesiveness were more, cooperative, friendly, and praise worthy of each others accomplishments (Shaw & Shaw, 1962; Stewart & Sham dasani, 1990). However, much of this research is outdate Although politicians make decisions on whats best for struggling readers, it is the classroom teachers responsibility to implem ent the curriculum. In todays age of accountability the classroom teacher is responsible for taking to accelerating the reading growth of elementary sc hool struggling readers (& Allington & Walmsley 1995). As the researcher observed in summer school classro oms and reflected on her knowledge of whats best for struggling read 228 declines. This decline in the attitudes of children as they age supports the importance of early intervention and assisting struggling readers at an early age (McKenna et al., 1995). During this mixed -meth od study, focus groups helped the researcher to cap ture the voice of the struggling reader. There is a ne ed for further research that captures the voice of the struggling reader (Langford & McDonag h, 2003; Lincoln & Guba, 1985). comme nts were made during focus groups by the students not just but about the pressure they felt in regards to the FCAT and Stanford 10 Test. There is a need for future resear ch that explores young childrens feelings and attitudes about standardized testing. Additionally, since during th e present study group cohesiveness appeared to play a big rol e in students responses during focus groups, there is an additional need for future rese arch on cohesiveness during focus Previous studies have shown that focus groups with high levels of cohesiveness were more, cooperative, friendly, and praise worthy of each others accomplishments (Shaw & Shaw, 1962; Stewart & Sham dasani, 1990). However, much of this research is outdate d. Closing Thoughts Although politicians make decisions on whats best for struggling readers, it is the classroom teachers responsibility to implem ent the curriculum. In todays age of accountability the classroom teacher is responsible for taking the primary role in regards to accelerating the reading growth of elementary sc hool struggling readers (& Allington As the researcher observed in summer school classro oms and reflected on her knowledge of whats best for struggling read ers, it seemed that the importance of the declines. This decline in the attitudes of children as they age supports the importance of early intervention and assisting struggling readers at an early age (McKenna et al., 1995). od study, focus groups helped the researcher to cap ture the voice of the struggling reader. There is a ne ed for further research that captures the & Guba, 1985). by the students not just but about the pressure they felt in regards to the FCAT and ch that explores young childrens standardized testing. Additionally, since during th e present study group cohesiveness appeared to play a big rol e in students responses during focus groups, there is an additional need for future rese arch on cohesiveness during focus Previous studies have shown that focus groups with high levels of cohesiveness were more, cooperative, friendly, and praise worthy of each others accomplishments (Shaw & Shaw, 1962; Stewart & Sham dasani, 1990). However, much Although politicians make decisions on whats best for struggling readers, it is the classroom teachers responsibility to implem ent the curriculum. In todays age of the primary role in regards to accelerating the reading growth of elementary sc hool struggling readers (& Allington As the researcher observed in summer school classro oms and reflected on her ers, it seemed that the importance of the
229 classroom teacher should have been considered when examining the attitudes of struggling readers. There was a difference not just in the appearance of the fidelity of the implementation of Voyager Passport in the different classrooms, but in the classroom environment. Overall, following summer school using Voyager Passport the children knew a lot about Voyager Passport and they felt they were ready to take the alternat ive test. Based on comments made during Focus Groups, i t appeared that many of the childrens confidence in their reading ability had improved. Many of the children gave the credit for being ready for the test to their te achers. Using scripted literacy goes against what reading t heorists and reading researchers have proven to be successful (Clay, 1991; Allington 1998, 2002; Slavin & Madden, 2001; Dolan, & Wasik, 1996; Duffy, 2001 ) Programs such as Reading Recovery and Success for All have been proven to show growth wit h struggling readers. In New Zealand 99 percent of the children in ten education al districts where Reading Recovery was fully implemented were reading at or above grad e level. However, Reading Recovery requires time, money and commitment. Succe ss for All includes both tutoring and family support services in its comprehensive sc hool restructuring program (Slavin & Madden, 2001). Duffy (2001) found success using a b alanced approach to literacy instruction in summer school. She promotes using a balanced approach to literacy instruction as opposed to a program-driven approach : Rather than purchasing fixed, commercial reading p rograms and training teachers to use these programs, perhaps a better in vestment of school districts time and resources would help teachers understand h ow principles of balance,
230 acceleration, and responsive teaching can be utiliz ed in multiple, purposeful ways in classrooms with struggling readers (p. 92). Under Reading First school districts are being placed in awkward situa tions. The Reading First initiative provides states, districts, and schools with funding to implement scientifically based reading instruction (SBRR) in kindergarten through third grade. To meet the needs of struggling readers under Reading First many schools have adopted supplemental programs such as Voyager Passport However, the approved materials are not necessarily based on reading research which has led to a great deal of controversy surrounding Reading First As a third-grade teacher, I know first-hand what it s like to tell a child they must repeat the third-grade. As a third-grade teacher, I have a responsibility to teach each of my children. As a third-grade teacher I know what i ts like to be forced to use scripted literacy as a remediation tool with struggling read ers. As a third-grade teacher who has just spent the last year researching scripted liter acy and conducting this study, I must admit I have biases towards the use of scripted lit eracy as an intervention tool with struggling readers. However, when curriculum decisi ons are dictated by politicians, teachers and districts are caught in the middle. I am caught in the middle. Although I am a Ph.D. candidate completing her dissertation in cu rriculum and instruction with a focus on reading, as a third-grade teacher I am being man dated to use Voyager Passport with my third-grade struggling readers four times a week in addition to the 90 minute reading block. The use of Voyager Passport in summer school did not lower struggling readers attitudes about reading. However, their attitudes d id not increase either. As an advocate
231 for struggling readers, struggling readers deserve instruction that is individualized not scripted. As an advocate for struggling readers, st ruggling readers deserve to be talked to authentically, not read a script. As an advocate fo r struggling readers, struggling readers have to right to responsive teaching. As an advocat e for teachers, teachers deserve to be trusted not trapped in scripted literacy. As a thir d grade teacher I do believe that educators have the will and the ability to teach al l children, including struggling readers to read. If I dont believe in educators who will? In 2001, President Bush enacted the No Child Left B ehind Act. At that time he called the legislation, The cornerstone of my admi nistration (retrieved 12-06-2006 from www.ed.gov ). Now, as the legislation approaches reauthorizat ion, it appears that in its quest to leave no child behind, Bushs corners tone has backed states and school districts into corners regarding how best to help s truggling readers. Ultimately, in the state of Florida school district s are penalized for failing report cards. However, it is the children who make pay the ultimate price. A total of three metaanalyses containing 700 analyses of achievement, fr om more than 80 studies published from 1925 to 2006 fail to support the use of retent ion as an early intervention to enhance academic achievement (Jimerson & Kaufman, 2003; Jim erson, 2001; Holmes, 1989; Silberglitt et al., 2006). Although the goal of No Child Left Behind is for every child to read on grade level by the end of third-grade, coun tless students continue to be left behind as retained students. During this small stud y in a small school district on the west coast of Florida, of the 547 students who qualifie d for summer school, 336 third-grade students attended summer school. Of the 336 student s who attended summer school on the first day, 285 students completed summer school
232 Of the 285 students who completed summer school, 83 students (29%) passed the alternative test at the end of summer school. The r emaining students were left behind.
