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 001919651
007 cr mnu|||uuuuu
008 071220s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002081
Prediction of English and Spanish early literacy skills of English language learners in the primary grades
h [electronic resource] /
by Giselle Sanchez.
[Tampa, Fla.] :
b University of South Florida,
Thesis (Ed.S.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic thesis) 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 90 pages.
ABSTRACT: This study explored how language, emergent literacy, and reading skills in both English and Spanish develop with a group of English language learners (ELLs) (n = 267). Specifically, the researcher investigated what early language and literacy skills were the most important predictors of reading abilities as indicated by the Book Task in prekindergarten through first grade. Early language and literacy skills were assessed utilizing subtest from the Woodcock Language Proficiency Batter Revised, the Woodcock Language Proficiency Battery Revised Spanish Form and the Phonological Awareness Task. Participants came from households where Spanish was one of the languages spoken. Multiple linear regression and path analyses were utilized to reveal the importance of each predictor variable during each grade level. Results indicated that vocabulary, listening comprehension, letter-word recognition, and phonological awareness are the most important predictors throughout the grade levels. These results are discussed in terms of their potential implications for research and practice with ELLs.
Advisor: Harold Keller, Ph.D.
x School Psychology
t USF Electronic Theses and Dissertations.
Prediction o f English a nd Spanish Early Literacy Skills o f English Language Learners i n t he Primary Grades by Giselle Sanchez A thesis submitted in partial fulfillment of the requirements for the degree of Education S pecialist Department of Psychological and Social Foundations College of Education University of South Florida Co Major Professor: Harold Keller Ph.D. Co Major Professor: Lisa M. Lopez Ph.D. Robert Dedrick, Ph.D. Date of Approval: July 9, 2007 Keywords: Hispanic, Latino, reading, bilingual, achievement Copyright 2007 Giselle Sanchez
i Table o f Contents List of Tables iii List of Figures iv Abstract v Chapter 1: Introduction 1 Statement of the Problem 1 Theoretical Framework 3 Purpose of the Study 4 Research Questions 5 Signi ficance of the Study 5 Organization of Remaining Chapters 6 Chapter 2: Review of the Literature 8 Reading Trends in the U.S 8 Reading Trends of the Hispanic Population 11 Importance of Early Identification of Readin g Difficulties 12 Early Literacy Skills 14 Phonological Awareness 14 Cross Language Transfer of Phonological Awareness 15 Letter and Word Recognition 17 Awareness of Print 19 Writing and Spelling Skills 20 Writing Conventions 21 Oral Language Skills 22 Vocabulary 23 Listening Comprehension Language Recall 25 Conclusions 28 Chapter 3: Methods 29 Participants 29 Ethical Considerations 32 Measures 32 The Phonological Awareness Task 32 The Woodcock Language Proficiency Battery Revised 33 Book Task 35
ii Procedure 36 Data Analysis 37 Chapter 4: Results 39 Overview 39 Descriptive Statistics 39 Reliability 42 Correlat ion Analyses 43 Regression Analyses 46 Assumptions 46 Multiple Regression 47 Path Analysis 52 English Path Analysis 54 Predicting Pre ki ndergarten Book Task Scores 55 Predicting Kindergarten Book Task Scores 56 Predicting First Grade Book Task Scores 56 Spanish Path Analysis 57 Predicting Pre kindergarten Book Task Scores 57 Predicting Kindergarten Book Task Scores 58 Predicting First Grade Book Task Scores 58 Chapter 5: Discussion 59 Summary of Findings 59 Implications for Research 6 5 Limitations 66 Implications for Practice 67 Directions for Future Research 68 Conclusion 70 References 71 Appendices 80 Appendix A: Eng lish Bo ok Task 81 Appendix B: Spa nish Book Task 84 Appendix C: English Book Task Scoring 87 Appendix D: Spanish Book Task Scoring 89
iii List of Tables Table 1 Demographical information of study participants 31 Table 2 Means and standard deviations of each measure by grade level 41 Table 3 Reliability of Book Task 43 Table 4 Correlations between the early literacy measures in 44 pre kindergarten, kindergarten, and first grade Table 5 Correlations between the predictive measures across 45 grade level Table 6 Correlations between predictive measures and Book Task 46 Table 7 Predicting Book Task scores in pre kindergarten 48 Table 8 Predicting Book Task scores in kind ergarten 49 Table 9 Predicting Book Task scores in first grade 51 Table 10 Standardized and unstandardized path coefficients 56
iv List of Figures Figure 1. Path Model 54 Figure 2. English Path Analysis 55 Figure 3. Spanish Pa th Analysis 57
v Prediction of English a nd Spanish Early Literacy Skills of English Language Learners in t he Primary Grades Giselle Sanchez ABSTRACT This study explore d how language, emergent literacy, and reading skills in both English and S panish develop with a group of English language learner s ELLs n = 267 Specifically, the researcher investigated what earl y language and literacy skills we re th e most important predictors of reading abilities as indicated by the Book Task in pre kinder garten through first grade. Early language and literacy skills were assessed utilizing subtest from the Woodcock Language Proficiency Batter Revised, the Woodco ck Language Proficiency Battery Revised Â€ Spanish Form and the Phonological Awareness Task Participants came from households where Spanish was one of the languages spoken. Multiple linear regression and path analyses were utilized to reveal the importance of each predictor variable during each grade level. Results indicated that vocabulary, listening comprehension, letter word recognition, and phonological awareness are the most important predictors throughout the grade levels. These results are discussed in terms of their potential implications for research and practice with ELLs.
1 Chap ter 1 Introduction Statement of the Problem The literacy gap in achievement that exists between Caucasian and Hispanic students is significant. This is a severe problem that needs to be addressed because Hispanic students represent one of the fastest g rowing minority groups enrolled in U.S. schools. Yet, according to current trends about one of every four Hispanic students 27% is bound to drop out before the completion of high school National Center for Educational Statistics, 2003. In a 2000 2001 study by the U.S. Department of Education, researchers noted that English language learners ELL s comprise d a little over one in every 10 elementary school students. Among these students, Spanish constitutes the language spok en by 79.2% of ELL Kindler, 2002. By the year 2030, this group of students will represent 40% of the school age population, yet the majority of U.S. schools are currently failing at providing them with adequate levels of e ducation Thomas & Collier, 2002 With the passage of t he No Child Left Behind Act NCLB in 2002, schools around the nation are being held accountable for the academic achievement of all students. As a direct result, NCLB ensures that educators are held accountable for the progress of ELLs. By the year 2013 NCLB wants all third grade children reading at grade level. Therefore, educators are in a rush to make sure ELLs are acquiring the level
2 of academic English necessary for passing the hig h stakes tests used in the accountability systems Additionally, s tates and school districts are mandated by Title III of the NCLB to develop coherent language policies that delineate how their schools address the needs of ELLs. States should also have standards set up describing the types of skills students should poss es s in core subject are as such as reading. S tandards should be determined by research highlighting skills that ELLs should possess, rather than on monolingual expectations. Regrettably, systematic studies of the earl y reading acquisition of ELLs are quit e limited Gerber, Jimenez, Leafstedt, Villaruz, Richards, & English, 2004. These standards may be unattainable when one considers the research on bilingual language development discussion to follow in Chapter 2. Finally, schools are also being requi red to implement theoretically sound, research based programs that can provide evidence of student learning and achievement Freeman, 2004. It is difficult for ELLs to meet reading benchmarks under a progress monitoring system such as that proposed by NC LB. By definition, ELLs have had very limited exposure to important pre requisite reading skills such as phonology, the alphabet and vocabulary in English. That is why this population is increasingly being identified as at risk for reading failure Gerbe r et al., 2 004. Freeman 2004 explains that the majority of U.S. schools have a language as problem ideological orientation. This ideology views Â languages other than English, and speakers of languages other than English, as problems to be overcomeÂ‹ p viii Such a deficit orientation is unfortunate because it perpetuates the subordinate status of non English languages, and contributes to the poor academic performance of speakers of other languages Freeman, 2004
3 The opposite ideology views langu ages other than English, and speakers of those languages as resources that need to be tapped Freeman, 2004 This is the philosophy of dual language programs that build on the linguistic and cultural resources that many students already possess at the st art of their schooling Freeman 2004 argues that when this resource is tapped, ELLs, their families, and the U.S. in general benefit Even though the research has indicated that effectively implemented dual language immersion programs provide the best long term results fo r ELLs Lindholm Leary, 2001; Thomas & Collier, 2002 opting to implement such programs is filled wit h much socio political debate Freeman, 2004 It may take more lon gitudinal studies to alter the widesprea d deficit ideology tha t exi sts around the nation. This study strives to provide some further understanding of how bilinguals acquire both English and Spanish under the current language as problem ideology. Perhaps such research may serve for future comparative studies with student s whose language is viewed as a resource. Theoretical Framework Two t heoretical frameworks guide this research study The first is the simple view of reading introduced by Hoover and Gough 1990. These researchers have explained that skilled rea ding can be considered as consisting of the product of two necessary skill s: decoding and listening comprehension. The simple view of reading can b e illustrated with the formula R = D x C, where R represents reading comprehension, D represents decodin g skills, and C represents listening comprehension. Such a formula can be interpreted as meaning that a child lacking decoding or listening comprehension skills results in a non read er Students have been found to posses s good decoding skills
4 and at the same time have demonstrated weak reading com prehension abilities as a consequence of their poor language comprehension skills Stothard & Hulme, 1992. Th e simple view of reading lends itself well with this study because the theoryÂšs predictive validi ty was tested with a group of bilingual children in grades 1 through 4. Hoover and Gough 1990 found that about half of the variance in reading comprehension was explained by decoding and listening comprehension. Therefore, these skills can be viewed as essential pre requisites to becoming a good reader. The second theoretical framework provides a basis for understanding the factors involved when students are learning a new language. Cummins 1979 introduced the developmental interdepen dence hypothesis to gain an understanding of how bilingual children learn two languages simultaneously. With this hypothesis, Cummins indicated that the level of second langua ge L2 competence reached by student s is to some exten t related to the competen ce student s demonstrate in their L1 native language at the start of intensive immersion into an L2 setting. Hence, the higher students in this study achieve in Spanish at pre school, the higher they should achieve in English at the end of first grade. Purpose of the Study This study utilize d data from the Early Childhood Study of Language and Literacy Development of Spanish speaking Children ECS 1 th at was developed to g ain information about the language and literacy development of a young group a ge = 4 7 years of ELLs. Additional analyses were conducted with the data co llected by the initial 1 This study is a sub project Prinicipal Investigators: Patton O. Tabors & Marie la M. Paez of a program project titled Acquiring Literacy in English directed by the Center for Applied Linguistics, Washington, DC. The program project is funded by the National Institute for Child Health and Human Development and the Office for Educati onal Research and Improvement, US Department of Education Grant No. P01 HD39530. Additional funding was provided through a National Science Foundation Minority Postdoctoral Fellowship to Lisa M. Lpez award # 0109201.
5 group of re searchers to further explore how la nguage, emergent literacy, and reading skills develop with ELLs. If monolingual students serve as a compa rison for understanding the language development of all children, then bilingual children present differences that need to be analyzed from a perspect ive that allow s for different ways of learning languages Genesee, Paradis, & Crago, 2004. Valdes and Fi gueroa 1994 explained that it is unclear how bilinguals may differ from monolinguals and as a result it is unknown how these differences can affect their performance on standardized measures As a result, this study examine d whether reading skill develo pment for bilinguals is parallel to that of monolingual, English speaking peers or if there are differences that need to highlighted. Research Question s 1. What early language and literacy skills are the most important predict ors of the reading abilities of a group of English language learners ages 4 Â€ 7 ? 2. Is there a differe nce across grade levels in the ability of these early lan guage and literacy skills to predict reading ability on the book task ? 3. What are the direct and indirect effects across grade l evels of the reading task predictors on the book task ? Significance of the Study There exists a gap in understanding how to progress monitor and intervene with ELLs who struggle to re ad in English and who are often misidentified as having learning disabi lities Researchers are establishing a consensus regarding benchmarks for identifying monolingual students who are non responsive to instruction, but there i s a great deal of uncertainty with respect to the application of such identification and
6 intervent ion strategies with ELLs Gerber et al., 2004 Graves, Gersten and H aager 2004 utilized the Dynamic Indicators of Basic Early Literacy Skills DIBELS as an outcome measure with ELLs and they o bserved that about 20 40% of students in first grade classr ooms with top rated teachers still fell below the 40 w ords p er m inute wpm benchmark. Such an observation brings to light the possibility that such a benchmark may be too high for first grad ers not yet fluent in English. These authors stated that resear ch is currently being conducted to address this issue and to determine whether revised benchmarks are indeed necessary Graves et al., 2004. The overarching goal of this present study is to examine how well certain language and literacy skills predict a group of ELLs ability to perform on a book reading task It is hoped that findings from this study can be used to guide future research for establishing standards for ELLs. By gaining a further understanding of the ir l anguage and literacy skills perha ps one day standards can be establish ed that a re more in line with the ELL sÂš true abilities The discussion above regarding NCLB demonstrates that our countryÂšs leaders are establishing national goals in order to diminish the gap that exists between minor ity and majority students Lindholm, 1991 Therefore, researchers are being encouraged to find answers to the multitude of que stions that remain regarding how best to remedy the problems ELLs are facing while learning how to read Organization of Remai ning Chapters The proceeding chapters will highli ght the specifics of this study. Included in Chapter 2 is a review of the literature already published that relates to the development of literacy skills for both monolingual an d bi lingual students. Chap te r 3 descri bes the methodology that was used to conduct the research study including: a description of the
7 participants, ethical considerations, assessment instruments, procedures, research design and data analysis. Chapter 4 provides the results of the current study. Finally, a summary of findings implications for research, limitations, implications for practice, and directions for future resear ch are presented in Chapter 5
8 Chapter 2 Review of the Literature U.S. schools are st ructured to prepare its young citizens to contribute to the nationÂšs economic system and perpetuate the beliefs of that system. One of the ways that U.S. schools prepare students for the labor force i s by teaching them the skills necessary for employment such as reading Ogbu, 1987. Therefore, it follows that e conomic s uccess in the U.S. can seldom occur without knowing how to read. At a minimum, basic reading skills are necessary across all academic disciplines and are generally required in order to a chieve personal, social and economic growth. Once children learn to read, they are endowed with the ability to become reflective and independent learners Good, Simmons, & Smith, 1998. DÂšAngiulli, Siegel and Maggi 2004 explained that a communityÂšs po tential for succes s is to a certain extent depende nt on the literacy levels of its children. Therefore, the success of countries with a great influx of immigrants, such as the U.S., may depend on English language learners ELLs developing adequate readin g skills in th eir second language DÂšAngiulli et al., 2004. Reading Trends in the U.S. A report from the National Center for Educational Statistics NCES has indicated that as much as 37% of children in the U.S. read below basic proficiency leve ls. This percentage has been a rather constant figure for over a decade. In addition, 26% of these students will reach the eighth grade, without attaining a level of basic reading proficiency
9 NCES, 2004. This translates into no less than 10 million ch ildren in the U.S. being poor re aders Fletcher & Lyon, 1998. Unfortunately, based on these findings, the reading ability of todayÂšs youth does not look too promising. The trend described by NCES where the majority of poor readers in fourth grade contin ue being poor readers in eighth grade has been documented by researchers such as Juel 1988 and Stanovich 1986, the latter of whom refers to the phenomenon as the ÂMatthew EffectÂ‹ in reading. Stanovich 1986 makes the comparison to the biblical Matthe w effect where the rich get richer and the poor get poorer. S tudents who possess the riches e.g. phonological awareness at the beginning of their schooling experience go on to acquire even more riches e.g. comprehension. In the mean time, those stu dents who do not initially posses s such riches end up unlikely to ever acquire subsequent wealth because of their inadequate exposure to reading. The trajectory starts very early in life. Stanovich 1986 discussed the importance of acquiring phonologic al awareness at an early age as a prereq uisite for success in reading. Phonological awareness refers to the conscious awareness of the sound structure of language. It includes the ability to detect, manipulate, and think about the various units in spoken language e.g., syllables and phonemes separate from meaning Sindelar et al., 2002. In addition, s tudents who do not gain an understanding of the alphabetic principle early on will tend to become poorer readers and get trapped in a negative spiraling trend. Understanding the alphabetic principle involves knowing that letters and sounds work tog ether to make words. Children must possess skills in both phon ological awareness and alphabetic principle plus certain basic print concepts before they can d ecode words. Weak decoding skills restrict the amount of text students get exposed to.
