|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
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 001681006
007 cr mnu|||uuuuu
008 051208s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001123
Sansosti, Jenine M.
General education teachers and classroom-based interventions
h [electronic resource] :
b knowledge, training, and building-level influences /
by Jenine M. Sansosti.
[Tampa, Fla.] :
University of South Florida,
Thesis (Ph.D.)--University of South Florida, 2005.
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 203 pages.
ABSTRACT: Intervention assistance (IA) programs have been developed as a mechanism for avoiding costly special education referrals and for supporting teachers in their instruction of students with varying needs within the general education classroom (Safran and Safran, 1996). Although IA programs are designed to be consultative, multidisciplinary approaches to assisting teachers, some studies report that teachers conduct the majority of classroom-based interventions for a given student on their own prior to referring students to an IA team (Wilson, Hagen, Gutkin, and Oats, 1998). It is important to determine what interventions or strategies teachers commonly consider and what factors are associated with breadth and depth of intervention knowledge. The purpose of the present study was to replicate a portion of the research of Wilson et al. (1998), which assessed general education teachers knowledge of classroom-based interventions.The present study also extended the work of Wilson et al. by using an exploratory descriptive/nonexperimental design to examine the degree to which teachers individual professional characteristics, as well as the IA practices of the schools in which they work, were related to their knowledge of interventions. Twenty-nine general education teachers in Hillsborough County, FL responded to a vignette describing a typical classroom-based problem in a structured-interview format. Participants responses were then counted and coded for (a) how specifically interventions were described, and (b) what types of interventions the teachers used (e.g., instructional, behavioral, etc.). Teachers also completed a brief demographic questionnaire, which included items about the IA programs at their schools, as well as their individual referral history over the last two years, and the degree to which they had been trained in classroom-based interventions.
Adviser: Linda M. Raffaele Mendez, Ph.D.
x Interdisciplinary Education
t USF Electronic Theses and Dissertations.
General Education Teachers and Classroom Based Interventions: Knowledge, Training, and Building Level Influences by Jenine M. Sansosti A thesis submitted in partial fulfillment of the requirements for the degree of Education Specialist Dep artment of Psychological and Social Foundations College of Education University of South Florida Major Professor: Linda M. Raffaele Mendez, Ph.D. Michael J. Curtis, Ph.D. John Ferron, Ph.D. Date of Approval: April 8, 2005 Keywords: educators, mult idisciplinary teams, prereferral intervention, intervention assistance, problem solving, school factors Copyright 2005, Jenine M. Sansosti
Acknowledgements This project could not have been completed without the support of a dedicated group of in dividuals. I am extremely grateful to my major professor, Dr. Linda M. Raffaele Mendez, for her patience, support, endless faith, and countless hours of problem solving the many logistical challenges that arose. I am also grateful to the members of my co mmittee, Dr. Michael Curtis and Dr. John Ferron, for their flexibility, availability, and guidance. I would also like to extend a special thanks to the 2004 2005 1 st year cohort of students from the School Psychology Program at the University of South Flo rida who volunteered to assist with the data collection for this project: Alana, Cesnae, Decia, Emily, Iravonia, Jason and Kristen I literally could not have done this without you. My husband, Frank, also has my endless gratitude for his support (both p ersonal and professional) in helping me to complete this project. In addition, I am deeply thankful to Dr. Caryl Palmer Wilson, whose past research helped inspire this project and whose materials and insight helped inform its method. Thank you for your generosity in sharing your work. Finally, I wish to thank the Florida Association of School Psychologists (FASP) for their financial support of this research project.