233 List of References Adler, P.A. & Adler, P. (1994). Observational techn iques. In N.K. Denzin & Y.S. Lincoln (Eds.), Handbook of Qualitative Research (pp. 337-392). Thousand Oaks, CA: Sage. Ajzen, I. (1989). Attitude structure and behavior. In A. R. Pratkanis, S.J. Breckler, & A. G. Greenwald (Eds.), Attitude structure and function (pp. 241-274). Hillsdale, NJ: Erlbaum. Alexander, J. E. & Filler, R. C. (1976). Attitudes and reading Newark, DE: International Reading Association. Allington, R. L. (1998). More on the schools we hav e versus the schools we need. In C. Weaver (Ed.), Practicing what we know: Informed reading instructi on (pp. 554565). Urbana: National Council Teachers of English. Allington, R. L. & Walmsley, S. A. (Eds). (1995). No quick fix rethinking literacy programs in americas elementary schools. Teachers College Press: New York. Allington, R. L. (2002). Troubling Times: A short h istorical perspective. In R. L. Allington (Ed.), Big Brother and the national reading curriculum: Ho w ideology trumped evidence. (pp. 3-46). Portsmouth: Heinemann. Anderson, M. A., Tollefson, N. A. & Gilbert, E. C. (1985). Giftedness and reading: A cross-sectional view of differences in reading atti tudes and behaviors. Gifted Child Quarterly, 29, 186-189.
234 Anderson, R. C., Hiebert, E. H., Scott, J. A., & Wi lkinson, I. A. (1985). Becoming a nation of readers: The report of the Commission on Reading. Washington DC: National Institute of Education. Ary, D., Jacobs, L. C., & Razavieh, A. (1996). Introduction to research in education (5th ed.). Fort Worth, TX: Harcourt Brace College Publishers. Askov, E. & Fishback, T. (1973). An investigation o f primary pupils attitudes toward reading. Journal of Experimental Education, 41 1-7. Austin, G. R., Roger, B. G., & Walbesser, H. H. (19 72). The effectiveness of summer compensatory education: A review of the research. Review of Educational Research, 42, 171-181. Barbour, R. S. & Kitzinger, J. (1999). (Ed.), Developing focus group research: Politics, theory and practice. London: Sage. Beck, R. C. (1983). Motivation: Theories and Principles Englewood Cliffs, NJ: Prentice Hall. Beck, E. (2005). Elements of Vocabulary Steck-Vaughn. Billson J.M. (1994). (Ed.), Conducting Focus Groups: A manual for sociologists on the use of focus groups as a tool in social and market research American Sociological Association. Bogdan, R. C., & Biklen, S. K., (1998). Qualitative Research for Education. Boston: Allyn and Bacon. Borman, G. D. (2000). The Effects of summer school, questions answered, questions raised. Monographs of the Society for Research in Child Dev elopment, 65 (1, Serial No. 260).
235 Borman, G. D. & Boulay, M. (2004). Summer learning: Research, policies and programs Mahwah, NJ: Lawrence Erlbaum Associates. Borman, G. D. & Dowling, N. M. (2006). Longitudinal achievement effects of multiyear summer school. Evidence from the teach baltimore ra ndomized field trial. Educational Evaluation and Policy Analysis, 28, 1-22. Borrelli, B., Sepinwall, D., Ernst, D., Bellg, A.J. Czajkowski, S., Breger, R., Defrancesco, C., Levesque, C., Sharp, D.L., Ogedeg be, G., Resnick, B., Orwig, D. (2005). A new tool to assess treatment fidelity and evaluation of treatment fidelity across 10 years of health behavior researc h. Journal of Consulting and Clinical Psychology, 73 852-860. Carey, M. A. (1995). Comment: concerns in the analy sis of focus group data. Qualitative Health Research, 5 (4), p. 487-495. Carter, L. F. (1984). The sustaining effects of co mpensatory and elementary education. Educational Researcher, 13 (7), 4-13. Charmaz, K. (2000). Grounded Theory: Objectivist an d constructivist methods. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp.509-535). Thousand Oaks, CA: Sage. Clay, M. (1991). Becoming Literate: The Construction of Inner Control Portsmouth NH: Heinemann. Cohen, J. (1965). Some statistical issues in psychological research In B.B. Wolman (Ed.), Handbook of clinical psychology (pp. 95-121). NY: McGraw-Hill.