10 Then the lack of practice with current reading materials leaves students unprepared to tackle more difficult curriculum as the school year progresses Stanovich, 1986 Finally, students perceive their efforts as worthless when faced with a curriculum that is too advanced and do not benefit from the rewarding experiences that reading may offer. In contrast, Stanovich 1986 reported that students who break the alphab etic code early get exposed to a much greater amount of text about three times as many words as poor readers. Such practice in reading is likely to result in a growth in vocabulary. In the end, these early decoders gain general knowledge and a greater understanding of complex syntactic structures through their continued exposure to print Stanovich, 1986. Although the Matthew effect can be observed with any group of students, a prime example exists between Hispanic and Caucasian students. When compar ed to Caucasian age mates, a greater proportion of Hispanic students are falling behind on reading achievement from an early age. Plus, when the reading achievement of Hispanic adults was recently assessed, the gap in reading achievement had grown NCES, 2003; NAAL, 2003. Therefore, ELLs are starting out with weaker literacy skills and that gap only continues to increase with time. In the previous chapter, references were made alluding to the severity of the problem that educators face in helping ELLs to acquire adequate reading skills. Therefore, f or the remainder of this chapter researc h will be reviewed highlight ing the following: the reading trends of the Hispanic population, importance of early identification of reading difficulties, early l iteracy skills and oral language skills. A focus will be maintained on studies conducted with bilingual Hispanic children Yet, the research conducted with monolinguals will be discussed when appropriate because the
11 greatest amount of research on readin g development has been conducted with this population. There will also be some interspersed discussion of whether or not these early language and literacy skills have been shown to transfer across English and Spanish. Reading trends of the Hispanic p opul ation Students who live in U.S. households where languages other than English are spoken represent approximately 9.9 million of 45 million school aged children. Such demographics denote a 35% increase in this population since 1980. Of the 9.9 million s tudents, about two thirds six million are children whose home language is Spanish Lindholm Leary, 2001 Hispanics represent the fastest growing minority group in the U.S. U.S. Census Bureau, 2000. Unfortunately, Hispanics also represent the poorest and least educated group in the nation Lopez & Cole, 1999. One of the primary academic areas that Hispanic students are struggling with is reading. Overall, Spanish speaking students represent the lowest achieving cultural group in the U.S. when it c omes to reading achievement. It is difficult to interpret the performance of ELLs on English reading tasks because of their limited exposure and proficiency in English Gerber et al., 2004. This results in disproportionately high levels of special educa tion referrals and climbing rates of learning disabled labels Gunn, Smolkowski, Biglan, & Black, 2002. Findings from the 24 th Annual Report to Congress on IDEA indicate that over 17% of students identified as learning disabled are Hispanic yet they only represent about 12 13% of the population Office of Special Education Programs, 2002. In addition, researchers have pointed out that an under identification problem may also exist with English language learners ELLs if educators erroneously attribute early
12 learning difficulties to delays in learning English Gerber & Durgunoglu, 2004; Geva, 2000. Out of fear of inaccurately attaching a disability to a student who just needs time to learn the language, educators may be refraining from seeking support s for struggling ELLs. Very little information is available regarding the prevalence of learning disabilities among ELLs, especially reading disabilities Gerber & Durgunoglu, 2004. Therefore, the reading development of ELLs needs to be studied in order to design assessment s that can differentiate children at risk for reading disabilities from those at risk in acquiring their second language Gerber et al., 2004. Importance of Early Identification of Reading Difficulties As reported above, learning to read can become quite a difficult feat for a great number of children, and mainly depends on developing language and literacy skills at an early age Fletcher & Lyon, 1998. Research on reading in the 1990Âšs focused on identifying prerequisite reading sk ills in attempts to decrease the number of students experiencing difficulty learning to read. Pre requisite skills for decoding have been identified to include phonological awareness, print awareness, and the alphabetic principle Sindelar et al. 2002. It is crucial to identify reading problems by assessing prerequisite skills because, as Juel 1998 reported, by the end of first grade, students with good reading skills see about 18,681 words in text. Yet, poor readers only see about half that many wor ds, 9,975. As exemplified with the Matthew effect, as early as first grade it is already difficult for poor readers to catch up to their peers. Good et al. 1998 illustrated that to reach their peers, poor readers need to increase their progress by a r ate of 3.5 standard deviations and acquire skills twice as fast as their peers. Such progress is quite a difficult feat for initially poor readers to
13 overcome. In essence, early id entification and intervention are crucial because the longer it takes to i dentify a child with basic reading difficulties, the more difficult it becomes to provide successful remediation Lyon, 1996. Shaywitz, Fletcher, and Shaywitz 1994 reported that 74% of students with a reading disability who do not receive intervention before the age of nine will continue exhibiting a reading disability throughout high school. These findings suggest that a critical period for intervening exists. Ideally, schools should strive to conduct diagnostic assessments in preschool Dufva, Niemi & Voeten, 2001. Although a number of studies with monolingual children have highlighted how critical early language and literacy skills are for later success in reading, only a handful have been conducted with bilingual students Whitehurst & Lonigan, 1998. Longitudinal studies analyzing the development of English reading skills of ELLs had not been documented until fairly recently e.g. Lindsey, Manis, & Bailey, 2003; Lesa ux & Siegel, 2003; Swanson Gerber, & Leafstedt, 2004. What follow s is a brief description of a study with such a research goal. This study utilized data collected by The Early Childhood Study of Language and Literacy Development of Spanish speaking Children ECS Tabors, Paez, & Lopez 2003; preliminary findings fr om this study will be discussed throughout this chapter. The ECS was developed to answer a number of questions relating to a group of English language learners Paez et al., 2007 Children were assessed in English and Spanish from pre kindergarten to se cond grade in order to identify factors related to their language and literacy development in each language. The main goal with this group of
14 students was to see if language and literacy development differed from that of monolinguals Paez et al., 2007 The principal investigators pointed out two groups of literacy related skills emphasized through research with monolinguals e.g. Dickinson & Snow, 1987; Lonigan, 2003; Storch & Whitehurst, 2002 that are important prerequisites of reading during the preschool years The first group of skills are referred to as early literacy skills and include phonological awareness, letter and word recognition, and writing and spelling skills. The other group consists of oral language abilities and includes voc abulary and listening comp rehension skills Tabors et al. 2003. Early Literacy Skills Phonological a wareness There has been surmounting evidence relating studentsÂš phonological awareness skills with reading achievement in English e.g. Adams, 1990 ; Swan & Goswami, 1997. Good et al. 1998 stressed that so much empirical support e.g. Stanovich, 1986; Torgesen & Hecht, 1996 has been found pointing to poor readers as having phonological deficits that Âit has been deemed a core deficitÂ‹ p. 46. De velopment of good phonological awareness skills depends on explicit reading and spelling instruction. Therefore, children do not naturally gain phonological awareness unless they are exposed to print Goswami, 2001. According to Lyon 1996, m ost c hildren easily acquire phonological awareness by six or seven years of age. Yet, about 17% of children experience significant difficulties picking up these skills and struggle learning to read Lyon, 1996 Unless phonological awareness is acquired, such children will continue to struggle with reading, regardless of their intellectual capabilities Lyon, 1996. Morris, Bloodgood, Lomax, and
15 Perney 2003 explain that children will be unable to make use of the alphabetic code and match the letters in word s with their corresponding sounds until they can segment spoken words into phonemes. Cross language transfer of phonological a wareness Durgu noglu, Nagy and Hancin Bhatt 1993 noted that Spanish phonological awareness also predicts studentsÂš word r ecognition skills in Spanish. Durgunoglu and colleagues set out to study whether phonological awareness skills transferred across English to Spanish and their findings pointed out that they did indeed transfer. It was observed that Spanish phonological a wareness was related to student s Âš ability to recognize words in English. If students scored high on measures of phonological awareness in Spanish, they were also more successful at reading English words and pseudowords. When bilingual students from the E CS sample were assessed during the end of pre kindergarten, they achieved higher scores in the Phonological Awareness Task in Spanish than in English Paez et al. 2007 Phonological awareness is a metalinguistic ability, and is not language specific D urgunoglu et al., 1993. In alphabetic languages students are required to Âidentify the phonological subcomponents of the spoken words and understand how orthographic symbols are mapped onto those phonological subcomponentsÂ‹ p. 462. Regardless of the alphabetic language in which students develop these metalinguistic skills, they utilize them when attempting to read in a new language. Durgunoglu 1998 later studied the word recognition and spelling performance of Spanish speaking students. Results i ndicated that Spanish word recognition and spelling are best predicted by the studentsÂš degree of phonological awareness and letter knowledge in Spanish. Confirming DurgunogluÂšs previous findings, evidence was also
16 found to support the ability of Spanish proficiencies such as phonological awareness, word recognition and spelling to predict performance in English word recognition. Although Dur gunoglu Âšs sample did not receive formal reading and writing instruction in English, these students were able to rea d environmental print and spell words in English as a result of developing such skills in Spanish. Phonological awareness has been observed to transfer across languages as early as preschool Dickinson, McCabe, Clark Chiarelli, & Wolf, 2004; Lopez & Green field, 2004. Since a number of studies have provided evidence for the cross language transfer of phonological awareness e.g. Cisero & Royer, 1995; Comeau, Cormier, Grandmaison, & Lacroix, 1999; Durgunoglu et al. 1993; Lindsey et al., 2003; Ri ccio, Amado, Jimenez, Hascrouck, Imhoff, & Denton, 2001, it was suggested by Durgunoglu 2002 that educators may assess studentsÂš L2 phonological awareness skills as long as they have received sufficient instruction in L1. Riccio et al 2001 pointed o ut that Spanish measures of phonological processing may be essential for early identification of students at risk of developing reading problems in English and Spanish. These authors emphasized that identification in early grades e.g. kindergarten, firs t grade may result in proactive programming that may prevent the development of reading problems. Such assessments can take place even if ELLs have not developed an extensive vocabulary in their L2. This recommendation is made because phonological awaren ess transfers even though vocabulary does not Durgunoglu, 2002. Phonological awareness was measured as early as age 4 with the ECS, pre kindergarten sample. Although they displayed stronger skills in English than in Spanish for all the skills assessed, they performed better in Spanish on the phonological awarenes s measure Paez et al., 2007
17 Letter and word r ecognition Letter knowledge is an emergent literacy skill and one of the best predictors of later reading achievement. The term letter recogniti on refers to a childÂšs knowledge of the letters of the alphabet. A child who enters school knowing the alphabet is likely to experience success with reading in the future Pullen & Justice, 2003; Whitehurst & Lonigan, 1998. When students have poor phon ological awareness they cannot decode recognize words at a fast rate and words are often read inaccurately Lyon & Chhabra, 1996. Of all the children in public schools identified as learning disabl ed, about 70 80% of the student s Âš primary area of impai rment is in reading. Of those students with reading impairments, 90% of them have problems with word recognition. Children with good word recognition skills realize Âthat words have an internal structure based on their sounds and represented by the alpha betÂ‹ Fletcher & Lyon, 1998, p. 58. Children with deficient word recognition skills are unable to segment syllables and words into their smallest sound units, phonemes Lyon, 1996. In order to determine the types of skills that students possess i n kindergarten that might predict their reading skills in second grade, Baker, Fernandez F ein, Scher and Williams 1998 studied a couple of prerequisite skills. They found that knowledge of nursery rhymes in kindergarten best predicted word attack 36% o f the variance and word identification skills 48% of the variance. The predictor that made the second strongest contribution was letter knowledge. An extra 11% and 18% of the variance for word attack and word identification, respectively, was explaine d by letter knowledge Baker et al., 1998. Quintessentially, the best way one can observe whether a child has a
18 re ading problem is to assess his or her ability to read single words from a list or text Lyon & Chhabra, 1996. Certain educators have expr essed concern about bilingual children mixing their languages and not being able to differentiate the structure and rules of each Durgunoglu, 2002. This occurs because negative transfer is oftentimes witnessed in word recognition and spelling. Durgunog lu et al. 2002 examined a number of circumstances where bilingual students inappropriately used their Spanish skills to spell in English and vice versa. One of the reasons why such a phenomenon occurs is because Spanish orthography has a more transparen t sound to spelling correspondence. To a greater degree, words in Spanish are written the way that they are heard. Therefore, Spanish speakers tend to make the mistake of thinking that English words are spelled as transparently as they are in Spanish. A n example of when English is mixed into Spanish is observed when English consonant clusters are used to spell in Spanish Durgunoglu et al., 2002. Such orthographic transparen cy may actually be beneficial to ELLs. An argument was made by Koda 1997 th at orthographic knowledge in L1 has the potential to assist in word recognition and lexical processing in L2. The amount of transfer depends on how orthographically similar both languages are. English and Spanish are quite orthographically similar and wi th the transparent nature of Spanish orthography, it may actually be easier for a Spanish speaking student to learn Geva & Wang, 2001. When word recognition and spelling performance were assessed by Durgunoglu 2002 at more comprehensive levels, po sitive correlations were still detected both within and between languages. Paez and Rinaldi 2006 investigated whether transf er of word
19 recognition skills f r o m Spanish to English was replicable with the ECS sample. They indeed found that Spanish word re cognition skills in kindergarten predicted English word recognition skills in first grade. Therefore, educators should be advised to look at the broader picture when they witness the errors bilingual students make while decoding words and spelling. In ge neral, students are transferring their decoding and spelling skills from one language to the other Durgunoglu, 2002. The couple of instances where incorrect transfer occurs should be considered similar to the invented spelling of monolinguals. With eno ugh practice and exposure to English and Spanish, bilingual students should eventually acquire the correct decoding and spelli ng abilities in each language. Paez et al. 2007 hypothesized that students would demonstrate significant improvements in early English literacy skills and such gains were not observed. In addition, they fel l even further behind the Spanish norm with Letter Word Identification. Awareness of p rint Pullen and Justice 2003 explain that Âa childÂšs awareness of the forms, functio ns and uses of print provide the foundation upon which reading and writing abilities are builtÂ‹ p. 89. Children who are read to often are aware that book language differs from spoken language Clay, 1991. Print knowledge was described by Manis, Lin dsey and Bailey 2004 as a composite variable resulting from combining two measures that assessed letter naming skills and concepts about print. Students were asked to provide the researchers with letter names and sounds in both English and Spanish. To assess concepts about print, the investigators utilized childrenÂšs books in Spanish to ask questions evaluating their understanding of book and print conventions. The assessment paralleled the English ver sion of ClayÂšs 1979 Concepts A bout Print task. M anis et al. 2004 analyzed four variables print
20 knowledge, phonological awareness, rapid automatic naming RAN and expressive language which all resulted as significant predictors from Spanish to English. Print knowledge turned out to be the single b est predictor across languages. According to these findings, quantity of early exposure to print in Spanish best predicts future reading skills in English Manis et al., 2004. The assessment of Concepts About Print 1979 has been described as a good too l for non readers because educators can determine whet her the skills that precede word reading are present. These skills should develop during the first two years of schooling and educators should witness an increase in these skills as reading improves. Test retest r eliability reported by Clay for a group of urban children was .95 Clay, 1970 The validity of Concepts Abo ut Print was also investigated by analyzing the measuresÂš correlation with word reading abilities of 100 six year old children. A cor relat ion of .79 was found Clay, 1966 A group of 56 kindergartene rs were also assessed and test retest reliability coefficients of .73 to .89 were obtained. Corrected split half coefficients with the same group of children were reported between .84 and .88 Day & Day, 1980 Writing and spelling s kills In order to be successful with spelling, a child needs to have good phonological awareness skills and knowledge of the alphabetic principle. MacDonald and Cornwall 1995 pointed out that the best predictor of spelling 11 years after kindergarten is phonological awareness. Their longitudinal study indicated that phonological awareness was a stronger predictor of a studentsÂš ability to spell than vocabulary development, word recognition and spellin g achievement. These results were obtained from a sample of 58 White monolingual students from Nova Scotia.