i Table of Contents List of Tables i v List of Figures v i A bstract vi i Chapter I: Introduction 1 Brief Review of Literature 1 Teacher Knowledge of Interventions 2 Teacher Training in Interventions 4 Building level IA Practices 5 Rationale 6 Purpose 7 Definitions 8 Pre re ferral Intervention 8 Knowledge 8 Specificity 9 Intervention Nature/Type of Intervention 10 Intervention Assistance (IA) Team 12 Chapter II: Review of the Literature 13 Historical Background 13 Challenges Facing Todays General Educators 16 Intervention Assistance Programs 17 Teacher Knowledge and Choice of Interventions 21 Determinants of Teachers Intervention Knowle dge : Training 29 Determinants of Teachers Intervention Knowledge: Building level IA Practices 31 System Factors 32 Process Factors 34 Purpose of Study 36 Research Questions 37 Chapter III: Research Met hods 39 Participants and Setting 39 Sampling Procedure 39 Setting 4 2
ii Materials and Measures 4 3 Demographic Questionnaire 4 3 Interview Instructions and Standardized Vignette 4 4 Interview Coding Form 4 7 Coding Definitions 4 7 Procedure 50 Protections of Confidentiality and Informed Consent 50 Preliminary Data Colle ction (Pilot Study Phase) 5 1 Main Study Data Collection 53 Data Entry 5 5 Interrater Reliability 57 Data Analyses 59 Independent Variables 62 Dependent Variables 62 Design 64 Statistical Analysis 65 Chapter IV: Results 68 Descriptive Analyses 68 Questionnaire 68 P articipant Demographics 68 Participant Problem Solving Characteristics 69 Vignette and Structured Interview 7 9 Post Hoc Analyses 82 Pe rception of Training Adequacy 82 Presence and Frequency of Hypotheses 83 Correlational Analyses 85 Rationale for Exploratory Correlational Analysis 8 7 Sample Size 8 7 Number of Correlations 84 Relationship Between Teacher Char acteristics and Interview Outcomes 8 7 Relationships Among Presence of Intervention Types 91 Post Hoc Analyses 92 Perception of Training Adequacy 92 Presence and Frequency of Hypotheses 92 Chapter V: Discussion 9 5 Teachers Self Reported Intervention Knowledge 9 6 Number of Interventions 9 8 Type of Interventions 9 8 Specificity of Intervention Descriptions 101 Teacher Problem Solving Characteristics 10 7 Existence and Practices of IA T eams 10 8 Self Reported Consultation Behaviors 1 10
iii Training Experiences 1 12 Relationships Between Selected Teacher Characteristics and Interview Outcomes 1 13 Limitations 1 15 Nonrandom Se lection of Schools 11 5 Homogeneity of Participants 11 6 Small Sample Size 11 6 Methodological Limitations 11 7 Implications for School Psychologists 1 20 Implications for Future Research 1 22 Conclusion 124 References 1 26 Appendices 1 34 Appendix A: Information Letter to Teachers 1 35 Appendix B: Summary of Pilot Results 13 7 Appendix C: Demographic Informat ion Form 1 44 Appendix D: Interview Instructions 14 8 Appendix E: Standardized Vignette 152 Appendix F: Coding Form 1 53 Appendix G: Code Definitions 1 56 Appendix H: Completed Coding Form for Participant 4A3 1 60 Appendix I: Completed Coding Form for Participant 3B3 1 67 Appendix J: Completed Coding Form for Participant 3D2 1 70 Appendix K: Completed Coding Form for Participant 2E2 1 75 Appendix L: C ompleted Coding Form for Participant 4F3 1 83 Appendix M: Correlations Among Specificity Ratings 1 91 Appendix N : Point Biserial Correlations (r pb ) Between Selected Teacher Characteristics and Intervention Types Suggested 1 9 2 Appendix O: Phi Coefficients (r ) Between Intervention Types 1 93
iv List of Tables Table 1: Demographic Characteristics of Participating Schools 4 2 Table 2: Independent and Dependent Variables, Types of Variables 59 Table 3: P earson Product Moment Correlations ( r ) Between the 2002 2003 and 2003 2004 Rates of Referrals to Problem Solving Teams, Referrals to School Psychologists, and Children Found Eligible for ESE Services 61 Table 4: Selected Participant Demograph ics 69 Table 5: Item Means By School: Existence, Schedules, and Requirements of Schoolwide Problem solving Teams 71 Table 6: Sub Item Means: Utilization of Best Practices among Schoolwide Problem Solving Teams 7 3 Table 7: Sub Item Means: Teachers Training Experiences in Classroom based Interventions 7 9 Table 8: Intervention Suggestions in Response to the Vignette and Structured Interview by Number of Teachers Suggesting, Relative Frequency, and Mean Specificity 8 3 Table 9: Sample Confidence Intervals for Obtained Pearson Product Moment Correlations ( r ) 8 7 Table 10: Pearson Product Moment Correlations ( r ) Between Selected Teacher Characteristics, Total Number of Interventions, and Overall S pecificity 8 9 Table 11: Pearson Product Moment Correlations ( r ) Between Selected Teacher Characteristics, Perception of Training Adequacy, Hypothesis Presence and Frequency 9 3 Table 12: Comparison of Findings Between Wilson et al. ( 1998) and Present Study, With Regard to Number and Type of Interventions 9 7
v Table 13: Comparison of Findings Between Wilson et al. (1998) and Present Study, With Regard to Specificity of Interventions 9 7
vi List of Figures Fig ure 1: Educational Professionals With Whom Teachers Consult About Difficult to Teach Students 7 7 Figure 2: Percent of Total Sample Suggesting Each Intervention Type 81 Figure 3: Total Number of Suggestions and Mean Specifici ty Rating by Intervention Type 82 Figure 4: Boxplot Depicting Distribution for Hypothesis Frequency 84