236 Cohen., L., & Manion, L. (1986). Research methods in education London: Croom Helm. Constas, M.A. (1992). Qualitative analysis as a pub lic event: The documentation of category development procedures. American Educational Research Journal 29, 253-266. Cooper, H., Nye, B., Charlton, K., Lindsay, J., & G reathouse, S. (1996). The effects of summer vacation on achievement test scores: a narr ative and meta-analytic review. Review of Educational Research, 66 (3), 227-268. Cooper, H., Charton, K., Valentine, J. & Muhlenbru ck, L. (2000). Making the most if summer school: A meta-anaytic and narrative review Monographs of the Society for Research in Child Development, 65 ,(260). Cronbach L. J, (1951). Coefficient alpha and the in ternal structure of tests. Psychometrika, 16, 297-334. Cunningham, J.W. (2001). The National Reading Panel Report. Reading Research Quarterly, 36 326-335. Darbyshire, P., MacDougall, C., & Schiller, W. (200 5). Multiple methods in qualitative research with children: More insight or just more? Qualitative Research, 5 (4), 417-436. Darling-Hammond, L. (1998). Alternatives to retent ion. The School Administrator August, 18-21. DIBELS (2006). Available from the DIBELS Web site, www.dibels.uoregon.edu
237 Downey, D. B., von Hippel, P. T., & Broh, B. A. (20 04) Are schools the great equalizer? Cognitive inequality during the summer months and t he school year. American Sociological Review, 69, 613-635. Duffy, A. (2001). Balance, literacy acceleration, a nd responsive teaching in a summer school literacy program for elementary school stru ggling readers. Reading Research and Instruction, 40 67-100. Dumas, J., Lynch, A., Laughlin, J., Smith, E., & Pr inz, R. (2001). Promoting intervention fidelity: Conceptual issues, methods and prelimina ry results from the EARLY ALLIANCE prevention trial. American Journal of Preventive Medicine, 20 (1S), 38-47. Durost, Bixler, Wrightstone, Prescott, & Balow (197 0). Measurement Educational and The Metropolitan Achievement Test. New York: Harco urt Brace Jovanovich. Edmunds, H. (1999). The Focus Group Research Handbook. NTC Business Books: Chicago. Engelmann, S., Haddox, P., & Bruner, E. (1984). Teach Your Child to Read in 100 Easy Lessons. Simon & Schuster: New York. Entwisle, D. R., Alexander, K. L., & Olson, L. S. ( 1997). Children, schools and inequality Boulder, Co: Westview Press. Ferguson, P., Jimerson, S. R., & Dalton, M.J. (2001 ). Sorting out successful failures: Exploratory analysis of factors associated with ac ademic and behavioral outcomes of retained students. Psychology in the Schools, Vol. 38 (4), 327-341.
238 Fishbein, M., & Ajzen, L. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison Wesley. Florida Center for Reading Research, (2006). Availa ble from FCRR Web site, http://www.fcrr.org. Frechtling, J. A., Zhang, X., & Siverstein, G. (200 6). The Voyager Universal Literacy System: Results from a study of kindergarten stude nts in inner city schools. Journal of Education for Students Placed at Risk, 1 1 (1), 75-95. Goodman, K. (Ed.). (2006). Examining Dibels: What it is and what it does. Brandon, Vermont: Vermont Society for the Study of Education, Inc. Greene, Caracelli, & Graham (1989). Toward a concep tual framework for mixed method evaluation designs. Educational Evaluation and Policy Analysis, 11 (3), 225-274. Gresham, F.M., MacMillan, D.L., Beebe-Frankenberger M.E., & Bocian, K.M. (2000). Treatment integrity in learning disabilities inter vention research: Do we really know how treatments are implemented? Learning Disabilities Research & Practice, 15 (4), 198-205. Grunwald, M. (2006, October 1). Billions for an in side game on reading, Washington Post Retrieved January 9, 2007, from http://www.washingtonpost.com Guidance for the Reading First Program, (April 2002 ). Retrieved January 9, 2007, from U.S. Department of Education Office of Secondary E ducation via Access: http://www.ed.gov/programs/readingfirst/guidance.pd f Guilford, J. P. & Fruchter, B. (1978). Fundamental statistics in psychology and education. New York: McGraw Hill.
239 Guthrie, J. T. & Wigfield, A. (2000). Engagement a nd motivation in reading. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson, & R. Barr (E ds.), Handbook of Reading Research: Volume III (pp. 403-422). New York: Erlbaum. Heathington, B. S. (1975). The Development of Scales to Measure Attitudes Towa rd Reading. Unpublished doctoral dissertation, University of Te nnessee. Heathington, B. S. & Alexander, J. E. (1984). Do cl assroom teachers emphasize attitudes toward reading? The Reading Teacher, 37, (6), 484-488. Hect, S.A. & Torgesen, J.K. (2002) Voyager Universal Literacy System Program: Evaluation in Orlando, Florida. Retrieved December 2, 2006, from http://www.fcrr.org Heyns, B. (1978). Summer learning and the effects of schooling New York: Academic Press. Heyns, B. (1987). Schooling and cognitive developme nt: Is there a season for learning? Child Development, 58, 1151-1160. Holmes, C. T. (1983). The fourth r: Retention. Journal of Research and Development in Education, 17, 1-6. Holmes, C. T. (1989). Grade level retention effects : a meta analysis of research studies. In L. A. Shepard & M. L. Smith (Eds.), Flunking grades: Research and policies on retention (pp. 16-33). London: Falmer. Institute for School Innovation Retrieved December 2, 2006 from www.ifsi.org Ishikawa, K. (1985). Developmental study of school childrens attitudes toward reading. The Science of Reading, 29 89-98.