21 Students from the ECS sample assessed during the end of pre kindergarten did not show substantial improvements in their spelling or writing ski lls. Even though these students did not demonstrate expected gains, they scored better on early literacy measures than they did on oral language measures in both English and Spanish. Nevertheless, their early literacy skills fell below the means of their monoling ual peers Paez et al., 2007 Writing c onventions Having an understanding of writing conventions require s more advanced skills than the previously mentioned language and literacy skills, but such skills are still found to transfer across lang uages. These skills can be assessed by providing students with pictures and asking them to write a story describing what is happening in the picture. A different picture can be used for each language being assessed. Durgunoglu 2002 explained that stud entsÂš awar eness of writing conventions was similar between languages. If students knew how to formulate detailed and well organized plots that included character descriptions, a conflict, a climax and an ending, they demonstrated this knowledge across lan guages Durgunoglu, 2002. Through analyses of studentsÂš writing assignments, Zecker 2004 observed spontaneous cross language use with students enrolled in a dual language program. The students were either English dominant or Spanish dominant. Yet, they incorporated words from their non dominant language as early as six weeks after they began school. This is an interesting finding because students were not asked or required to write in their non dominant language so early in their bilingual instruc tion. Zecker 2004 provided examples of how children applied their developing understanding about writing towards writing in their second language. When students were finally required to write in their non dominant language, they did not go back to more elementary forms of writing.
22 Instead, they applied their already established knowledge of writing in biliterate ways Zecker, 2004. Oral Language Skills Research conducted with monolingual students has indicated that oral language is highly correlated with achievement in reading comprehension Biemiller, 2003. The oral language abilities of preschool children were evaluated by Catts, F ey, Zhang and Tomblin 1999 in order to analyze their relationship with reading disabilities in second grade. Their findings revealed that over 70% of the weak readers studied exhibited language deficits in kindergarten. These deficits we re primarily related to difficulties with phonological processing and oral language Catts et al., 1999 Having adequate oral langua ge skills in English is also crucial for ELLs as they learn to read in English Carlisle, Beeman, Davis & Spharim, 1999; Proctor, Carlo, August, & Snow, 2005. Lindholm 1991 summed up the findings between oral language skills and reading as follows: re ading and academic language skills are highly dependent, oral English proficiency and academic English proficiency are not correlated, and both types of language proficiency are correlated with studentsÂš ability to read in English. Such findings have sign ificant implications for working with ELLs because both types of language skills, academic and conversational, should be developed in each language before educators can observe high levels of reading achievement Lindholm, 1991. With proper instruction, it takes about five to seven years for ELLs to be able to achieve grade level norms Thomas & Collier, 2002
23 Vocabulary To decrease the number of ELLs being misidentified for special education services, it is vital to evaluate the validity and relia bility of vocabulary scores. D urgunolu et al. 1993 are some of many researchers e.g., Cobo Lewis, Eilers, Pearson, & Umbel, 2002; Manis et al., 2004 who have emphasized that oral language proficiency measures in Spanish are not reliable for predicting studentsÂš reading abilities in English. The single greatest indicator of oral language proficiency is vocabulary. Vocabulary knowledge is vital to comprehending both spoken and written language Proctor et al., 2005. No significant correlations were f ound between oral proficiency levels and word recognition or phonological awareness skills Durgunoglu et al., 1993. A within language relationship was found by Lopez and Greenfield 2004 between language proficiency and phonological awareness, but not across languages. Their findings also confirm that oral language proficiency has not been observed to transfer across English and Spanish. Nevertheless, oral proficiency is necessary for phonological awareness to transfer. Therefore, if students are ev aluated based solely on oral proficiency and they perform well, they may be denied the supportive services needed to enhance their poor reading abilities. Such findings also draw attention to the fact that vocabulary needs to be taught in each language. When ELLs Âš vocabulary development occurs at a slow rate, their ability to comprehend text is affected, and their chances of being misdiagnosed with a learning disability increases August, Carlo, Dress ler, & Snow, 2005. The findings of Umbel, Pearson, F ernandez and Oller 1992, revealed that students who came from bilingual homes scored significantly below the norming sample in English vocabulary despite the fact that the SES of the ELLs was higher than the norming sample. Struggling Hispanic
24 readers w ere assessed and demonstrated that L2 vocabulary and phonological awareness each independently contributed to reading comprehension in L2 Ca rlisle et al. 1999. The opposite was seen when word reading abilities were assessed and the effects of SES and literacy instruction were controlled DÂšAngiulli et al. 2004. By Grades 3 or 5, English language learners on the lower and higher end of the SES spectrum were observed to have better word reading abilities than their L1 comparison group. Such a fi nding may be the result of the intensive literacy program that was in place for the intervention students, and not simply SES. In addition, the students in this study may have been able to read the words, but not have understood what the words meant. Some precautions need to be taken when interpreting the results of DÂšAngiulli et al. 2004 because their sample of ELLs consisted of students in North Vancouver rather than Hispanic students in the U.S. A longitudinal study similar to that of DÂšAn giulli et al. 2004 that controls for SES and instruction would be ideal for investigating the potential growth in vocabulary of Hispanic ELLs. Adding a measure of vocabulary can inform educators whether certain literacy instructional programs also help to diminish the gap that exists in word knowledge between Caucasian and Hispanic students. When no intervention is in place to adequately teach vocabulary, the gap has been observed to get wider through the years. As the school year progressed, pr e kindergarten students from the ECS sample were observed to lose ground in Spanish vocabulary. Overall, their oral language skills in Spanish fell well below norms. As early as pre kindergarten the pattern of Hispanic students losing skills in Spanish i n exchange for acquiring the English needed for school success wa s evident Paez et al., 2007
25 Listening comprehension language r ecall A bidirectional relationship has been observed between reading and listening comprehension. Generally, high leve ls of listening comprehension correlate with high levels of reading comprehension and vice versa. It has been observed that such correlations are not as strong early on but become stronger once word recognition becomes automatic Curtis, 1980; Palmer, McC leod, Hunt, & Davidson, 1985 Nevertheless, some students may exhibit poor comprehension despite having well developed word recognition skills. Such a phenomenon can be attributed to general language comprehension skills deficits Stothard & Hulme, 1992 In a study conducted by Duvfa et al. 2001, reading comprehension was assessed and observed to be more highly correlated with listening comprehension than word recognition in a group of Finnish students. Two hundred twenty two students were followed from preschool through second grade in one of the most extensive analyses of phonological memory and reading. The variables studied included verbal abilities, phonological memory, phonological awareness, word recognition, listening and reading comprehens ion. The authors concluded that phonological memory was a weak predictor of word recognition and reading comprehension if phonological awareness and listening comprehension were utilized in their structural equation model. High stability was observed whe n the development of comprehension skills were examined from preschool listening comprehension to second grade reading comprehension. They therefore concluded that the two most reliable skills for predicting word recognition a nd reading comprehension were phonological awareness and listening comprehension Duvfa et al., 2001.
26 Garcia 1 991 administered expository tex t passages to a group of fifth an d sixth grade Hispanic students. The children were provided with the passages in order to follow along with the examiner. Nevertheless, the examiner read the passage out loud to make the assessment a measure of comprehension and not solely a measure of studentsÂš ability to read in English. Results indicated that the ELLs studied were able to produce longe r and more accurate recalls of the English text when they were allowed to use their first language. Garcia 1991 claimed that such findings were in line with previous research conducted with ELL s. Langer, Bartolome, Vasquez and Lucas 1990 found that s tudentsÂš ability to comprehend text in English and Spanish depended on how well their meaning making strategies were developed in either language. The only thing required of students was basic proficiency in English to be able to comprehend text, once the y demonstrated that such strategies were being used. Students who had good meaning making strategies in Spanish were observed to transfer such skills when reading in English and vice versa Langer et al 1990. Another interesting finding was that thes e aforementioned reading comprehension strategies distinguished good readers from poor readers, not their level of English fluency. Yet, the degree of competence in Spanish was related to the studentsÂš enhanced ability to make meaning of text in both lang uages. Students claimed that they thought of words and concepts in Spanish while reading in English, especially when they experienced difficulties with the text. In general, students more successfully recalled content, generated hypotheses, and provided elaborate recalls while reading in Spanish. The students were also better able to respond to decontextualized questions when such
27 questions were asked in Spanish and they were allowed to respond in Spanish Langer et al 1990. GarciaÂšs 1991 findings were discrepant wit h those of Paez et al. 2007 because the latter group indicated that as early as pre kindergarten ELLs Âš language recall skills were superior in English. These differences may be a result of the instructional programs t he bilingual students were enrolled in during their early elementary years. Garcia 1991 reported that only the students who stated being literate in both English and Spanish were included in their study. Only a few of those students had been enrolled i n bilingual programs and transitioned into English only classrooms in second or third grade. Nevertheless, the exact percentage of bilingual program enrollment was not reported. In addition, no assessments were conducted in Spanish to verify the validity of the student claims. It is also possible that the students who reported being literate in both languages may have received instruction in their country of origin. Too many variables were left uncontrolled in GarciaÂšs 1991 study. In spite of that, t he discrepancies observed may have resulted because the students received reading instruction in their first language, as opposed to those studied by Paez et al. 2007 Paez et al. 2007 obtained results similar to previous research in dicating that the oral language skills of bilinguals are quite limited. Therefore, both essential components of reading comprehension, word recognition and listening comprehension skills, should be measured in order to obtain a more accurate assessment of students true reading abilities.
28 Conclusions School psychologists sensitive to linguistic and cultural differences of Latino families may be instrumental in optimizing the educational attainments of ELLs. Their awareness of data based instructional practices in early language and literacy can be beneficial when collaborating with parents and teachers of ELLs. This knowledge is especially valuable in states or school districts that only provide instruction in English for bilingual children Quiroga e t al., 2002. The study was conducted in hopes of adding some understanding to the field. The findings may add to the existing knowledge base available for making data based instructional decisions.
29 Chapter 3 Methods Participants The participants in the Early Childhood Study ECS were recruited by contacting the parents of children in Head St art and public preschool program s in three communities in Massachusetts Boston, Framingham, and Lawrence, and one community in M aryland Montgomery County All of the chi ldren were 4 years old at the beginning of pre kindergarten and were age qualified to attend kindergarten the following year. Additionally, the children in the sample in Massachusetts and Maryland were livi ng in homes where Spanish was at least one of the languages spoken. The ECS sample wa s made up of 267 50% female, 50% male chi ldren who were assessed at three different times. The first assessme nts occurred in the spring of 2002 as the students exited their pre kindergarten programs T ime 1. Follow up assessments occurred as students completed kindergarten Time 2 and their first Time 3 grade year. The mean age for the ECS sample at Time 1 was 4.46 years at Time 2 the students were 5.41 years ol d and at Time 3 they were 6.38 years old. The majority of the study participants were born in the United States 80.2 % and 4 .5% were born in the U.S. territory of Puerto Rico. The remaining percentages consisted of children born in various countries in Latin America. Although most of the children in this sample were born in the U.S. their parents ca me f r o m 22 countries in
30 Latin America and the Caribbean, including the U.S. territory of Puerto Rico. Twenty three percent of the sample did not have a fat her or male figure present in the home. Paez et al. 2007 described the children in the ECS sample as being dual language learners. Therefore, they have either learned both English and Spanish from infancy or are learning English as a second language onc e enrolled in their pre kindergarten programs. Since the f indings of Hammer, Miccio and Wag staff 2003 indicated that these two groups of young children showed similar early literacy skills, they were Âconsidered as one group with diverse dual language a bilities based on their personal hi storiesÂ‹ Paez et al., 2007 The sample of children represent a variety of family backgrounds in regards to language use at home, parental years of education, and family income. Sixty three percent of the mothe r s interviewed reported using only Spanish at home and an additional 20% spoke mostly Spanish Thus for 83 % of the families, Spanish was the language mainly used to communicate at home. Of the remaining families, 12.5% claimed to use both Spanish and En glish at home and only 5% of mothers claimed to speak mostly in English to their children Levels of parental education ranged from 0 to 22 years A third of the sample of mothers reported having some higher education. Close to a third of the fathers also reported having some higher education. Overall, there did not seem to be great discrepancies between the educational levels of mother s and fathers in the sample. As for family income, 79.2 % of the families in the sample reported making less than $3 0,000 with 18 .4 % making less than $10,000. Of t he remaining families, only 11 % had a household income of more than $40,000. These numbers are expected, considering
31 that the m ajority of the student s in the study attend ed Head Start programs which primari ly serve low income families. For more details on the participantsÂš demographical information, see Table 1 below. Table 1 Demographical Information of Participants Language use at home n = 269 % of students Lan guage Spanish only 63% Mostly Spanish 20% Spanis h and English equally 12.5% Mostly Engl ish 5% Level of parental education Years of education Mother Father 8 or less 20% 27% 12 26% 19% 13 + higher education 33% 30% Family Income Income % of familie s Less than $10,000 18.4% $10,000 Â€ $19,999 31.4% $20,000 $29,999 29.4% $30,000 $39,999 9.8% $40,000 $49,999 4.3% $50,000 $59,999 2.7% $60,000 $69,999 2.4% $70,000 $79,999 .4% $80,000 or more 1.2% The bilingual children in the ECS sample attended 68 pre kindergarten classrooms, 58 of which were Head Start classrooms, and the remainder w ere preschool programs in public schools. English was the dominant classroom language for all except
32 one all Spanish classroom; however, 21 classrooms experienced varying levels of language use in Spanish. A greater degree of specification regarding the amount of instruction that took place in Spanish was not available to the researcher. Ethical Considerations The researchers of the ECS following ethical guidelines sought permission from their Institutional Review Boa rd IRB to conduct the assessmen ts being utilized for this study. Nevertheless, the current research er also sought permission from the University of South FloridaÂšs USF IRB in order to ensure that no analys es were conducted w ith the data unless they also m et USFÂšs ethical guidelines. No analy s es were conducted until the study wa s approved by the IRB committee. Measures The lan guage and literacy battery us ed in this study wa s based on p r e vious work on the language and literacy skills of young children Dickinson & Tabors, 2001 ; Snow, Tabors, Nicholson, & Kurlan d, 1995 while taking into consideration three further criteria: a a need to have tasks in both Spanish and English, b the need to have as many instruments as possible that are of high reliability and validity, and c the need to have tasks that are appropriate for the age range ages 4 to 7 years under consideration. The complete battery has been designed to provide data about a group of constructs that have been shown to be related to childrenÂšs later literacy achi evement Dickinson & Tabors, 2001. The phonological awareness t ask Lopez, Tabors, & Paez, 2002 The Phonological Awareness Task was developed by the research team specifically for the ECS since equivalent tests in Spanish and English were not avail able that were
33 appropriate across the needed age range. There are five subtests: rhyme recognition, rhyme production, initial phoneme recognition, sentence segmenting, and syllable segmenting. There are two versions of the test, one in Spanish and one in English. These two versions tap the same skills, but have been constructed separately to demonstrate the childrenÂšs phonological abilities in each of their languages. A full description of this task can be found in Tabors et al. 2003. Adequate test retest reliability was reported for both the English .68 and Spanish .59 versions of the test. Internal consistency for the English assessment was reported as moderately high = .81and .86. For the Spanish assessment s, a moderate consistency = .78 and .79 was found for more infor mation see Tabors et al. 2003. Such internal reliability in both languages allows the researcher to proceed with the statis tical analysis of this measure using mean s cores The woodcock language proficiency b attery r evised WLPB R Woodcock, 1991 The Woodcock Language Proficiency Battery Revised is a standardized assessment consisting of a set of subtests used to meas ure different aspects of language and literacy skills. There are two versions of these tests, one in Spanish and one in English. Standard scores for all of t he WLPB R subtests are normed with a mean of 100 and a standard deviation of 15. The English For m of the subtests was normed on a randomly selected population of 6,359 English speakin g individual s in the United States. The sample was stratified and weighted so that the population is representative of the distribution and characteristics of the US po pulation. Consequently, the norms for these assessments were developed from monolingual English speaking children.