vii General Education Teachers and Classroom Based Interventions: Knowledge, Training, and Building Level Influences Jenine M. S ansosti ABSTRACT Intervention assistance (IA ) programs have been developed as a mechanism fo r avoiding costly special education referrals and for supporting teachers in their instruction of students with varying needs within the general education classro om (Safran & Safran, 1996) A lthough IA programs are designed to be consultative, multidisciplinary approaches to assisting teachers s ome studies report that teachers conduct the majority of classroom based interventions f or a given student on their own prior to referring students to an IA team (Wilson, Hagen, Gutkin, & Oats, 1998). I t is important to determine what interventions or strategies teachers commonly consider and what factors are associated with breadth and depth of intervention knowledge. The purpose of the present study was to replicate a portion of the research of Wilson et al. (1998), which assessed general education teachers knowledge of classroom based interventions. T he present study also extended the work of Wilson et al. by using a n exploratory descriptive/nonexperimental design to examine the degree to which teachers individua l professional characteristics, as well as the IA practices of the schools in which they work, were related to their knowledge of interventions. Twenty nine ge neral education teachers in Hillsborough County FL responded to a vignette describing a typical classroom based problem in a structured interview format. Participants responses were then counted and coded for (a) how specifically interventions were de scribed, and (b) what types of interventions the teachers used (e.g., instructional, behavioral, etc.). Teachers also completed a brief demographic questionnaire, which included items about
viii the IA programs at their schools, as well as their individual refe rral history over the last two years, and the degree to which they had been trained in classroom based interventions. Results were similar to Wilson et al. with regard to number of intervention ideas, but teachers were more specific than in previous invest igations. Descriptive data regarding teachers characteristics as problem solvers and their perceptions of IA at their school are offered, but few noteworthy relationships were identified between these variables and structured interview outcomes. Neverth eless, the present study offers a glimpse into the intervention practices of general education teachers. Implications for both school psychology practice and research are offered.
1 Chapter I Introduction Brief Review of Literature Tradit ional special education practices, including refer test place procedures resulting in restrictive placements for students, have been found to be minimally successful in improving student outcomes (Reschly, 1989). Furthermore, research suggests that this a pproach to eligibility determination results in overidentification and inappropriate placements in special education classes ( Fuchs, Fuchs, Bahr, Fernstrom, & Stecker, 1990; Safran & Safran, 1996) Efforts to remedy these problems have pushed for students with special needs to be integrated in general education classrooms whenever possible, and have emphasized students rights to receive services without having to be conferred a diagnostic label (NASP, 1995). As a result of these initiatives and movement s, there is today an increased number of difficult to teach students (with and without diagnosed disabilities) who are educated in a least restrictive environment (LRE). A variety of students and needs comprise todays general education classroom. While this change in student demographics is positive in that it suggests the realization of inclusion movements, the increase of difficult to teach students in general education also poses some serious challenges for teachers. I n response to national reform mo vements, h igher academic standards in education have increased pressure on teachers to move rapidly through complex curricula, leaving difficult to teach students behind (Rathvon, 1999). Although the spirit of the inclusion/LRE movement is laudable, the r ealities of these ideals may include unintended negative outcomes for difficult to teach students. Intervention assistance (IA) programs have been developed as a mechanism for avoiding unnecessary and costly special education referrals, as
2 well as for supp orting teachers in their instruction of students with varying needs within the general education classroom. Multidisciplinary IA teams, typically consisting of teachers, school psychologists, special educators, counselors, and other relevant personnel, re present a viable alternative to traditional refer test place practices, as they have been found to be generally successful in developing interventions and promoting collaboration among school personnel within an ecological/problem solving model (Harrington & Gibson, 1986; Nelson, Smith, Taylor, Dodd, & Reavis, 1991; Rathvon, 1999). Furthermore, IA programs facilitate compliance with LRE of IDEA 1997 by strengthening teachers capacities to meet the increasing diversity of student needs within the general e ducation context, and by challenging teams to find solutions to problems, rather than diagnoses. Although IA programs offer much in the way of innovation and support to schools and teachers, the literature suggests several areas in which these teams and pr ocedures must improve. Some studies of IA activities report that although IA programs are designed to be consultative, multidisciplinary approaches to assisting difficult to teach students, teachers actually conduct the majority of classroom based interve ntions for a given student on their own, prior to referring students to an IA team (Wilson, Hagen, Gutkin, & Oats, 1998). Even when teachers do consult with IA teams, research suggests that general education teachers often bear the heaviest burden of desi gning, implementing and evaluating the efficacy of those interventions (Bahr, 1994). In light of these findings and teachers considerable influence on the IA process, an examination of teacher knowledge and relevant characteristics (e.g., training experi ences and building level influences) seems warranted. Teacher knowledge of interventions Unfortunately, there is little research to date that quantifies teacher knowledge of interventions and/or intervention skills. This may be due to the inherent comple xity of defining and adequately assessing the whole of teachers knowledge of intervention strategies. Of the minimal data available, the picture of teachers skills in developing and