240 Jackson G. (1975). The research evidence on the eff ects of grade retention. Review of Educational Research, 45, 613-635. Janis, I. L. (1965). The problem of validating cont ent analysis. In H.D. Lasswell & N. Leites (Eds.), Language of Politics : Studies in Quantitative Semantics. (pp. 424439). Cambridge: MIT Press. Jimerson, S. (2001). Meta-analysis of grade retenti on research: implications for practice in the 21st century. School Psychology Review, 30 3-25. Jimerson, S. & Kaufman, A. (2003). Reading, writing and retention: A primer on grade retention research. The Reading Teacher, 56 622-633. Jimerson, S.R., Carlson, E., Rotert, M., Egeland, B ., & Sroufe, L.A. (1997). A prospective, longitudinal study of the correlates and consequences of early grade retention Journal of School Psychology, 35, 3-25. Johnson, B. & Christensen, L. (2004). Educational Research Quantitative, Qualitative and Mixed Approaches Boston: Pearson. Kidd, P. S., & Parshall, M. B. (2000). Getting the focus and the group: enhancing analytical rigor in focus group research. Qualitative Health Research, 10 (3), p. 293-308. Krashen, S. (2001). More smoke and mirrors: A criti que of the National Reading Panel Report on Fluency. Phi Delta Kappan, 83 119-124. Krashen, S. (2005). Is school free reading good for children? Why the National Reading Panel Report is (still) wrong. Phi Delta Kappan 86 32-35. Krippendorf, K. (1980). Content Analysis: an introduction to its methodolog y. Beverly Hills, CA: Sage.
241 Kuder, S. J. (1990) Language abilities and progress in a direct instruction reading program for students with learning disabilities. Journal of Learning Disabilities, 14, 124-129. Kuder, S. J. (2001). Effectiveness of the DISTAR re ading program for children with learning disabilities. Journal of Learning Disabilities 23 69-71. Kulynych, J. (2001). No playing in the public spher e: Democratic theory and the exclusion of children. Social Theory and Practice, 27 (2), 231-264. Kush, J. C. & Watkins, M.W. (1996). Long term stabi lity of childrens attitudes toward reading. Journal of Educational Research, 89, 315-325. Langford, J. & McDonagh, D. (Ed) 2003, Focus groups: Supporting effective product development. New York: Taylor & Francis. Lincoln, Y. S. & Guba, E. G. (1985). Naturalistic inquiry Beverly Hills, CA: Sage. Liska, A. E. (1984). A critical examination of the causal structure of the fishbein/ajzen attitude-behavior model. Social Psychology Quarterly, 47, (1), 61-74. Mann, D. (1987). Can we help dropouts? Thinking abo ut the undoable. In: G. Natriello, (Ed.), School Dropouts: Patterns and Policies (pp. 3-19). New York: Teachers College Press. Manzo, K. (2006, October 4). Scathing report casts cloud over reading first Education Week Retrieved February 16, 2007, from http://www.edweek.org Martin, C. E. (1984). Why some gifted children do n ot like to read. Roeper Review, 7 (2) 72-75.
242 Mathewson, G. C. (1994). Model of attitude influenc e upon reading and learning to read. In R. B. Ruddell, M. R. Ruddell, & H. Singer (Eds. ), Theoretical models and processes of reading 4th Ed. (pp.1131-1161). Newark, DE: International Reading Association. McAfee, J. K. (1981, April). Toward a theory of promotion: Does retaining studen ts really work? Paper presented at the annual meeting of the Ameri can Educational Research Association, Los Angeles, CA. McKenna, M.C. (1994). Toward a model of reading att itude acquisition. In E.H. Cramer & M. Castle (Eds.) Fostering the Lifelong Love of Reading: The Affecti ve Domain in Reading Education (pp. 18-40). Newark, DE: International Reading Association. McKenna, M. C. & Kear, D .J. (1990). Measuring atti tude toward reading a new tool for teachers. The Reading Teacher 43 626-639. McKenna, M. C., Kear D. J. & Ellsworth, R. A. (1995 ). Childrens attitudes toward reading: A national survey. Reading Research Quarterly, 30 934-955. McKenna, M. C., Stratton, B.D., Grindler, M. C. & J enkins, S. (1995). Differential effects of whole language and traditional instructi on on reading attitudes. Journal of Reading Behavior, 27 19-44. McGuire, W .J. (1989). The structure of individual attitudes and attitude systems. In A R. Pratkanis, S. J. Breckler, & A.G. Greenwald (Eds.). Attitude structure and function (pp. 37-69). Hillsdale, NJ: Erlbaum. Meyer, R. (2002 ) Phonics exposed: Understanding and resisting syst ematic direct intense phonics instruction Mahwah, NJ: Lawerence Erlbaum.