34 The Woodcock Language Proficiency Battery Revised Spanish Form Woodcock & Muoz Sandoval, 1995 is parallel in content and structure to t he English Form. The Spanish Form of the subtests was normed on 3,911 native Spanish speaking subjects from both inside and outside the United States Of these participan ts, 116 were tested in Costa Rica, 1,512 in Mexico, 196 in Peru, 634 in Puerto Rico, 128 in Spain, and 1,325 in the United States. A lthough some of the participant s used to provide norming data for these assessments lived in the US, these children were, by design, monolingual Spanish speakers Woodcock & Mu oz, 1995 Consequently, the norms for these assessments were essentially developed from monolingual Spanish speaking children. The reliability and validity characteristics of both forms of the WLPB R meet basic technical requirements see Woodcock, 1991b, p. 124. The four subtest s being used in this study from the WLPB R include Letter Word Identification Identificacin de Letras y Palabras, Dictation Dictado, Picture Vocabulary Vocabulario Sobre Dibujos, and Memory for Sentences Memoria para Frases. The Letter Word Ident ification subtest first measures symbolic learning through the use of rebuses the use of a pictogram to represent a syllabic sound followed by identification of letters and then word decoding. This test is being used in the present study to index child renÂšs letter and word recognition abilities. The first items in the Dictation subtest measure childrenÂšs prewriting skills, followed by items measuring their knowledge of letterforms, spelling, punctuation, capitalization, and word usage. This test is be ing used in the present study to index childrenÂšs writing and spelling skills. In the Picture Vocabulary subtest children are asked to select pictures to match words and to say a word when shown a picture. Although a childÂšs receptive vocabulary
35 skills are measured at the beginning of this test, this is primarily an expressive vocabulary task. This test is being used in the present study to index childrenÂšs vocabulary. In Memory for Sentences children are asked to repeat words, phrases, and then whole sentences. This subtest requires the use of both short term memory and ability to extract meaning from the sentences in order to aid recall. This test is being used in the present study to index childrenÂšs listening comprehension/ language recalling skil ls. Book t ask. The book task i s a measure similar to that of ClayÂšs Concepts About Print 1979 task see Appendix A & B Â€ for a copy of the measure s The assessme nt i s divided into four sections and utilizes the childrenÂšs book Carrot Seed by Ruth Krau ss 1989 If a student accurately completes a section of the b ook task, the examiner continues administration with the next sec tion. Otherwise, the measure i s discontinued if the child scores below the established criteri on Sample questions in the fir st s ection of the book task involve asking a child to point to the front of the book and later to open the book for the examiner. If the child successfully completes these basic book tasks then section two i s administered. In the second sect ion, the ex aminer read s the book to the child and asks questions regarding the ev ents in the story. The child i s also asked to point to certain characters in the story. Again, if a s ufficient number of questions a re answered corre ctly then the examiner proceeds to the third section. Here the examiner hands the child the book and asks the child to tell the examiner the story. If the child attempts to decode text, the examiner prompts the child to first just describe what the story is about because they will read i t later. Finally, if the child i s able to retell the story at least partia lly, then the
36 examiner proceeds to the fifth section. At this point the child i s asked to read the story to the examiner. The same procedures a re followed with th e Spanish versi on of the book, La Semilla De Zanahoria Palacios 1996 see Appendix B Each of the versions contained 16 items and students co uld obtain a maximum score of 24 points. For mos t of the questions, the child i s able to obtain two points for answering cor rectly and half the credit 1 point for partially correct r esponses. Plus, if the child i s able to read the b ook in its entirety, 3 points a re awarded in the third s ection of the task. Children a re not allowed to code switch Therefore, if students ans wer correctly in English while being assessed in Sp anish, the response is scored a s incorrect and vice versa see Appendix C & D for scoring Students were reminded to only respond in the language of the assessment throughout the sessions. Procedure Assessment sessions were conducted one on one at the school sites and lasted approximately 45 minutes. During the assessment session, children were allowed to discontinue the testing situation at any time. Children were assessed twice, once in English a nd once in Spanish, at three time points: in the spring of 2002 Time 1 as they exit ed their preschool programs, in the spring of 2003 as the y completed kindergarten Time 2 and in the spring of 2004 as they completed first grade Time 3 For the E CS sample, there were two teams of assessors, one for each language. The assessors received extensive training on administering the assessment battery. Prior to assessing a child, the assessor spent some time in the classroom getting to know the child. Assessors spoke only in the language of the assessment during both the warm up
37 sessions in the classroom and the assessment sessions. These procedures having separate language teams and using only the language of the assessment were used to minimize c ode switching during testing sessions. Data Analysi s Some descriptive statistics were c alculat ed as preliminary analyses to this study. For example, t he means and standard deviations for each of the measures are reported for each grade level. Data analyses follow the same procedures for the Spanish measures. Therefore, analyses of the Engl ish and Spanish measures were conducted separately for each language. The main fo cus of this study wa s to determine which early literacy and oral language s kills are most importa nt for predicting studentsÂš reading abilities as indexed by the book task A simultaneous multiple linear regression was used to calculate R which is often used as a measure that quantifies model fit. R may also be interpreted as the proportion of the variance in Y that can be accounted for by p predictors. A simultaneous multiple regression was chosen because there was an interest in exploring the importance of all the predictor variables across grade levels. The researcher proceed ed by check ing whether the assumptions of the linear regression model we re met. N ormality of the variable s were ex amined through the use of a stan dardized residual plot and calculation of skewness and kurtosis The presence of outliers or influent ial data points were evaluated to investigate whether any values affected either the regression equation or a s ingle predictor Stevens, 2002 CookÂšs distance was used to find influential data points The intercorrelations between the predictor varia bles were also examined. Ideally, the research er was looking for each of the predictors early language and literacy
38 variables to be highly correlated with the dependent variable book task and for low correlations between each of the predictors Colli nearity was explored through the calculation of tolerance statistics and variance inflation factors VIF. T here wa s an interest in comparing the contributions made by the individual early language and literacy variable s to determine which ones ar e o f greatest importance for ELLs while reading To determine the importance of the predictors, s tandardized beta weights were examined. In a study by Baltes, Parker, Young, Huff, and Altmann 2004 the investigators did not find other importance indices to be superior tha n traditional standardized beta weights. This mu ltiple regression procedure was first conduc ted with the maximum number of participants in each language. Then the same procedure was completed with the participants who had data in both lan guages. Obtaining models between languages with an equal number of participants allowed for comparison across English and Spanish. F inally, a path analysis was conducted in each language to answer the third research question. This statistical pro cedure allow ed the researcher to investigate both the direct and indirect effects of the variables for predict ing word reading. Such a procedure allow ed the researcher to observe effects that may not have been highlighted using the previously mentioned st atistical methods. The path analyse s also allowed the researcher to observe whether the importance of the predictors changed across grades.
39 Chapter 4 Results Overview E arly literacy measures were administered to a group of pre school students each year until they reached first grade. These students came from homes where one of the languages spoken was Spanish. Therefore, each of the measures were administered in both English and Spanish. In addition to the five early l iteracy measures, each year the students were also assessed on a book task. Students were presented with a storybook and were assessed on how f amiliar they were with books, whether they comprehended when an examiner read to them, and whether they could re ad on their own. This measure was also ad ministered in both languages. Multiple l inear regression model s were utilized to explore the ability of the early literacy measures to predict student performance on the book task for each of the three years of as sessment Finally, path analyses were conducted in order to examine the direct and indirect effects of the measures throughout the years. Descriptive Statistics The means and standard deviations of each of the measures are listed in Tab le 2 accordin g to grade level. As reported in Paez et al. 2007, w ith the exception of the pho nological awareness measure in pre k indergarten, the participants generally scored higher on the English measures. The student s Âš poorest performance s in Spanish were
40 observ ed in the picture vocabulary and memory for sentences measures. In each of these measures, students Âš pre kindergarten scor es in English surpassed scores obtained in Spanish during first grade. In other words, during first grade, students did not reach th e performance levels in Spanish that they had obtained in English as early as pre kindergarten. For further discussion regardin g the participant s Âš performance on these measures and how they compared to their monolingual peers see Paez et al. 2007 and P aez and Rinaldi 2006. In thi s study, the researcher reports findings from the book task which had been assessed by the ECS research group, but never explored. What follows is therefore a summary of the descriptive statistics Table 2 and reliabil ities Table 3 of the book task. Th e measure was designed to be identical between the languages. With an average score of about nine in English and seven and a half in Spanish, students were at least making it to the second section of the book task duri ng pre kindergarten. This means that students knew how to complete basic tasks such as identifying the front of a book, point to the authorÂš s name and open the book. Mean scores for the book task in each language demonstrated developmental progression, w ith scores increasing across grade level. In kindergarten, t he gap between what students could do in En glish compared to Spanish began to widen. Students tended to comprehend the story best and answer questions involving comprehension when assessed in English during Kindergarten. Finally, students were likely to be able to read at least part of the storybook in English. In contrast, they struggled to simply be able to retell the story when being assessed in Spanish. It is apparent that although thes e students speak Spanish at home, the absence
41 of academic instruction in Spanish results in quite limited academic skills in that language. Table 2 Means and standard deviations of each measure by grade level Measure Mean SD ES n Picture Voca bulary Pre K 1 6.34 13.84 4.75 4.14 .56 265 264 Kinder 1 9.69 14.66 3.68 4.11 1.29 260 257 Grade 1 2 2.77 15.87 3.89 4.45 1.65 260 257 Letter Word Identification Pre K 6.85 5.05 3.44 1.90 .65 265 264 Kinde r 13.32 8.42 4.34 6.61 .88 264 257 Grade 1 22 .67 15.27 7.30 12.55 .72 260 257 Dictation Pre K 6.96 6.08 2.15 1.91 .43 265 264 Kinder 11.47 9.42 2.43 3.63 .67 264 257 Grade 1 1 5.26 12.39 3.46 5.3 9 .63 260 257 Memory for Sentences Pre K 2 9.33 24.55 5.19 6.12 .84 261 258 Kinder 3 1.24 25.05 5.53 6.70 1.01 264 257 Grade 1 35.22 27.73 5.00 5.51 1.42 260 257 Phonological Awareness Pre K 7. 05 7.56 5.08 4.13 .11 264 261 Kinder 1 6.77 13.72 5.34 4.34 .63 265 257 Grade 1 20.98 17.17 3.77 4.08 .97 262 260 Book Task Pre K 9.01 7.50 5.66 5.19 .28 262 256 Kinder 17.41 13.15 4.30 5.17 .90 264 257 Grade 1 22.19 16.79 3.32 5.87 1.13 258 257 Note. All values are raw scores. Values enclosed in parentheses represent the Spanish measures.
42 The normality of the variables was examin ed by computing skewness and kurtosis values. A couple of the predictor variables, such as letter word identification in Spanish raised some concerns with skewness levels higher than the absolute value of 1. Attempts were made to normalize the data by computing log transformations, but such a proce dure did not remedy the non normal distributions. As a result, the data were analyzed with the original student scores. To asse ss whether there was heterogeneity of variance, standardized residual plots were generated from the multiple regressions No ne of the plots raised concern except for the English book task in first grade. Visual inspection revealed the possibility of a model violation during this year. For lower predicted sc ores on the book task, there was considerable variability of the resid uals Then for higher predicted scores on the book task, there wa s less variability and therefore more accurate predictability with the model. Although these possible violations were found, the researcher proceeded with the multiple regression analyses b ecause the procedure is robust against violations of normality and variance. Nevertheless, some caution should be taken with the interpretation of the results. Reliability In order to obtain a measure of reliability, CronbachÂšs alpha was calculated for each grade level the book task was administered. An overall alpha coefficient was also calculated for all three grade levels. A summary of these findings can be found in Tab le 3 Adequate reliability was obtained in each grade level except during first g rade in English = .64. In general, the Spanish book task demonstrated higher reliabilities each year excluding pre kindergarten t han the English book task. The highest internal
43 consistency reliability was found when CronbachÂšs alpha was calculated throughout the grades. Table 3 Reliability of Book Task Grade CronbachÂšs alpha n Pre kindergarten 02 .86 .84 262 259 Kindergarten 03 .75 .78 263 257 First Grade 04 .64 .81 263 264 Overall 02 04 .88 .91 256 245 Note. Values enclosed in parentheses represent the Spanish measures. Correlation Analyses Initial ly, PearsonÂšs c orrelations between each of the early language and literacy measure s were calculated for each grade level. As illustrated in Table 4 the English measures showed higher correlations between each other than the Spanish measures during pre kindergarten During kindergarten, a couple of the Spanish measures demonstrated hi gher correlations than the same measures in English. For example, dictation scores were more correlated with both the picture vocabulary and letter word identification measures when assessed in Spanish. In first grad e, higher correlations were evident be tween the Spanish measures than between the English measures. In sum, higher cor relations between the measures we re first seen in English during pre kindergarten. Then the Spanish measures demonstrate d higher correlations as students reach ed first grade.
44 Table 4 Correlations Between the Early Literacy Measures in Pre kindergarten N = 247 238 Measure PV LW D MS PA PV 1 .49 .42 .66 .38 LW .27 1 .74 .43 .46 D .30 .60 1 .45 .49 MS .62 .24 .39 1 .47 PA .22 .34 .39 .30 1 Correlations Be tween the Early Literacy Measures in Kindergarten N = 247 238 Measure PV LW D MS PA PV 1 .50 .44 .63 .42 LW .45 1 .69 .39 .52 D .55 .80 1 .38 .54 MS .59 .31 .36 1 .44 PA .23 .31 .30 .24 1 Correlations Between the Early Literacy Measures in First Grade N = 247 238 Measure PV LW D MS PA PV 1 .53 .55 .51 .42 LW .54 1 .75 .38 .42 D .60 .86 1 .37 .43 MS .61 .48 .54 1 .26 PA .27 .36 .40 .27 1 Note. Shaded value s represent the Spanish measures. PV = Picture Vocabulary; LW = Letter Word Identification; D = Dictation; MS = Memory for Sentences; PA = Phonological Awareness All correlations were significant at p < .001.