243 Meyer, L. A., Gersten, R. M., & Gutkin, J., (1983). Direct instruction: A project follow through success story in an inner-city school. The Elementary School Journal, 84 241-252. Mihalic, S. (2002, April). The importance of implementation fidelity, Boulder, Colorado: Center for the Study and Prevention of Violence. Miles, M. B. & Huberman, A.M. (1994). Qualitative Data Analysis, Second Edition Thousand Oaks, CA: Sage. Morgan, D. L. (1988). Focus Groups as Qualitative Research. Newburry Park: CA. Morgan, D. L. & Kruegger, R.A. (1997). Focus Group Kit, Vols 1-6 London: Sage. Morrow, L. M. (1983). Home and school correlates of early interest in literature. Journal of Educational Research, 76, 221-230. Moustafa, M. & Land, R. (2005). Scripted reading in struction: Help or hindrance? In B. Altwerger (Ed.) Reading for profit: How the bottom line leaves kids behind (pp. 63-77). Portsmouth, NH: Heinemann. Mowbray, C. T., Holter, M. C., Teague, G.B. & Bybee D. (2003). Fidelity criteria: development, measurement, and validation. American Journal of Evaluation, 24 315-340. National Reading Panel (2000). Report of the National Reading Panel: Teaching chil dren to read: reports of the subgroups. Washington DC: National Institute of Child Health and Human Development. No Child Left Behind Act of 2001, Pub. L. No. 107-1 10, 1201, 115 Stat. 1535. (2002). Access: http://edworkforce.house.gov/issues/107th/education /nclb/nclb.htm
244 OPPAGA, The Florida Office of Program Policy, Analy sis, and Government Accountability (2006) Access: http://www.oppaga.state.fl.us Onwuegbuzie A. J. (2002, November). Validity and qualitative research: An oxymoran? Paper presented at the Annual Conference of the Mi d-South Educational Research Association, Chattanooga, Tennesee. Onwuegbuzie, A. J. (2003). Expanding the framework of internal and external validity in quantitative research. Research in the Schools, 10(1) 71-90. Onwuegbuzie, A. J. & Collins, K.M.T. (2004 ). A framework for developing sampling designs in mixed method research. Paper presented at the annual meeting of the Australian Association for Research in Education, Melbourne, Australia. Parker, D. R. (2001). Social promotion or retention ? Two wrongs still dont make a right. Leadership, 30, 12-16. Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, IA: William C. Brown. Peterson, S. E. (1987). A longitudinal study of the effects of retention/promotion on academic achievement. American Educational Research Journal, 24 107-118. Rafoth, M. A. (2002). Best practices in preventing academic failure and promoting alternatives to retention. In A. Thomas and J. Grim es, editors, Best Practices in School Psychology IV, (pp. 789-801). The National Association of School Psychologists. Bethesda: Maryland, Reading First (2006). Access: http://www.readingfirstsupport
245 Roberts, G. (2002). Evaluation report on the impact of the voyager univ ersal literacy system in Birmingham schools Retrieved December 2, 2006, from http://www.fcrr.org Roberts, G. & Allen, A.S. (2003). Impact of the Voy ager Universal Literacy System as Measured by PALS in Virginia. Ross, E.P. & Fletcher, R. K. (1989). Responses to childrens literature by environment, grade level, and sex. Reading Instruction Journal, 32 (2), 22-28. Rumberger, R. W. (1987). High school dropouts: A re view of issues and evidence. Review of Educational Research 101-121. Sainsbury, M. (2004). Childrens attitudes to readi ng Educational Review, 79 (15). Sainsbury, M. & Schagen, I. (2004). Attitudes to re ading at ages nine and eleven. Journal of Research in Reading 27 (4), 373-386. Sexton, C.W. Effectiveness of the DISTAR reading pr ogram in developing first graders language skills. Journal of Educational Research 82, 291-293. Shapiro, S.S., & Wilk, M.B. (1965). An analysis of variance test for normality. Biometrika 52, (3), 591-599. Shaw, M. E., & Shaw, L. M. (1962 ). Some effects of sociometric grouping upon learning in a second grade classroom. Journal of Social Psychology 57 453-458. Shepard, L. S., & Smith, M. L. (1990). Synthesis of research on grade retention. Educational Leadership, 47 84-88.
246 Silberglitt, B., Appleton, J.J., Burns, M.K., & Jim erson, S. (2006). Examining the effects of grade retention on student reading performance: A longitudinal study. Journal of School Psychology, 44 255-270. Slavin, R., & Madden, N. (2001). One Million Children: Success for All. Thousand Oaks, CA: Corwin Press. Stanford 10 Achievement Test, 10th Edition (2003). Harcourt Publishing. Stewart, D.W. & Shamdasani, P.N. (1990). Focus Groups Theory and Practice. Newbury Park: Sage Publications. Strauss, A. & Corbin, J. (1990). Grounded theory re search: procedures, canons, and evaluative criteria, Qualitative Sociology, 13 (1), 3-21. Stemler, S. (2001, June 7). An overview of content analysis. Practical Assessment Research & Evaluation, 7 (17). Retrieved November 9, 2007 from http://PAREonline.net/getvn.asp?v=7&n=17 Stone, S., Engel, M., Nagaoka, J., & Roderick, M. ( 2005). Getting it right the second time around: Student classroom experience in Chicagos summer bridge program. Teachers College Record, 107 (5), p. 935-957. Swanson. B. (1982), The relationship between attitu de toward reading and reading achievement. Educational and Psychological Measurement, 42, 1303-1304. Tashakkori, A. & Teddlie, C. (Eds.) (2003). Handbook of Mixed Methods in Social and Behavioral Research Thousand Oaks, CA: Sage. Taylor, B., Short, R., Shearer, B., & Frye, B. (199 5) First Grade Teachers Provide Early Reading Intervention in the Classroom. In R. L. All ington & S. A. Wallmsleys (Eds.) No Quick Fix. pp. 159-176. Teachers College Press: New York.