45 A second s et of correlations explored the relationship of each predictive measure across grade le vel Table 5 For example, the English picture vocabulary score at pre kinder garten was significantly correlated to the English picture vocabulary score at kindergarten r = .71. Overall, higher correlations were seen for grades closest in time. I t w as more likely for a predictive measure to demonstrate a higher correlation from pre kindergarten to kindergarten than from pre kindergarten to first grade. An exception occurred with memory for sentences in Spanish where the correlation from pre kinderga rten to first grade r = .66 was greater than the correlations from kindergarten to first grade r = .55. Table 5 Correlations between the predictive measures across grade level N = 247 238 Pre K to Kinder Pre K to 1st grade Kinder to 1 st grade PV .71 .73 .68 .66 .72 .79 LW .58 .43 .46 .30 .70 .70 D .49 .52 .42 .47 .55 .81 MS .68 .66 .56 .66 .57 .55 PA .50 .27 .33 .24 .56 .47 Note. Values enclosed in parentheses represent the Spanish measures. PV = Picture Vocabulary; LW = Letter Word Identification; D = Dictation; MS = Memory for Sentences; PA = Phonological Awareness All correlations were significant at p < .001. Simple correlations were also calculated each year be tween the predictive measures and the book task Table 6 Picture vocabulary showed a higher correlation with book task in English during pre kindergarten. However, higher correlations were seen in Spanish during kindergarten and first grade. An intere sting phenomenon was observed for the remaining measures. During pre kindergarten and kindergarten the remaining measures demonstrated higher correlations in English. Ye t, once students reached first grade the correlations were higher in Spanish between letter word
46 identification, dictation, memory for sentences, and phonological awareness with the book task. Table 6 Correlations Between Predictive Measures and Book Task N = 247 238 Pre K Kinder First Grade PV .67 .50 .62 .70 .51 .67 LW .48 .34 .60 .54 .53 .56 D .50 .43 .58 .57 .50 .61 MS .62 .48 .63 .55 .45 .60 PA .41 .33 .56 .42 .43 .47 Note. Values enclosed in parentheses represent the Spani sh measures. PV = Picture Vocabulary; LW = Letter Word Identification; D = Dictation; MS = Memory for Sentences; PA = Phonological Awareness All correlations were significant at p < .001. Regression Analyses Assumptions Multiple regression analys es are based on seve ral assumptions and the data were examined in order to justify the use of such procedures The assumptions that errors are independent, normally distributed, and with constant variance were explored with residual plots Stevens, 1999. A graphical display of the residual s against predicted values were created for each grade level and language. No mode l violations u were indicated based on observation of the plots. The presence of influential data points were surveyed by calculating st udentized residuals and CookÂšs distance values. The maximum values found included a studentized residual value of 3. 05 and a CookÂšs D values of 0.14 As a result, no alarms were raised in regards to the poss ibility of any case having an undo influence on the regression analyses Collinearity, the undesirable circumstance where high correlations exist between the independent variables, was examined. Tolerance statistics were calculated for each of
47 the models. This value indicates the proportion of variance that is not accounted for by the other variables in the model Kinnear & Gray, 2006. Another measure of collinearity that was computed was the variance inflation factor VIF Neither one of these collinearity gauges revealed that intercorr elations among the predictors were problematic. Multiple Regression S imultaneous m ultiple linea r regression analyse s were used to develop model s for predicting students Âš book task scores from their early language and literacy scores e.g., picture vocab ulary, letter word identification, dictation, memory for sentences, a nd phonological awareness. The s e a nalyses were conducted each year pre kindergarten, kindergarten, and 1 st grade and in each language English and Spanish. In addition, multiple reg ressions were analyzed again for participants who had no missing data in either language. This latter procedure allowed the researcher to make direct comparisons across the lang uages since the same participants were used in each analysis N evertheless, t he data were analyzed by language in order to make observations wi th the maximum data points availa ble for each language. In each of the following tables Tables 7a Â€ 9a multiple regressions with the maximum number of NÂšs will be found above the multipl e regressions with the same sample size Tables 7b 9b
48 Table 7a Predicting Book Task Scores in Pre kindergarten N = 247 238 Predictor B Picture Vocabulary .49 .37 40 *** .29 *** Letter Word Identifi cation .10 .10 .06 .04 Dictation .41 .59 .16 .22 ** Memory for Sentences .25 .15 .24 *** .17 Phonological Awareness .06 .16 .05 .13 Note. R = .54*** .37*** Raw scores were used for calculations. Values enclosed in parentheses represent the Spanish measures. p <.05 ** p < .01 *** p < .001. Table 7b Predicting Book Task Scores in Pre kindergarten N = 224 Predictor B Picture Vo cabulary .48 .38 .40*** .29*** Letter Word Ident ification .12 .11 .07 .04 Dictation .36 .59 .14* .21 ** Memory for Sentences .28 .16 .26 *** .19 ** Phonological Awareness .06 .17 .06 .13 Note. R = .56*** .38*** Raw scores were used for calculations. Values enclosed in parentheses r epresent the Spanish measures. p <.05 ** p < .01 *** p < .001. During pre kindergarten Table 7a the variables found to significantly predict the book task in English included picture vocabulary, dictation, and memory for sentences = .40, .16, and .24 respectively The five early literacy and language variables produced an R of .54 F 5, 241 = 57.02, p < .01 for the prediction of the English book task. In Spanish, these predictor variables were also found to significantly pr edict the book task. Plus, phonological awareness also significantly contributed to the predictio n of the book task in Spanish. An R of .37 F 5, 232 = 27.65, p < .01 was obtained for the prediction of the book task in Spanish.
49 When the data w e re analyzed with the same number of participants no drastic changes were observed Table 7b Nonetheless, m emory for sentences d e monstrated an increase in significance in predicting the Spanish book task and the R for both the English and Spanish model s slightly increased. The En glish model produced an R value of .56 F 5, 218 = 54.67, p < .01 Finally, the comparable Spanish model produced an adjusted R value of .38 F 5, 218 = 26.97, p < .01 In both models and both languages, picture vocabul ary produced the highest standardized beta coefficient. These finding s are aligned with the aforementioned correlation analyses that identified picture vocabulary as having the highest correlation with the book task in each language during pre kindergarte n CohenÂšs 1992 effect size f = R/ 1 R was computed for the English model in pre kindergarten d = 1.27 This value can be interpreted as indicating a large effect size using CohenÂšs rough guidelines .02 small, .15 medium, .35 large. For the Spanish model, a large ef fect size of 0 .61 was also obtained Effect sizes will only be repo rted for the models with 224 participants which allows for comparison across languages. Table 8a Predicting Book Task Scor es in Kindergarten N = 247 238 Predictor B Picture Vocabulary .22 .57 .20*** .44 *** Letter Wor d Identification 2 3 .14 .22*** .19 ** Dictation .26 .07 15 * .05 Memory f or Sentences .25 .15 .31*** .18 ** Phonological Awareness .11 .25 .15** .20 ** Note. R = .62*** .62*** Raw scores were used for calculations. Values enclosed in parentheses r epresent the Spanish measures. p <.05 ** p < .01 *** p < .001.
50 Table 8b Predictin g Book Task Scores in K indergarten N = 224 Predictor B Picture Vocabulary .23 .58 .21*** .45*** Letter Word Identification .21 .14 .21** .19** Dictation .25 .07 .15* .05 Memory fo r Sentences .26 .13 .32*** .16** Phonolo gical Awareness .11 .23 .14* .19** Note. R = .62*** .61*** Raw scores were used for calculations. Values enclosed in parentheses r epresent the Spanish measures. p <.05 ** p < .01 *** p < .001. During kindergarten, all of the predictor variables significantly contributed to the prediction of the book task in English Table 8a The gre atest predictors were memory for sentences = .31, letter word identification = .22 and picture vocabulary = .20. At the kindergarten level, phonological awareness was found to significantly contribute to the prediction of the book task in English = .15. The predictor variables pro duced an R value of .62, F 5, 241 = 76.33, p < .01 for the prediction of the English book task. In Spanish, the only predictor that was not found to significantly contribute to the prediction of the book task was dictation. The variable with the great est predictive power in Spanish turned out t o be picture vocabulary = .44 The Spanish early literacy and language variables yielded an R value of .62, F 5, 232 = 74.26, p < .01 When participants were matched in English and Spanish to create the model some decreases were seen in the signi ficance of the predictive ability of letter word identification, dictation and phonological awareness in English Table 8b The R value for the comparable model was also .62, F 5, 218 = 69.67, p < .01 No changes in significance levels were observed in Spanish when the comparable participant model was utilized. The R value i n Spanish turned out to be .61, F 5, 218 = 69.07, p < .01 R
51 values for all the models analyzed in kindergarten were at about 62 indicating more similarity between the lang uages than in the other grades Effect sizes in kindergarten were 1.63 in English and 1.56 in Spanish. The effect sizes were again large and quite similar between languages in kindergarten. Table 9a Predicting Book Task Scores in First Grade N = 247 238 Predictor B Pict ure Vocabulary .13 .47 .15* .37 *** Letter Word Identification .10 .01 .22** 03 Dictation .10 .17 .10 .17 Memor y for Sentence s .14 .23 .22*** .22 ** Phonological Awareness .17 .34 .19** .24 *** Note. R = .42*** .61*** Raw scores were used for calculations. Values enclosed in parentheses represent the Spanis h measures. p <.05 ** p < .01 *** p < .001. Table 9b Predictin g Book Task Scores in First Grade N = 224 Predictor B Pict ure Vocabulary .11 .49 .14 .39*** Letter Word Identification .10 .02 .23** .04 Dictation .08 .14 .09 .13 Memor y for Sen tences .15 .24 .21** .23 ** Phonolo gical Awareness .18 .34 .20** .24*** Note. R = .43*** .61*** Raw scores were used for calculations. k Values enclosed in parentheses represent the S panish measures. p <.05 ** p < .01 *** p < .001. The contributions of greatest influence during first grade Table 9a were seen with memory for sentences = .22, letter word identification = .22 and phonological awareness = .19. Dictation no longer contributed to the prediction of the book task during first grade. An R of .42, F 5, 241 = 35.33, p < .01 was observed for the prediction of the E nglish book task
52 In Spanish, letter word identification and dictation made no significant contributions to the prediction of the book task scores. Picture vocabulary = .37, phonological awareness = .24 and memory for sentences = .22 were a l l significant predictors of the book task p < .001 An R value of .61, F 5, 232 = 71.17, p < .01 was generat ed for the prediction model in Spanish during first grade. Once the comparative model was analyzed, picture vocabulary was observed to join dictation in not contributing to the prediction of the book task in English Table 9b The R levels also remained about the same with these models. In English an R of .42, F 5, 218 = 32.21, p < .01 was generated. In Spanish, no great changes were o bserved. Picture vocabulary contributed a bit more to the prediction of the Spanish book task = .49. There was no change in R again producing a value of .61, F 5, 218 = 67.77, p < .01 when the groups were compared using the same number of stude nts. During first grade, the effect size of the model in English was 1.38. For the first time the Spanish model surpassed the English model with an effect size of 1.56. Path Analysis The following figure Figure 1 represents the path model that wa s tested in both English and Spanish with AMOS version 7.0 Paths were drawn from the predictor to the outcome variables for each grade level. Table 10 pr esents the results of the model, specifically unstandardized and standardized estimates, and their s ignificance The indirect effects were tested, but none of the effects were found to be statistically significant. In order to determine the appropriateness of the proposed model, the researcher reviewed the fit indices for each language.
53 For the Engli sh language model, indices were mixed making it difficult to firmly state whether the model was a good fit. This wa s evidenced by the Chi square statistic, 92, 267 = 319.40, p < .0 01 which indicated statis tically significant lack of fit In addition, the root mean square error o f approximation RMSEA = .10 wa s above the recomme nded cut off of .08 or less as an indication of an adequate fit. However, t he comparative fit index CFI = .93 produced a va lue close to one which indicated a good fit. Although no conclusions were made regarding the degree of fit the model generated, the researcher believed it was important to highlight some of the notable fin dings. A similar pattern for the fit indices was observed for the Spanish mode l Nevertheless, the fit of the Spanish model was not as good as that of the English model. The Chi square statistic, 98, 267 = 486.40, p < .0 01 also revealed statistical significance for the Spanish model. The root mean square error of approximation RMSEA = .12 was higher for th e Spanish model, indicating less than adequate fit. T he compa rative fit index CFI = .88 wa s below the widely used cutoff of .95 Taken together, these values would indicate less than acceptable fit
54 Picture Vocabulary Letter Word Identification Dictation Memory for Sentences Phonological Awareness Book Task Picture Vocabulary Letter Word Identification Dictation Memory for Sentences Phonological Awareness Book Task Picture Vocabulary Letter Word Identification Dictation Memory for Sentences Phonological Awareness Book Task Pre Âƒ K K 1 st Grade Figure 1 This figure represents the path model chosen for the path ana lysis. Thicker arrows indicate direct paths from independent to dependent variables and vice versa. Th inner arrows indicate predictor variables across grade level. English path a nalysis T he final path model for the English language test scores is display ed in Figur e 2 The pathway coefficients are represented as sta ndardized beta weights to allow for comparison s with beta weights obtained with the multiple regression analyses.
55 Picture Vocabulary Letter Word Identification Dictation Memory for Sentences Phonological Awareness Book Task R Â¤ = .55 Picture Vocabulary R Â¤ = .39 Letter Word Identification R Â¤ = .21 Dictation R Â¤ = .16 Memory for Sentences R Â¤ = .33 Phonological Awareness R Â¤ = .30 Book Task R Â¤ = .58 Picture Vocabulary R Â¤ = .38 Letter Word Identification R Â¤ = .30 Dictation R Â¤ = .15 Memory for Sentences R Â¤ = .25 Phonological Awareness R Â¤ = .29 Book Task R Â¤ = .36 Pre Âƒ K K 1 st grade English .42*** .03 .19** .22*** .05 .15* .22*** .16** .32*** .15** .11* .14* .10 .16** .12* .63*** .61*** .29*** .11* .45*** .40*** .58*** .24* .55** .39*** .50*** .39*** Figure 2 Pathway coefficients are represented as standardized beta weights. Thicker arrows indicate direct path from independent to dependent variables. Thinner arrows indicate pred ictor variables across grade level. p <.05 ** p < .01 *** p < .001. Predicting pre kindergarten book task s cores The ear ly language and literacy skills that predicted studentsÂš book task scores in pre kindergarten were picture vocabulary = .42, memory for sentences = .22, and dictation = .19 Each of the aforementioned predictors were signif icant at the .01 lev el Although the beta weights ranged from small to moderate, they were all positive. This indicates that higher scores on these three early language and literacy skills were associated with higher scores on the English book task. The model accounted for 55% of the variance in studentsÂš book task scores in pre kindergarten.
56 Predicting kindergarten book task s core s. All five predictors were statistically significant p < .05 during kindergarten. Memory for sen tences = .32 was the predictor that produced the greatest beta weight in comparison to the other measures. Letter word identification = .22 and phonological awareness = .15, predictor s that did not have significant association s with the b ook t ask at pre kindergarten were now significant in kindergarten. Picture vocabulary = .15 and dictation = .16 remained significant predictors of the book task at kindergarten. The model accounted for 58% of the var iance in the studentÂšs book task s cores in kindergarten. Predicting first grade book task s cores The set of first grade predictors, as well as scores on the book task in first g rade, reflect ed a change in scores from pre kindergarten through first grade. All of the predictors, except fo r dictation = .10, were statistically sign ificant p < .05 The full unconstrained model accounted for 36% of the variance in the final criterion variable studentÂšs book task scores in first grade Table 10 Standardized and Unstandardized Path Coefficients Pa th Name Unstandardized Path Coefficient Standardized Path Coefficient Sig. Test To Pre K Book Task Picture Vocabulary 50 .34 .42 .27 y y Letter Word Identification .05 .08 .03 .03 n n Dictation 50 .55 .19 .21 y y Memory for Sentences 24 .17 .22 20 y y Phonological Awareness .06 .12 .05 .10 n n To Kindergarten Book Task Picture Vocabulary .17 .46 .15 .37 y y Letter Word Identification .22 .16 .22 .21 y y Dictation .27 .03 .16 .02 y n Memory for Sentences .24 .14 .32 .20 y y Phonological Awareness .11 .17 .15 .16 y y To 1 st grade Book Task Picture Vocabulary .09 .32 .11 .25 y y Letter Word Identification .06 .04 .14 .10 y n Dictation .09 .05 .10 .05 n n Memory for Sentences .10 .16 .16 17 y y Phonological Awareness .10 .20 .12 .16 y y Pre K to Kindergarten Book Task .08 .22 .11 .24 y y Kindergarten to 1 st grade Book Task .22 .38 .29 .36 y y
57 Spanish path a nalysis The final path model for the Spanish early language and literacy measures is displayed in Figure 3 Picture Vocabulary Letter Word Identification Dictation Memory for Sentences Phonological Awareness Book Task R Â¤ = .35 Picture Vocabulary R Â¤ = .37 Letter Word Identification R Â¤ = .07 Dictation R Â¤ = .09 Memory for Sentences R Â¤ = .31 Phonological Awareness R Â¤ = .06 Book Task R Â¤ = .58 Picture Vocabulary R Â¤ = .46 Letter Word Identification R Â¤ = .35 Dictation R Â¤ = .46 Memory for Sentences R Â¤ = .24 Phonological Awareness R Â¤ = .18 Book Task R Â¤ = .48 Pre Âƒ K K 1 st Grade .61*** .26*** Spanish .3*** .56*** .25*** .10 .20** .21** .03 .27*** .24*** .36*** .37*** .21*** .02 .20*** .16*** .67*** .59*** .68*** .49*** .43*** .16*** .17*** .05 .10 .25*** Figure 3 Pathway coefficients are represented as standardized beta wei ghts. Thicker arrows indicate direct path from independent to dependent variables. Thinner arrows indicate pre dictor variables across grade level. p <.05 ** p < .01 *** p < .001. Predicting pre kindergarten book task s cores Interestingly, the same pred ictors that were statistically significant in English during pre kindergarten resulted as signif icant predictors in Spanish as well. Picture vocabulary = .27, dictation = .21, and memory for sentences = .20 scores were all statistically signific ant at a level of .01. Letter word identification and phonological awareness were not statist ically
58 significant predictors of the Spanish book tas k. Th e model accounted for 35% of the variance in studentsÂš scores during pre kindergarten. Predicting kindergarten book task s core s. All the predictors, excluding dictation, were significant predict ors of the book task in kindergarten. S tudentsÂš picture vocabulary = .37, letter word identification = .21, memory for sentences = .20, and phonological awareness = .16 scores significantly predicted book task scores at the .001 level. Th e model accounted for 58% of the variance in change in studentsÂš boo k task scores from pre kindergarten to kindergarten. Predicting first grade book task s cores As with the English language analyses, the set of first grade predictors and the book task reflect ed scores from pre kindergarten through first grade. In this grade level, letter word identification no longer made a statistically significant contribution to predicting book task scores. That is, studentsÂš picture vocabulary = .25, memory for sentences = .17, and phonological awareness = .16 skills were all significantly p < .001 related to studentsÂš book task scores in first grade. Th e model accounted for 48% of the variance in change in studentsÂš book task scor es from pre kindergarten to kindergarten to first grade.
59 Chapter 5 Discussion The main purpose of this study was to explore how well early language and literacy skills predict the reading abilities of a group of bilingual students. A number o f analyses were run in order to explore thr ee main questions in each of the languages spoken by the students First, the researcher examined the most impor tant skills for predicting reading abilities This was done in each of the grade levels assessed in order to inv estigate whether there were differences across the grade levels. Finally, the direct and indirect effects of these predictors across grade levels were explored Throughout this chapter, a summary of the findings along with implications f or r esearch and practice are discussed. Summary of Findings The res earcher began to explore t he first research question by analyzing the simple correlations between the predictor s and the book task. Throughout the grades, it was observed that picture vocabu lary and memory for sentences had some of the highest correlations with the book task in both languages. Picture vocabulary had a correlation of .67 in English during pre kindergarten and then the same correlation of .67 was seen in Spanish during first g rade. Such results are not surprising considering the abundance of studies that emphasize the strong relationship that exists between vocabulary and reading skills e.g. Lonigan, Burgess, & Anthony, 2000 ; Wagner, Torgesen, Rashotte, Hecht,
60 Baker, Burgess et al., 1997. Memory for sentences had correlations of .62 in pre kindergarten and .63 in kindergarten with the English book task. A similar correlation of .60 was seen with memory for sentences during first grade. In sum, after calculating some bas ic bivariate correla tions, it was apparent that picture vocabulary and memory for sentences would likely result in important predictors for the book task. These initial findings were confirmed when the multiple regression and path analyses were conducte d. Picture vocabulary was one of the most important predictors of the book task in both languages throughout the years An exception occurred in English during first grade when picture vocabulary was not found to be significant with the multiple regressi on analyses During first grade, letter word identification, memory for sentences, and phonological awareness were among the most dominant predictors of the book task in English Yet, the path analysis conducted in English found picture vocabulary to rem ain a significant predictor of the book task in first grade. Memory for sentences remained a significant predictor in both languages throughout each g rade level Paez and Rinaldi 2006 analyzed the same data set but focused mainly on how kindergarten or al language skills e.g., picture vocabulary, memory for sentences, phonological awareness predicted first grade word reading abilities as indicated by the letter word identi fication measure. Their analyse s revealed vocabulary and phonological awareness to be the best predictors of word reading abilities. Although different outcome measures were utilized by Paez and Rinaldi 2006, findings were rather parallel with this study This study extended their research by including early literacy variables in the prediction of reading ability e.g. letter word identification, dictation In addition, the
61 outcome variable in this present study assessed not only word reading abilities, but also comprehension abilities. The bivariate correlations displa yed in Table 4 show that picture vocabulary and memor y for sentences demonstrate d one of the highest correlations between the predictor variables. Taking in to consideration the findings reported by Proctor et al. 2005 that the strength of student s Âš voca bulary i s greatly associated with student Âšs ability to comprehend both spoken and written language, the reason for the high correlation becomes apparent. If a student does not possess the vocabulary dictated during the memory for sentences measure, that s tudent will strugg le to comprehend and have difficulties repeat ing the sentence back to the examiner. The effects of this finding as they relate to results obtained in the book task will be discussed subsequently. In pre kindergarten, dictation was found to significantly contribute in both the multiple regression and path analyses, to the prediction of the book task in b oth English and Spanish Yet, letter word recognition was not found to be significant. This finding is quite surprising considering the extensive amount of literature linking letter word recognition skills as an important predictor of reading abilities Lyon & Chhabra, 1996; Pullen & Justice, 2003; Whitehurst & Lonigan, 1998. T he researcher was aware that a side from the intercorrelations found with picture vocabulary and memory for sentenc es, there were also high correlations between letter word identification and dictation. It makes sense that if a student is able to identify a letter or a word, they are just as likely to be able to wri te that letter or word down for the examiner when it is dictated to them. As a result, t he researcher suspected that the correlation dictation demonstrate d with let ter
62 word i dentification impacted the predictors Âš ability to appear as significant in the pr ediction model. Phonological awareness was found to be a significant predictor of the book task in Spanish in the multiple regression analysis. Significance was not maintained once the path analysis was conducted p = .09. Students scored higher on the phonological awareness measure in Spanish during pre kinde rgarten, but it seems that students did not score high enough to produce significant differences with the path analysis. The results in English are not surprising considering that children do not develop good phonological awareness skills until about six years of age when they have received explicit instruction in reading Goswami, 2005. Plus, a s pr eviously no ted by Gerber et al. 2004, by definition, ELLs have had limited exposure to import ant pre requisite skills such as phonology. Such limited phonological awareness in English and greater skills in Spanish implied that the majority of children had a greater understanding of Spanish until school enrollment in an English only environment When one observes how students fell further behind Spanish norms as they progressed through school, it became evident that s tudents had to abandon their Spanish at the expense of learning English. During kindergarten, some additional interesting find ings were observed. Both the multiple regression and path analyses indicated that dictation was a significant predicto r of the book task in English. Yet i n Spanish dictation was not found to be a significant predictor of the book task. Such a finding m ay be a result of the students increased skills in phonological awareness in English. As noted earlier with the findings of MacDonald and Cornwall 1995, the best predictor in kindergarten of future spelling abilities is phonological awareness. In both languages and with both methods of
63 analyses, letter word recognition turned out to be a significant predictor of the book task during kindergarten. It is perhaps likely that after phonological awareness and letter word identification skills have been util ized in a model to predict reading abilities, a measure of writing and spelling may have little to add to the prediction model. Finally, in first grade, letter word identification was found to be a significant predictor of the English book task us ing both methods of analyses. Pict ure vocabulary and dictation were not found to be statistically significant with the multiple regression analysis. Only dictation was not found to be significant when the path analysis was used. The results of both meth ods of analyses i n Spanish indicated that both letter word identification and dictation did not significantly predict the book task. In conclusion, the most important variables observed to predict the reading abilities of bilingual students from pre kinde rgarten through first grade include vocabulary, letter word recognition, listening comprehension, and phonological awareness. These findings along with the descriptive statistics reported for the book task in C hapter IV can be integrat ed to produce a greater understanding of the skills these bilingual students posses s In Chapter IV i t was noted that as early as kindergarten, the gap between what students could do in English compared to Spanish began to widen. At kindergarten, studen ts tended to comprehend the story best and answer questions involving comprehension when assessed in English. The phenomenon reported in the previous paragraph is likely a result of student s Âš limited vocabulary in Spanish. Proctor et al. 2005 discusse d the importance of having a strong vocabulary in order to be able to comprehend both spoken and written language Paez et al. 2007 reported that students in the ECS study fell about two standard
64 deviations below their monolingual peers in both picture vocabulary and me mory for sentences. Memory for sentences was a significant predictor of the book task each year, in each language, and with each method of analysis. Such findings place an emphasis on the importance of listening comprehension in predicti ng reading ability. Further support for the simple view of reading Hoover & Gough, 1990 was established with the results of this study. To recap, the theory states that a student with good decoding and listening comprehension skills is likely to be a g ood reader. In both kindergarten and first grade, the analyses in English indicated that decoding and listening comprehension skills were the two most important predictors of reading abilities. Therefore, support was added to Hoover and GoughÂšs 1990 cl aims that good decoding and listening comprehension skills lead to more skilled reading in English. On the contrary, the same results were not obtained in Spanish. Vocabulary was observed to make the most important contributions in Spanish from pre kinder garten throughout first grade. Decoding skills were not observed to be significant predictors of the Spanish book task in first grade. Nevertheless, listening comprehension skills maintained their importance throughout each of the grades in Spanish. The se results are likely to have been affected by the low scores obtained in Spanish by the sample studied. Further discussion regardi ng the effect of low scores on the prediction model can be found in the limitations section. The similar findings of thi s study with those of the simple view of reading theory highlight the importance of having included early literacy measures such as letter word recognition in the prediction models of readings abilities in this study Oral language measures alone may unde restimate the reading abilities of bilingual students. Paez and
65 Rinaldi 2006 reported that English oral language skills lagged significantly behind monolingual norms. In contrast, letter word recognition skills were developing at a normal rate. In ess ence, a lthough t he importance of oral language abilities has alr eady been established, these abilities should not be the only indicators of reading abilities This is especially important for bilingual students whose performance on early literacy measures has been observed to surpass their oral language performance Paez et al., 2007. Implications for Research The researcher did not want to place restrictions on the variables entered into the model by using a stepwise o r hierarchical model. Therefore, a simultaneous multiple regression procedure was chosen in order to utilize all the predictor variables in each grade level and allow for comparisons. Such a procedure may have limited a variable from appearing as s t atis tical ly significant because of its correlation with another variable in the model. Dickinson McCabe, Anastasopoulous, Peisner Feinberg and Poe 2003 analyzed their data utilizing a hierarchical method and found that vocabulary and phonological awarenes s contributed equally when pr edicting print knowledge among three and four year olds. Both the multiple regression and path analyses conducted in this study did not identify phonological awareness in English to be a significant predictor for the four year olds in the sample. As a result, the utilization of a methodological procedure such as hierarchical regression may indicate whether the findings of Dickinson et al. 2003 replicated with this group of students.
66 Considering the possible impact of int ercorrelations noted above between the predictor variables, i t seems reasonable to suggest remov ing the dictation measures from the model. R unning these same analyses conducted in this study with just the picture vocabulary, letter word identification me mory for sentences, and phonological awareness measures in the models would be interesting to explore Dictation did not result to be a significant predictor in first grade with any of the methods of analysis. Providing educators with four simple measure s that could be administered early and be most indicative of their success or failure in reading would be ideal. For that reaso n, the importance of those four measures should be evaluated in future research. Limitations Some precautions should be taken when interpreting the results of this study because several limitations were found. Perhap s the greatest limitation existed with the measure chosen as the dependent variable, the book task. Although the book task was developed by the ECS research gro up to parallel early literacy measures, there is a lack of informat ion regarding the measure. Besides the r eliability analysis calculated in this study, no other examinations of reliability and validity have been conducted with this measure. Another limitat ion of the book task is that there are no norms available with which to compare the performance of this group of bilingual students. A finding noted by Dickinson et al. 2003 was that language was a much stronger predictor when students achieved wit hin normal levels in phonological sensitivity than when students score d i n the low end of phonological sensitivity. It was also found that phonological sensitivity was a much stronger predictor when students displayed normal language skills than when thes e language skills were deficient. Such fin dings suggest
67 that when studentÂš s skills are developing within normal ranges, literacy is supported by both language and phonological sensitivity skills. Yet, when students experience difficulties in one language area e.g. phonological awareness, vocabulary, the impact of this deficiency may generalize into other skills very early in the literacy process. Students in this study were performing well below norms. It may therefore be possible that there was some loss in the pr edictive power of the v ariables examined in this study because of the low achieving group of students sampled Implications for Practice With the reauthorization of the Individuals with Disabilities Educational Act IDEA of 2004, the use of Response to Intervention RTI is permitted. One of the key components of RTI involves systematic screening and progress monitoring. Such procedures are implemented in order to catch students before they fail. Researchers such as Coleman, Buysse and Neitzel 2006 have emphasized in a research synthesis they conducted of the use of RTI that intervening in kindergarten, or ideally earlier, is a framework that can make an impact on at risk students. Research has pointed out speech and language del ays, and phonological processing deficits as being some of the precursors of learning disabilities Coleman, Buysse & Neitzel, 2006 Therefore, educators need to conduct universal screenings of key language and early literacy skills such as the ones hi ghlighted as important in this study, to identify these at risk students. A lack of information still exists regarding the identification of pre school students at risk for learning disabilities and age appropriate interventions. This study was an attem pt to better understand what areas bilingual students struggle with that are associated with reading. These areas highlighted as deficient as early a s pre kindergarten
68 can be target ed for intervention. Much emphasis has been placed on the importance of p honological awareness in predicting literacy e.g. Adams, 1990; Swan & Goswami, 1997 That importance of phonological awareness should not be downplayed but educators should make sure to balance the curriculum by building language skills such as vocabul ary. Bilingual students split their vocabulary between two languages and as a result score well below average on monolingual norms in assessments of vocabulary Cobo Lewis et al. 2002 ; Umbel et al., 1992 Their vocabulary skills should not be assessed in English only because such practices only demonstrate half their knowledge. As Dickinson et al. 2003 reported, when students perform within no rmal ranges in both phonological awareness and vocabulary their literacy skills increase in a strongly pred ictable way. However, when they demonstrate deficiencies in either vocabulary or phonological sensitivity skills, their overall literacy skills suffer. So it benefits educators to make every effort to build both skills. Directions for Future Research Since the researcher is unaware of any early language and literacy measure that explores the development of reading in both English and Spanish, more investigations of the validity of the measure are called for. If the book task does indeed turn out to be a validated tool for assessing early literacy skills, then it should be used with other sample s of bilingual students. Ideally, the same measures should be implemented with similar low income bilingual students who receive instruction in Spanis h. Finally, a group of monolingual peers should also be assessed utilizing the English measures in order to have a basis for comparison.