247 Tjora, A. H. (2006). Writing small discoveries: an exploration of fresh observers observations. Qualitative Research, 6 (4), 429-451. Trophies (2004). Harcourt School Publishers. Twist, L., Gnaldi, M., & Morrison, J. (2004). Good readers but at a cost? Attitudes to reading in England. Journal of Research in Reading 27 387-400. U.S. Department of Education Office of Inspector Ge neral (2006). The Reading First Programs Grant Application Process, Final Inspecti on Report (ED-OIG/113F0017). Washington, D.C. Voyager Expanded Learning (2006). Voyager Principals Handbook (2006). Walberg, H.J. & Tsai, S. (1985). Correlates of read ing achievement and attitude: A national assessment study. Journal of Educational Research, 78, (3), 159-167. Wallbrown, F.H., Brown, D.H., & Engin, A.W. (1978). A factor analysis of reading attitudes along with measures of reading achievemen t and scholastic aptitude. Psychology in the Schools, 15 160-165.
249 Appendix A: The Elementary Reading Attitude Survey (ERAS)
250 Appendix A: (Continued)
251 Appendix A: (Continued)
252 Appendix A: (Continued)
253 Appendix A: (Continued)
254 Appendix A: (Continued)
255 Appendix B: Informed Consent Form
256 Appendix B: (Continued)
257 Appendix C: Sample Voyager Lesson A sample third grade lesson began with the children gathered around a table. They each had a soft back reading work-book in front of them. The teacher opened the lesson by specifically teaching a prefix or suffix. Each o f the vocabulary words from that lesson had a prefix or suffix in it. After a mini lesson, the teacher introduced 6 more vocabulary words that were in the passage the group was about to read. The teacher then introduced the story and read aloud the passage. Next the teac her led the group as they chorally read the paragraph. Following the choral read the teacher led the stude nts through a series of questions and answers about the paragraph. Next, th e students read the paragraph again chorally. Then the teacher asked the students a ser ies of comprehension questions. Then Depending the group moved on to the next activity w hich was a vocabulary center. The teacher then called a new group to her table and be gan to work with the group on the same lesson. Later that afternoon, the teacher taught the second part of the lesson which was on fluency. Each child had their own fluency workbo ok, which they worked in during the lesson. During the fluency lesson, the teacher intr oduced a reading passage and lead the group in a choral. Then, the teacher asked the stud ents to identify words in the text that were unfamiliar. The teacher then reviewed the word s and asked one child to do an oral retelling of what the story was about.
258 Appendix C: Continued Next, she asked another child to build on that rete lling. Then the students partner read the passage. Finally, the lesson ended with a timed read where the students re-read the passage in a 1-minute timed reading. Once again the lesson was taught small group. This time the other students were at the computer o r doing independent reading. When the teacher had finished the lesson she switched gr oups.
259 Appendix D: Voyager Passport Training Agenda
260 Appendix E: ERAS Teacher Training Protocol Elementary Reading Attitude Survey Group Protocol T eacher Training Agenda for Teacher Training: 1. First provide a brief overview of the study. i. Explain that the study is measuring childrens atti tudes about reading. ii. Ask if anyone is familiar with The Elementary Readi ng Attitude Survey (ERAS). iii. Explain what the survey measures: The Elementary R eading Attitude Survey provides a quick indication of stud ents attitudes toward reading. It consists of 20 items and can be administered to an entire class in about 10 minutes. Each item pres ents a brief, simply worded statement about reading, followed by four pictures of Garfield. Each pose is designed to depict a diff erent emotional state, ranging from very positive to very negative 2. Now explain to teachers how to administer the asses sment. i. Begin by telling students that you wish to find out how they feel about reading. Emphasize that this is not a test an d there are no right answers. Encourage sincerity. ii. Distribute the surveys and ask the students to writ e their names in the space at the top.
261 Appendix E: (Continued) iii. Hold up a copy so that the students can see the fir st page. Point to the picture of Garfield at the far left of the firs t item. iv. Ask the students to look at this same picture on th eir own survey form. Discuss with them the mood Garfield seems to be in. (very happy). Then move to the next picture ad again disc uss Garfields mood. In the same way move to the third and fourth pictures and talk about Garfields moods-a little upset and very upset. It is helpful to point out the position of Garfields mou th, especially in the middle two figures. v. Explain to the students that together you will read some statements about reading and that the students should think ab out how they feel about each statement. vi. Instruct the students that then they will circle th e picture of Garfield that is closest to their own feelings. vii. Emphasize to the students that they should respond according to their own feelings not as Garfield might respond! viii. Read each item slowly and distinctly. ix. Read each item a second time while students are thi nking. x. Remind teachers to make sure and read the item numb er and to remind them of page numbers when new pages are reac hed.
262 Appendix E: (Continued) 3. Explain to the teachers that I will be scoring the surveys. Ask them if they will gather the surveys up once complete and send t hem to the office to put them in a large envelope which has the research ers name on the it.
263 Appendix F: ERAS Group Protocol Directions for use Begin by telling the students that you wish to find out how they feel about reading. Emphasize that this is not a test an d there are no right answers. Encourage sincerity. Distribute the surveys and ask the students to writ e their names in the space at the top. Hold up a copy so that the students can see the fir st page. Point to the picture of Garfield at the far left of the first it em. Ask the students to look at this same picture on th eir own survey form. Discuss with them the mood Garfield seems to be in. (very happy). Then move to the next picture ad again discuss Garfield s mood (a little happy). In the same way move to the third and fourt h pictures and talk about Garfields moods (a little upset and very ups et). It is helpful to point out the position of Garfields mouth, especia lly in the middle two figures. Explain to the students that together you will read some statements about reading and that the students should think about ho w they feel about each statement. Instruct the students that then they will circle th e picture of Garfield that is closest to their own feelings. Emphasize to the students that they should respond according to their own feelings not as Garfield might respond! Read each item slowly and distinctly. Read each item a second time while students are thi nking. Remind teachers to make sure and read the item numb er and to remind them of page numbers when new pages are reached. When done turn into the front office for me to coll ect.