69 Perhaps the best use of resources should be utilized to investigate what practices in pre kindergarten and kinderga rten work best to foster rapid growth in the vocabulary skills of bilingual students. P lacing bilingual students in English only environments has not shown to build upon their already deficient vocabulary Paez et al., 2007 It is therefore imperative t o explore alternative s in the curriculum that most effectively enrich the vocabulary of bilingual students. Paez and Rinaldi 2006 demonstrated that Spanish word reading skills in kindergarten was a significant predictor of English word reading ability d uring first grade. Such findi ngs imply that building student s Âš literacy skills in Spanish may help them to better achieve in English literacy. These findings lend more support to CumminÂšs 1979 developmental interdependence hypothesis. Furthermore, if students receive instructional support in Spanish, the gap in skills with their monolingual peers may diminish As a result, stronger prediction models are likely to be obtained with scores within normal ranges. The study of cross language transfer may a lso be more appropriate when students receive instruction in Spanish and score within normal ranges. The majority of the students 83% in this study ca me from homes where their parents mainly communicate d with them in Spanish. These parents may have fe l t incapable of helping with their children Âšs literacy in English. Yet, multiple research findings indicate that the best results may be obtained if parents help educators in building the skills stu dents already possess in Spanish Manis et al., 20 04 The achievement gap reported in the review of literature highlighted the gravity of the reading problem educators face with ELLs. It is unlikely that the achievement levels demanded by NCLB will be attained unless resources such as parents and the c ommunity are also tapped.
70 Conclusion Only the tip of the iceberg has been explored with the research questions posed in this study. The reality is that much remains to be invest igated in terms of the impact of early language and literacy skills on b ilingual student s becoming proficient readers. The prominent variables found in this study across pre kindergarten through first grade include d vocabu lary, letter word recognition listening comprehension and phonological awareness. These are the same sk ills that have been demonstrated to be of most importance for monolingual students. Nonetheless, some uncertainty still remains in regards to whether bilingual students benefit from the same instructional supports that monolingual students benefit from or whether different instructional approaches may be needed.
71 References Adams, M. J. 1990. Beginning to read: Thinking and learning about print Cambridge MA: The MIT Press. August, D., Carlo, M., Dressler, C. & Snow, C. 2005. The critical role of vocabulary development for English Language Learners. Learning Disabilities Research & Practice, 20 50 57. Baker, L., Fernandez Fein, S., Scher, D., & William s, H. 1998. Home experiences related to development of word recognition. In J. L. Metsala & L. C. Ehri Eds., Word recognition in beginning literacy pp. 263 287. Hillsdale, NJ: Erlbaum. Baltes, B. B., Parker, C. P., Young, L. M., Huff J. W., & Altmann, R. 2004. The practical utility of importance measures in assessing the relative importance of work related perceptions and organizational characteristics on work related outcomes. Organizational Research Methods, 7 326 340. Biemiller, A. 2003. Oral comprehen sion sets the cei ling on reading comprehension. American Educator 27 23 44. Carlisle, J. F., Beeman, M., Davis, L. H., & Spharim, G. 1999. Relationship of metalinguistic capabilities and reading achievement for children who are becoming bilingu al. Applied Psycholinguistics, 20 459 478. Catts, H. W., Fey, M. E., Zhang, X. & Tomblin, J. B. 1 999. Language basis of reading and reading disabilities: Evidence from a longitudinal investigation. Scientific Studies of Reading 3 331 Â€ 361 Clay, M. M. 1991. Introducing a new storybook to young readers. The Reading Teacher, 45 264 273. Clay, M. M. 1979. The early detection of reading difficulties: A diagnostic survey with recovery procedures Portsmouth, NH: Heinemann. Cl ay, M. M. 1970. Research on language and reading in Pakeha and Polynesia n children. In D.K. Bracken & E. Malmquist Eds ., Improving reading ability a round the World Newark, Delaware: In ternational Reading Association
72 Clay, M. M. 196 6. Emergent Reading Behaviour Unpubl ished doctoral dissertation, University of Auckland Library Cisero, C. A., & Royer, J. M. 1995. The development of cross language transfer of phonological awareness. Contemporary Educational Psychology, 20 275 303 Cobo Lewis, A. B., Pearson, B. Z., Eilers, R. E., & Um bel, V. C. 2002. Effects of bilingualism and bilingual education on oral and written Spanish skills: A multifactor study of standardized test outcomes. In D. K. Oller, & R. E. Eilers Eds., Language and literacy in bilingual children pp. 3 21. Clevedon, England : Multilingual Matters. Cohen, J. 1992. A power primer. Psychological Bulletin, 112 115 159. Coleman, M. R., Buysse, V., & Neitzel, J. 2006. Establ ishing the evidence base for an emerging early childhood practice: Recognition & Response. In V. Buysee & P.W. Wesley Eds., Evidence based practice in the early childhood field pp. 195 225. Washington, DC: ZERO TO THREE Press. Comeau, L., Comier, P., Grandmaison, E., & Lacroix, D. 1999. A longitudinal study of phonological processing skills in children learning to read in a second language. Journal of Educational Psychology, 91 29 Â€ 43. Cummins, J. 1979. Linguistic interdependence and t he educational development of B ilingual children. Review of Educational Research, 49 222 251. Curtis, M. E. 1980. Development of components of reading skill. Journal of Educational Psychology 72 656 Â€ 669. DÂšAnguilli, A., Siegel, L. S. & Maggi, S. 2004. Lite racy instruction, SES, and word r eading achievement in English language learners and children with English as a first language: A longitudinal study. Lea rning Disabilities Research & Practice, 19 202 213. Day, H. D., & Day, K. 1980. The reliability and validity of the Concepts About Print and Record of Oral Language Arlington, Virginia: Resources in Education. ERIC Document Reproduction Service No ED179932 Dickinson, D. K., McCabe, A., Clark Chiarelli, N., & Wolf, A. 2004. Cross language transfer of phonological awareness in low income Spanish and English bilingual preschool children. Applied Psycholinguistics, 25 323 347. Dickinson, D. K., McCabe, A., Anastasopoulous, L., Peisner Feinberg, E. S. & Poe, M. D. 2003. The comprehensive language approach to early literacy: The interrelationships among vocabulary, phonological sensitivity, and print knowledge among preschool aged children. Journal of Ed ucational Psychology, 95 465 481.
73 Dickinson, D. K., & Tabors, P. Eds. 2001. Beginning literacy with language: Young children learning at home and school Baltimore, MD: Paul H. Brookes Publishing. Dickinson, D. K., & Snow, C. E. 1987. Inter relationships among pre reading and oral language skills in kindergartners from two social classes. Early Childhood Research Quarterly, 2 1 25 Dufva, M., Niemi, P., & Voeten, M. 2001. The ro le of phonological memory, word recogni tion, and comprehension skills in reading development: from preschool to grade 2. Reading and Writing: An Interdisciplinary Journal, 14 91 177. Durgunoglu, A. Y. 2002. Cross linguistic transfer in literacy development and implications for language learners. Annals of Dyslexia, 52 189 204. Durgunoglu, A. Y. 1998. Acquiring literacy in En glish and Spanish in the United States. In A.Y. Durgunoglu & L. Verhoeven Eds., Literacy development in a multilingu al context pp. 135 145 Mahwah, NJ: Lawrence Erlbaum Associates. Durgunoglu, A. Y., Nagy, W.E. & Hancin Bhatt, B.J. 1993. Cross language transfer of phonological awareness. Journal of Educational Psychology, 85 453 465. Fletcher, J. M., & Lyon, G. R. 1998. Readi ng: A researched based approach. In W. M. Evers Ed., WhatÂ„s gone wrong in AmericaÂ„s classrooms pp. 50 72. Stanford: Hoover Institution Press. Freeman, R. 2004. Building on community bilingualism Philadelphia: Calson, Inc. Garcia, G. E. 1991. Factors influencing the Engl ish reading test performance of Spanish speaking Hispanic children. Reading Research Quarterly, 26 371 392. Genesee, F., Paradis, J., & Crago, M. B. 2004. Dual language development and disorders B altimore: ML Paul H. Brookes Publishing Co. Gerber, M., & Durgunoglu, A. Y. 2004. Reading risk and intervention for young English learners: Evidence from longitudinal intervention research. Learning Disabilities Research & Practice, 19 199 201. Gerber, M., Jimenez, T., Leafstedt, J., Villaruz, J., Ric hards, C., & English, J. 2004. English reading effects of small group intensive intervention in Span ish for K 1 English learners. Learning Disabilities Research & Practice 19 239 251. Geva, E. 2000. Issues in the assessment of reading disabilit ies in L2 children Beliefs a nd research evidence. Dyslexia, 6 13 28. Geva, E. & Wang, M. 2001. The development of basic reading skills in children: A cross language perspective. Annual Review of Applied Linguistics, 21 182 204.
74 Good, R. H., Simmons, D. C., & Smith, S. B. 1998. Effectiv e academic interventions i n t he United States: Evaluating and enhancing the acquisition of early reading skills. School Psychology Review, 27 1 45 56. Goswami, U. 2001. Early phonological development and the acquisiti on of literacy pp. 111 125. In S. B. Neuman & D. K. Dickinson Eds. Handbook of Early Literacy Research New York: The Guildford Press. Graves, A W., Gersten, R., & Haager, D. 2004. L iteracy instruction in multiple language first grade classrooms: Linking student outcomes to observed instructional practice. Learning Disabilities Research and Practice, 19 262 272. Gunn, B., Smolkow skil, K., Biglan, A., & Black, C. 2002. Supplemental instruction in decoding skills for Hispanic and non Hispanic students in ea rly elementary school: A follow up. Journal of Special Education, 36 69 79. Hammer, C. S., Miccio, A. W., & Wag staff, D. A. 2003. Home literacy experiences and t heir relationship to bilingual preschoolersÂš developi ng English literacy abilities: A n initial investigation. Language, Speech, and Hearing Services in Schools, 34 20 30. Hoover, W A., & Gough P. B. 1990. The simple view of reading. Reading and Writing: An Interdisciplinary Journal, 2, 127 Â€ 160. Juel, C. 1988. Learning to read and write: A longitudinal study of 54 children from first through fourth grades. J ournal of Educational Psychology, 80 437 447. Kindler, A. 2002. Survey of the statesÂ„ limited English proficient students and available educational programs and services, 1999 2000 summary report Washington, DC : National Clearing House for English Language Acquis ition and Language Instruction Educational Programs. Kinnear, P. R., & Gray, C. D. 2006. SPSS 14 Made Simple New York: Psychology Press. Koda, K. 1997. Orthographic knowledge in L2 lexical processing: A cross l inguistic perspective. In J. Coady & T. Huckin Eds., Second lan guage vocabulary acquisition: A rationale for pedagogy pp. 35 52. Cambridge, England: Cambridge University Press. Krauss, R. 1989. Carrot s eed N ew Y ork: Scholastic Langer, J. A., Bartolome, L., Vasquez, O., & Lucas, T. 1990. Meaning construction in school literacy tasks: A study of bilingual students. American Educational Research Journal, 27 427 471.
75 Lesaux, N. K & Siegel, L. S. 2003. The development of reading in children who speak English as a second language. Developmental Psychology, 39 1005 1019. Lindholm, K. J. 199 1. Theoretical assumptions and empirical evidence for academic achievement in two languages. Hispanic Journal of Behavioral Sciences, 13 3 17. Lindholm Leary, K. 2001. Dual language education Clevedon: Multilingual Matters. Lindsey, K. A., Manis, F. R., & Bailey, C. E. 2003. Prediction of first grade reading in Spanish speaking English langua ge learners. Journal of Educational Psychology, 95 3, 482 494. Lonigan, C. J. 2003. Development and promotion of emergent literacy skill in preschool children at risk of reading difficulties. In B. R. Foorman Ed., Preventing and rem ediating reading difficulties: Bringing science to scale pp. 23 50. Lonigan, C. J. Burgess, S. R., & Anthony, J. L. 2000. Development of emergent literacy and early reading skills in preschool childre n: Evidence from a la tent variable longitudinal study. Developmental Psychology, 36 596 Â€ 613. Lopez, A. & Co le, C. L. 1999. Effects of a parent i m plemented intervention on the a cademic readiness skills of five Puerto Rican ki ndergarten students in an urban s chool. School Psychology R eview, 28 439 447. Lpez, L. M. & Greenfield, D. B. 2004. The cross language tra nsfer of phonological skills in Hispanic Head Start children. Bilingual Research Journal, 28 1 18 Lopez, L. M., Tabors, P. O., & Paez, M. M. 2002. The Phonolo gical Awareness Test. Cambridge, MA: Harvard Graduate School of Education. Lyon, G. R. 1996. Learning disabilities. The Future of Children, 6 54 76. Lyon, G. R. & Chhabra, V. 1996. The current state of science and the future of specific r eading disability. Mental Retardation and Deve lopmental Disabilities Research Reviews, 2 2 9. MacDonald, W. G., & Cornwall, A. 1995. The relationship between phonological a wareness and reading and spelling achievement eleven years later. Jo urnal of Learning Disabilities, 28 523 527. Manis, F. R., Linsey, K. A., & Bailey, C. E. 2004. Development of reading in grades K 2 in Spanish speaking English Language learners. Learning Disabilities Research and Practice, 19 2 14 224.
76 McCardle, P., McCarthy, J., & Leos, K. 2005. English language learners and learning disabilities: Research agenda and implications for practice. Learning Disabilities Research and Practice, 20 68 78. Morris, D., Bloodgood, J. W., Lomax, R. G., & Perney, J. 2003. Developmental steps in learning to read: A longitudinal study in kindergarten and first grade. Reading Research Quarterly, 38 302 328. National Center for Educational Statistics. 2004. The nationÂšs r eport card: Reading highlights 2003. Washington, DC: U.S. Department of Education. National Assessment of Adult Literacy 2003. National Center for Educational Statistics Home Page. Retrieved May 10, 2006 from http://nces.ed.gov/naal/ No Child Left Behind NCLB. United States Department of Education Home Page Retrieved May 1 0, 2006 from http://www.ed.gov/nclb/landing.jhtml Office of Special Education Programs. 2002. Twenty fourth an nual report to Congress on the implementation of the Individuals with Disabilities Educational Act Washington, DC : US Department of Education. Ogbu, J. U. 1987. Variability in minority school perform ance: A problem in search of an ex planation. Anthropology & Education Quarterly, 18 312 334. Paez M. M., & Rinaldi, C. 2006. Predicting English word reading skills for Spanish speaking students in first grade. Topics in Language Disorders, 26 338 350. Paez, M. M. Tab ors, P. O & Lopez, L. M. 2007 Dual language and literacy development of Spanish speaking preschool children. Journal of Applied Developmental Psychology 28 85 102 Palacios, A. 1996. La semilla de zanahoria New York: Scholastic. Ori ginal work p ublished 1989 Palmer, J., McCleod, C., Hunt, E., & Davidson, J. 1985. Information processing correlates of reading. Journal of Memory and Language, 24 59 Â€ 8 8. Pedhazur, E. J. 1997. Multiple regression in behavioral resear ch 3 rd ed.. Fort Worth, TX: Wadsworth. Proctor, C. P., Carlo, M., August, D., & Snow, C. 2005. Native Spanish speaking c hildren reading in English: Toward a model of comprehension. Journal of Educational Psycholo gy, 2 246 256.