264 Appendix G: Informed Consent Pilot Study
265 Appendix G: Continued
266 Appendix H: ERAS Pilot Study Group Protocol Group Protocol for teachers to administer the Eleme ntary Reading Attitude Survey Directions for use Begin by telling the students that you wish to find out how they feel about reading. Emphasize that this is not a test an d there are no right answers. Encourage sincerity. Distribute the surveys and ask the students to writ e their names in the space at the top. Hold up a copy so that the students can see the fir st page. Point to the picture of Garfield at the far left of the first it em. Ask the students to look at this same picture on th eir own survey form. Discuss with them the mood Garfield seems to be in. (very happy). Then move to the next picture ad again discuss Garfield s mood (a little happy). In the same way move to the third and fourt h pictures and talk about Garfields moods (a little upset and very ups et). It is helpful to point out the position of Garfields mouth, especia lly in the middle two figures. Explain to the students that together you will read some statements about reading and that the students should think about ho w they feel about each statement. Instruct the students that then they will circle th e picture of Garfield that is closest to their own feelings. Emphasize to the students that they should respond according to their own feelings not as Garfield might respond! Read each item slowly and distinctly. Read each item a second time while students are thi nking.
267 Appendix I: ERAS Pilot Study Scores Pilot Study ERAS Scores ___________________________________________________ _____________________ Pretest Posttest ID # Rec. Ac. Full Scale Rec. Ac ad. Full Scale ___________________________________________________ _____________________ 1 33 43 73 27 34 61 2 33 35 68 19 27 46 3 31 30 61 23 32 55 4 28 30 58 28 36 64 5 32 37 69 33 37 70 6 35 37 72 30 35 65 7 31 27 58 25 21 46 8 31 31 62 30 32 62 9 28 31 59 29 27 56 10 32 35 67 16 16 32 11 35 32 67 30 28 58 12 36 37 73 33 37 70 13 31 34 65 37 34 71 14 33 34 67 29 34 63 15 36 36 72 34 33 67
268 Appendix J: Coding Training Protocol Protocol for Coding Training Go over the rules for assigning units to categories (Krippendork, 1980) 1. Review where data came (The transcripts of the focu s groups conducted during summer school). 2. Review specific characteristics of the coders: a. The primary coder was the researcher who was a grad uate student trained in qualitative methods. b. The secondary coder was a fellow graduate students also trained in qualitative research. The secondary coder reviewed 20% of the transcripts to verify the accuracy. i. The 20% was made up of one question at four of the five sites. Two questions were verified from the fifth site. ii. Differences were discussed and an agreement was mad e as to how to code the data in question. Teach the secondary coder how to code the data. 1. Use a sample question that will not be scored by th e secondary coder. 2. Review the Use cut and paste technique to identify section of the research that was relevant to the question. 3. Take the Question and review sample answers. 4. Now explain to secondary coder how the research ass igned the codes or chunks 5. Explain to researcher that she will be working from a typology that the researcher has already established. 6. Review Sample codes. 7. Give the secondary researcher the transcripts from 20% of the questions. At this time also give her the typology the researcher had already established in regards to possible codes fo r each answer.
269 Appendix J: (Continued) 8. Discuss what will be done if there is a discrepancy amongst the two coders. a. In the event of a discrepancy the two coders will d iscuss the specific response and come to agreement as to how t o code it. b. In the event that a consensus cannot be agreed on, both answers will be accepted.
270 Appendix K: Focus Group Protocol Protocol for Focus Groups Agenda 1. Introductions: Introduce the children to one anothe r. Go around the circle and state your name and your home school. 2. Introduce self (moderator) 3. Breaking The Ice: Talk with children about pets. (H ave you ever had a pet? Does anyone have a dog?) 4. Explain the purpose of the session to talk about re ading 5. Explain the moderators role (to run the session) a nd the observers role (to observe and tape record). 6. Review ground rules 7. Ask questions (be sure to use clarification and pro bing routines as necessary). Focus Group Ground Rules Remind the students to speak clearly one at a time: It will be important for each of you to speak clearly and one at a time. No right or wrong answers Need for active participation Clarification and Probing Routines Because children will agree with others to avoid st anding out, it will be important to frequently ask if anyone has other ideas or diff erent opinions.
271 Appendix K: Continued Watch for gestures and facial expressions that may reveal something about the accuracy of a comment, or suggest that someone has a strong feeling about a question being asked. If a talker takes over the conversation, thank them for sharing and call on another student. Begin with voluntary responses, if some children ar e not participating, call on them. Questions 1. Tell me about summer school. Which part of summer s chool do you like best? 2. Do you like reading? Why or why not? 3. What types of books do you like to read the most?