77 P ullen, P. C. & Justice, L. M. 2003. Enhancing phonological awareness, print awareness, and oral language skills in preschool children. Intervention in School and Clinic, 39 87 98. Quiroga, T., Lemos Britton, Z., Mostafapour, E., Abbott, R. D., & Berninger, V. W. 2002. Phonological awareness and beginning reading in Span ish speaking ESL first graders: Research into practice. Journal of School Psychology, 40 85 111. Riccio, C. A., Amado, A., Jimnez, S., Hasbrouch, J. E., Imhoff, B., & Denton, C. 2001. Cross linguistic transf er of phonological processing: D evelopment of a measure of p honological processing in Spanish. Bilingual Research Journal, 25 583 603. Shaywitz, B. A., Fletcher, J. M., & Shaywitz, S. E. 1994 Issues in the definition and classification of attention deficit disorder. Top ics in Language Disorder, 14 1 25. Sindelar, P. T., Lane, H. B., Pullen, P. C., & Hudson, R. F. 2002. Building, implementing, a nd sus taining a beginning reading improvement model: Lessons learned school by school. In M. R. Shinn Ed. Interventions for Academic and Behavior Problems II: Prevention and Remedial Approaches pp.703 727. Bethesda, MD : National Associatio n of School Psychologists. Snow, C. E., Tabors, P., Nicholson, P., & Kurland, B. 1995. SHELL: A method for a ssessing oral language and early literacy skills in kindergarten and first grade children. Journal of Research in Childhood Education 10 37 48. Stanovich, K. E. 1986. Matthew effects in reading: Some consequences in individual differences in the acquisition of literacy. Reading Research Quarterly, 21 360 406. Stevens, J. P. 2002. Applied multivariate statistics for the social sciences 4 th ed.. Mahwah, NJ: Lawrence Erlbaum Associates. Stevens, J. P. 1999. Intermediate statistics: A modern approach 2 nd ed.. Mahwah, NJ: Lawrence Erlbaum Associates. Storch, S. A., & Whitehurst, G. J. 2002. Oral language and code related precursors to reading: Evidence from a longitudinal structural model. Developmental Psychology, 38 934 947. Stothard, S. E. & Hulme, C. 1992. Reading comprehension difficulties in children: The r ole of la nguage comprehe nsion and working memory skills. Reading and Writing 4 245 256. Swan, D., & Goswami, U. 1997. Phonological awareness deficits in developmental dyslexia and the phonological representations hypothesis. Journal of Expe rimental Child Psychology, 66 18 41.
78 Swanson, H. E., Saez, L., Gerber, M. M., & Leafstedt, J. 2004. Literacy and cognitive functioning in bilingual and non bilingual children at or not at risk for reading disabilities. Journal of Educ ational Psychology, 96 3 18. Tabors, P., Pez, M., & Lpez, L. 2003. Dual language abilities of bilingual four year olds: Initial findings from the Early Childhood Study of Language and Literacy Development of Spanish speaking Children. NABE Journal of Research and Practice Winter, 70 91. http://www.uc.edu/njrp Thomas, W., & Collier, V. 2002. A national study of school effectiveness for language m inority studentsÂ„ long term academic achievement Final Report: Project 1.1. Santa Cruz, CA: Center for Research on Education, Diversity, and Excellence. Torgesen, J. K., & Hecht, S. A. 1996. Preventing and remediating reading disabilities: Instructional variables that make a difference for special students. In M. F. Graves, P. van de Broek, & B. M. Taylor Eds., The first R: Every childÂšs right to read pp. 133 159. New York: Teachers College Pres. Umbel, V. M., Pearson, B. Z., Fernandez, M. C., & Oller, D. K. 1992. Measuring b ilingual childrenÂšs receptive vocabularies. Child Development, 63 1012 1020. U.S. Census Bureau. 2000 Population projections program, population division, WDC 20233. [Online]. Available: www.census.gov Valdes, G., & Figueroa, R. A. 1994. Bilingualism and testing: A special case of bias Norwood, NJ: Ablex. Wagner, R. K., Torgesen, J. K., Rashotte, C. A., Hecht, S. A., Barker, T. A., Burgess, S. R., et al. 1997 Changing relations between phonological processing abilities and word level reading as children develop from beginning to skilled readers: A 5 year longitudinal study. Developmental Psychology, 33 468 Â€ 479. Whitehurst, G. J. & Lonigan, C. J. 1998. Child deve lopment and emerging literacy. Child Development, 69 848 872. Woodcock, R. W. 1991a. Woodcock Language Proficiency Battery Revised. Itasca, IL : Riverside Publishing. Woodcock, R. W. 1991b. Woodcock Language Prof icien cy Battery Revised: English and Spanish Forms: ExaminerÂ„s manual. Itasca, IL: Riverside Publishing. Woodcock, R. W. & Muoz Sandoval, A. F. 1995. Woodcock Language Proficiency Battery ÂŸ Revised: Spanish Form. Itasca, IL: Riverside Pub lishing.
79 Zecker, L. B. 2004. Learning to read and write in two lang uages: The development of e arly biliteracy abilities pp. 248 265. In C. A. Stone, E. R. Silliman B. J. Ehren, & K. Apel Eds., Handbook of Language and Literacy: Deve lopment and Disorders New York, NY : The Guilford Press.
81 Appendix A: English Book Task BOOK TASK *************** ENTIRE TASK IS TAPE RECORDED************ **** SECTION I 1. Hand the book to the child upside down and backwards. Say to the child: WHEREÂšS THE FRONT OF THE BOOK? 2. THE TITLE OF THIS BOOK IS THE CARROT SEED CAN YOU SHOW ME WHERE IT SAYS THAT ON THE COVER? 3. WHATÂšS THE BOY DOING? 4. THE CARR OT SEED BY RUTH KRAUSS Point to Âby Ruth Krauss.Â‹ WHAT DID RUTH KRAUSS DO? 5. PICTURES BY CROCKETT JOHNSON Point to ÂPictures by Crockett Johnson.Â‹ WHAT DID CROCKETT JOHNSON DO? 6. IÂšM GOING TO READ YOU THIS BOOK. CAN YOU OPEN THE BOOK FOR ME? 7. Point to the second page; not to the print. SHOW ME WHERE TO START READING. Read the story. Finish with THE END. If three or more squares are checked off for Section I, continue to Section II; otherwise end task here
82 SECTION II 9. Go back to the s econd page. NOW LETÂšS LOOK AT THE BOOK AGAIN. A LITTLE BOY PLANTED A CARROT SEED. HIS MOTHER SAID, IÂ“M AFRAID IT WONÂ“T COME UP. CAN YOU TELL ME WHAT THE BOYÂšS FATHER SAID? 10. HIS FATHER SAID, IÂ“M AFRAID IT WONÂ“T COME UP. AND HIS BIG BROTHER SAID IT WONÂ“T COME UP. SHOW ME WHERE THE WORD BROTHERÂš IS ON THIS PAGE. 11. EVERY DAY THE LITTLE BOY PULLED UP THE WEEDS AROUND THE SEED AND SPRINKLED THE GROUND WITH WATER SHOW ME WHAT HE PULLED OUT OF THE GROUND. 12. BUT NOTHING CAME UP. AND NOTHING CAME UP. EVERYONE K EPT SAYING IT WOULDNÂ“T COME UP IS SOMETHING GOING TO COME UP? If child answers only ÂYes,Â‹ then WHAT WILL COME UP? 14. BUT HE STILL PULLED UP THE WEEDS AROUND IT EVERY DAY AND SPRINKLED THE GROUND WITH WATER. AND THEN, ONE DAYA CARROT CAME UP. JUST AS THE LITTLE BOY HAD KNOWN IT WOULD. WHAT DOES THE BOY HAVE IN HIS WHEELBARROW? A CARROT!!! If four or more squares are checked off in Section II, continue to Section III; otherwise end task here SECTION III 15. NOW YOU TELL ME THE STORY. Hand the book to the child open to the second page. Let the child turn the pages. If child begins to read i.e. decode text, the n JUST TELL ME THE STORY FIRST AND WEÂšLL READ IT LATER. If child repeats the text either partially or substantially continue to Section IV; otherwise end task here
83 SECTION IV 16. Turn to the first page of text IÂšD LIKE YOU TO TRY TO READ THIS BOOK FOR ME. LETÂšS START HERE turn to the second page of the book. IÂšLL HELP YOU IF YOU GET STUCK. Point to each word. Supply correct word whenever the child hesitates or misreads a word. Stop the child after ÂHis father saidÂ‹ if assessor is providing more than half of the words on a page. Assessor finishes reading the book. Stop the child after ÂEvery dayÂ‹ if assessor is providing more than two words per page. Assessor finishes reading the book. THE END! WEÂšRE DONE!
84 Appen dix B: Spanish Book Task PRUEBA DEL LIBRO *************** GRABAR TODA LA PRUEBA**************** SECCI"N I 8. D el libro al nio/a boca abajo y al revs. CUL ES LA PORTADA DEL LIBRO? 9. EL TTULO DE ESTE LIBRO ES LA SEMILLA DE ZANAHORIA D"NDE EN LA P ORTADA CREES QUE DICE ESO? 10. QU EST HACIENDO EL NIO? 11. LA SEMILLA DE ZANAHORIA POR RUTH KRAUSS Seale Âpor Ruth Krauss.Â‹ QU" HIZO RUTH KRAUSS? 12. DIBUJOS DE CROCKETT JOHNSON Seale ÂDibujos de Crockett Johnson.Â‹ QU HIZO CROCKETT JOHNSON? 13. VOY A LEER EL LIBRO. PUEDES ABRIR EL LIBRO PARA LEERLO? 14. Seale la segunda pgina; pero no las palabras. SEALA D"NDE DEBO EMPEZAR A LEER. Lea la historia. FIN. Si ha marcado tres o ms cuadrados en la Seccin I, contine con la Seccin II; de otra ma nera pare la prueba aqu SECCI"N II AHORA VAMOS A VER EL LIBRO DE NUEVO. 9. Abra el libro en la segunda pgina. UN NIITO SEMBR" UNA SEMILLA DE ZANAHORIA. SU MAM LE DIJO, ME TEMO QUE NO CRECER. PUEDES DECIRME QUE DIJO EL PAP? 10. SU PAP DIJO, ME TE MO QUE NO CRECER. Y SU HERMANO MAYOR LE DIJO: -SEGURO QUE NO CRECER SEALA LA PALABRA HERMANOÂš EN ESTA PGINA.
85 11. TODOS LOS DAS EL NIO ARRANCABA LA MALEZA QUE CRECA ALREDEDOR DE LA SEMILLA Y ROCIABA EL SUELO CON AGUA. ENSAME QUE ARRANCABA DEL SUEL O. 12. PERO NADA CRECA. Y NADA CRECA. TODO EL MUNDO LE DECA QUE NADA CRECERA ALGO VA A CRECER? Si el nio responde solo ÂS'Â‹, entonces pregunte QU" VA A CRECER? 14. PERO TODOS LOS DAS L SEGUA ARRANCANDO LA MALEZA Y ROCIANDO EL SUELO CON AGUA. Y E NTONCES, UN DÂšACRECIÂ› UNA ZANAHORIA. TAL Y COMO EL NIO SABA QUE CRECERA. QU TIENE EL NIO EN LA CARRETA? UNA ZANAHORIA!!! Si ha marcado cuatro o ms cuadrados en la Seccin II, contine con la Seccin III; de otra manera pare la prueba aqu SECCI"N III 15. AHORA CUNTAME LA HISTORIA. Dele el libro al nio/a abierto en la segunda pgina. Deje que el nio/a pase las pginas. Si el nio/a empieza a leer, entonces diga CUNTAME LA HISTORIA PRIMERO Y LUEGO LO LEEMOS. Si el nio/a repite el tex to parcialmente o mayormente contine con la Seccin IV; de otra manera pare la prueba aqu.
86 SECCI"N IV 16. Pase a la segunda pgina del libro. ME GUSTARA QUE TRATARAS DE LEERME EL LIBRO. EMPEZEMOS AQU. TE AYUDAR SI TIENES PROBLEMAS. Seale cad a palabra. Lea la palabra cuando el nio titubea/vacila, tiene dificultad con la palabra o la lee incorrectamente. Pare al nio despus de ÂSu pap dijoÂ‹ si tiene que leerle al nio ms de la mitad de las palabras en la pgina. Examinador acaba de lee r el libro. Pare al nio despus de ÂTodos los d'asÂ‹ si tiene que leerle al nio ms de dos palabras por pgina. Examinador acaba de leer el libro. Â¡FIN! Â¡ACABAMOS, MUCHAS GRACIAS!
87 Appendix C: English Book Task Scoring B ook Task Scoring Â¡ is always zero points Code switching is always zero points 1. WHEREÂšS THE FRONT OF THE BOOK? 1 point: Correct 2. CAN YOU SHOW ME WHERE IT SAYS THAT ON THE COVER? 1 point: Points to print not in title 2 points: Points to any w ords in the title 3. WHATÂšS THE BOY DOING? 1 point: Dropping/ planting something, any reasonable answer 2 points: Correct planting a seed 4. WHAT DID RUTH KRAUSS DO? 1 point: Wrote/made the book, author 5. WHAT DID CROCKETT JOHNSON DO? 1 point: Drew/made the pictures 6. CAN YOU OPEN THE BOOK FOR ME? 1 point: Turns to first or second page 7. SHOW ME WHERE TO START READING. 1 point: Points to any print on page not beginning of sentence 2 points: Points to beginning o f sentence 9. CAN YOU TELL ME WHAT THE BOYÂšS FATHER SAID? 1 point: Some but not all of the words or correct meaning 2 points: Completely correct 10. SHOW ME WHERE THE WORD BROTHERÂš IS ON THIS PAGE. 1 point: Points to any print not bro therÂ‹ 2 points: Points to the word ÂbrotherÂ‹ 11. SHOW ME WHAT HE PULLED OUT OF THE GROUND. 1 point: Points at the weed
88 12. IS SOMETHING GOING TO COME UP? 1 point: Plant, flower, tree 2 points: Carrot/carrot seed 14. W HAT DOES THE BOY HAVE IN HIS WHEELBARROW? 1 point: A plant/flower/tree 2 points: A carrot 15. NOW YOU TELL ME THE STORY 1 point: Repeats the text of the book: partially correct 2 points: Repeats the text of the book: substantially correc t 16. IÂšD LIKE YOU TO TRY TO READ THIS BOOK FOR ME. 1 point: Child reads to page 3 2 points: Child reads to page 5 3 points: Child reads entire book
89 Appendix D: Spanish Book Task Scoring Prueba del Libro Scoring Â¡ es siempre cero puntos ÂCode switchingÂ‹ es siempre cero puntos 15. CUL ES EL FRENTE DEL LIBRO? 1 punto: Correcto 16. D"NDE EN LA PORTADA CREES QUE DICE ESO? 1 punto: Seala unas palabras pero no del ttulo 2 puntos: Seala cualquier palabra del ttulo 17. QU EST HACIENDO EL NIO? 1 punto: Tirando/ sembrado algo, cualquier respuesta que sea razonable 2 punto: Correcto sembrando/plantando una semilla 4. QU HIZO RUTH KRAUSS? 1 punto: Escribi/hizo el libro, autora 5. QU HIZO CROCKETT JOHNSON? 1 punto: Dibuj/hizo los dibujos 6. PUEDES ABRIR EL LIBRO PARA LEERLO? 1 punto: Abre el libro en la primera o segunda pgina 7. SEALA D"NDE DEBO EMPEZAR A LEER. 1 punto: Seala cualquier palabra en la pgina no el prin cipio de la oracin 2 puntos: Seala el principio de la oracin 9. PUEDES DECIRME QUE DIJO EL PAP? 1 punto: Algunas pero no todas las palabras o significado correcto 2 puntos: Completamente correcto 10. SEALA LA PALABRA HERMANOÂš EN ESTA PG INA. 1 punto: Seala cualquier palabra no hermanoÂš 2 puntos: Seala la palabra hermanoÂš 11. ENSAME QUE ARRANCABA DEL SUELO. 1 punto: Seala la maleza
90 12. ALGO VA A CRECER? 1 punto: Planta, flor, rbol 2 puntos:Zanahoria/semilla de zanahoria 14. QU TIENE EL NIO EN LA CARRETA? 1 punto: Una planta/flor/rbol 2 puntos: Una zanahoria 15. AHORA CUNTAME LA HISTORIA. 1 punto: Repite el texto del libro: parcialmente correcto 2 puntos: Repite el texto del libro: mayormente corre cto 16. ME GUSTARA QUE TRATARAS DE LEERME EL LIBRO. 1 punto: Lee el libro hasta la pgina 3 2 puntos: Lee el libro hasta la pgina 5 3 puntos: Lee todo el libro