272 Appendix L: Attitude Scale for Classroom Observatio n Attitude Scale Used for Classroom Observations Gath ered from Field Notes ___________________________________________________ _____________ Enthusiastic Engaged Indifferent Off Task Withdraw n ___________________________________________________ _____________ Enthusistic Sitting on edge of seat Comments such as, Please pick me! Smiling Raising hand as high as possible On task Engaged Raising Hand normally On task Indifferent Going through the motions No verbal comments No expression
273 Appendix L: (Continued) Off Task Playing with items in desk Turned around during lesson Frequent trips to bathroom or water fountain Fake reading Rolling around on ground during independent reading Withdrawn Head down Jacket over Head No comments No expression
274 Appendix M: Example of Transcript Coding Example of Transcript Coding 1. Use cut and paste technique to identify section of the research that was relevant to the question. 2. Question 1: Tell me about summer school. What part of summer school do you like best? Sample answers included: a. I like recess. b. I like reading independently. c. The timed reading. You know you can beat your time . d. You can learn more. 3. After reading all of the transcribed responses the following codes or chunks emerged. Sample codes included: a. Timed reading b. Independent reading c. Read aloud d. Pass test e. Learning f. Recess g. Playing with friends h. Meeting new people i. Games j. Teachers 4. Next look at the series of words and phrases to det ermine which ones could be clustered together into a category. Sample categori es included: a. Social (recess, playing with friends, meeting new p eople, games) b. Academic (timed reading, independent reading, read loud, teachers, pass test) c. Learning (You can learn more, learning, learn how t o read) 5. Now use attribution analysis (Janis, 1965) to exami ne the frequency of the categories. a. Social: 4 b. Academic: 14 c. Learning: 2 d. No Response: 1
275 Appendix N: Permission Letter from School District
276 Appendix O: Request Letter for Study to School Dist rict
277 Appendix P: Summer School Day Structure
278 Appendix Q: Voyager Fidelity Measure
279 Appendix Q: (Continued)
280 Appendix R: Protocol for Gathering and Analyzing F ield Notes Gathering Field Notes 1. During the initial classroom visit introduce self t o the summer school teacher. Ask for a class schedule and ask the teacher to point o ut the targeted students. 2. Visit each classroom weekly at a different time. Sc hedule visits when children are in the classroom. 3. Observations should last approximately 45 minutes. 4. During the visits observe the targeted students. 5. Throughout the observations record field notes usin g a laptop computer. 6. Use anecdotal records to record critical incidents taking place in the classroom during the observation, as well as the specific act ivities the targeted children were engaged in during the observation. (Brandt, 1972). 7. These anecdotal notes contained the classroom activ ity as well as the targeted students attitude during the observation. 8. Next, from the anecdotal records, provide a general ized description of the observation described in the anecdotal records (Tjo ra, 2006). 9. Use Attitude Scale for Classroom Observation (Appen dix R) to determine attitude of child during observation. 10. Then categorize the observation and interpret the c hilds attitude. 11. In an effort to protect the identity of the partici pants, change the names of the schools, teachers and students.
281 Appendix R: (Continued) Analyzing Field Notes 1. Take anecdotal records and determine the unit of an alysis. 2. Develop units into emerging categories. 3. Examine anecdotal records and refer to Appendix R ( Attitude Scale for Classroom Observation) to determine the childs att itude during the observation. 4. Count how many times the category was observed (Tjo ra, 2006). 5. Count how many times the different attitudes were o bserved. 6. The results were then analyzed and presented using charts and histograms.
282 Appendix S: Focus Group Clarification and Probing R outines Clarification and Probing Routines 1. Because children will agree with others to avoid st anding out, it will be important to frequently ask if anyone has other ideas or different opinions. 2. Watch for gestures and facial expressions that may reveal something about the accuracy of a comment, or suggest that so meone has a strong feeling about a question being asked. 3. If a talker takes over the conversation, thank th em for sharing and call on another student. 4. Begin with voluntary responses, if some children ar e not participating, call on them.
ABOUT THE AUTHOR Katie Fradley received her Bachelors degree in Ele mentary Education from Florida State University in 1990. That year she began a career as an elementary school teacher teaching third grade. During the next 18 years she taught every grade from kindergarten through fifth grade. In 1998 after becoming a Natio nally Board Certified Teacher, Fradley realized she needed to further her educatio n. During this same period, while teaching first grade, Fradley developed a fascinati on with struggling readers. This inspired her to pursue a Masters Degree in Reading in 2001 from The University of South Florida. The next year Fradley began taking c ourses towards a Ph.D. in curriculum and instruction at the University of South Florida s Tampa campus. During this time Fradley continued to teach elementary school full t ime, as well as working as a graduate assistant and teaching undergraduate courses on the University of South Floridas Sarasota campus.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001970416
007 cr mnu|||uuuuu
008 081126s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002315
The effect of a summer school literacy program on the reading attitudes of elementary school struggling readers
h [electronic resource] /
by Katie Fradley.
[Tampa, Fla.] :
b University of South Florida,
ABSTRACT: This mixed-method study explored and examined the reading attitudes of thirdgrade struggling readers (n=91) following six weeks of summer school using a scripted literacy program (Voyager Passport). During the quantitative portion of the study students (n=91) from five different summer school sites were given the Elementary Reading Attitude Survey (McKenna & Kear, 1990). The survey, which was administered by the classroom teacher the first day of summer school and the last week of summer school, provided scores for academic, recreational and total reading attitude. Following data collection the results of the ERAS surveys were analyzed using a dependent measures t-test as well as descriptive statistics. Results revealed no significant differences in recreational or total reading attitude following summer school using a scripted literacy program. Gender and school site were both examined using a multivariate analysis.^ Results indicated no statistically significant differences based on gender. However, when academic attitude was examined the results for school site were found to be significant F (4, 90) = 2.87, p = .03. A follow-up Tukey test revealed that although there was a difference in academic attitudes between the school sites, the variation could not be pinpointed to particular sites. The qualitative portion of the study relied on both field notes gathered through classroom observations (n=113) and focus groups. One focus group was held at each of the five summer school sites. During focus groups a group moderator asked the students a series of six questions. Results were analyzed using semantic content analysis (Stewart & Shamdasani, 1990) to identify themes related to students' attitudes about reading. After a cross case analysis of the targeted classrooms was conducted, triangulation was used to compare the findings from the ERAS survey, classroom observations, and focus groups.^ The qualitative findings revealed that following summer school students liked to read, felt they were better readers, and felt prepared to take the standardized test. However, only 29% of the students passed the alternative assessment. The results also revealed questions regarding the fidelity of the implementation and concerns with the lack of norming data on the fidelity measure.
Dissertation (Ph.D.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 282 pages.
Adviser: Susan P. Homan, Ph.D.
x Early Childhood Education
t USF Electronic Theses and Dissertations.