3 implementing interventions is somewhat bleak. For example, Pugach (198 5) reported that the majority of general education teachers she interviewed attempted intensive, high quality, ecologically focused interventions with difficult to teach students before initiating a referral for suspected disability However, more recent research by Myers and Holland (2001) indicates that teachers rarely consider the function of behaviors (e.g., attention, escape, tangible, sensory) before suggesting an intervention. The authors of this study concluded that teachers may take a cookbook a pproach to intervention selection, simply choosing from a list of commonly accepted strategies for a given problem. These data indicate that teachers may not individualize interventions appropriately, rendering them less effective. This finding is parti cularly concerning, as IDEA 1997 regulations require a functional behavioral analysis (FBA) for some disability determinations. There is additional support for this trend in the intervention literature. Professional best practices for intervention design and implementation are frequently overlooked. Many studies do not include operational definitions of behaviors or independent variables, and often lack a detailed intervention plan. Gresham (1989) speculated that the ineffectiveness of many prereferral interventions can be attributed to poor treatment integrity, which is likely a result of treatment plans that are low in specificity and precision. Flugum and Reschly (1994), in a review of permanent product data from teacher implemented interventions, fo und that the typical prereferral intervention does not include many of the elements considered essential to quality intervention development (e.g., behavioral definitions, direct measure of student outcome, systematic and detailed intervention plan, grap hic representation of results, comparison of student progress to baseline levels). Wilson et al. (1998) used a two part interview to investigate general education teachers knowledge of classroom based interventions. First, the authors administered a st andardized vignette describing a hypothetical classroom behavior problem to all teachers and asked them to list all the
4 interventions they knew of to help the hypothetical child reach two behavioral goals. In the second phase, teachers were asked to recal l an actual student they had worked with who eventually qualified for services under a mildly handicapped category. Teachers were prompted to recall all of the intervention strategies they had attempted with the student at varying points in the referral process. Throughout the interview, the interviewer reminded teachers to describe interventions and strategies as specifically as possible and provided examples and nonexamples of specific responses. In analyzing teachers performance on both of these tas ks, Wilson et al. (1998) found that teachers intervention descriptions were generally lacking in specificity. Ten percent of responses in the standardized vignette condition and 13% of responses in the referral case condition were rated highly specific. Specificity of intervention description, Wilson et al. note, has been linked to treatment integrity, or teachers adherence to the intervention plan (Gresham, 1989). Interventions were lacking in variety most were behavioral in nature and teachers data collection strategies were rated as mostly haphazard in approach. The authors concluded that teachers may be inadequately trained to fulfill this important IA team role. Teacher training in interventions Clearly, preservice and inservice training are excellent strategies for developing teachers skills in the area of interventions, but are not discussed extensively in IA literature. Wood, Lazzari, Davis, Sugai, and Carter (1990) found that nearly 25% of states require or recommend intervention assista nce programs, but only 3 states reported that training in this area was provided at the preservice level by universities and colleges. Within the relevant teacher training literature, few studies directly address the goal of increasing teachers skills fo r designing and conducting classroom based interventions. For example, an experimental analysis of a supervised training experience for teaching interns to practice intervention development and implementation (Newman, 1999) used interns perceptions of se lf efficacy and locus of control as outcome measures.
5 Though innovative in their approach, studies such as Newmans (1999) that measure perceptions of efficacy or increased competence do not provide necessary information about how such programs might imp act teachers skills or intervention practices. Given the importance of this role for teachers, especially in states in which IA processes are mandated, it is surprising that teacher training experiences in classroom based interventions have garnered so l ittle attention in the extant literature. Standards for training teachers to provide necessary accommodations to difficult to teach students have not been well delineated in the theoretical or empirical IA literature. Building level IA practices. Althoug h the aforementioned literature has demonstrated the need for research examining teacher knowledge and training in interventions, it is also imperative that teachers intervention efforts are considered within the greater context of the school in which the y work. Teachers do not operate in a vacuum and, as part of a complex educational system, any investigation of teachers skills must consider the impact that existing prereferral programs within their schools may have on intervention knowledge and practic es. There is some evidence that participation in IA teams and programs can improve individual teachers classroom based interventions. Results from a controlled experimental design by Pugach and Johnson (1988) indicated that participation in an IA like pr ogram increased teachers tolerance for a broad range of cognitive abilities, improved 91% of their target behavior definitions, and generated apparently successful interventions (teachers perceptions of effectiveness were reported in lieu of data on actu al behavior change). These findings suggest that teachers intervention skills and perceptions regarding IA programs can be impacted merely by participating in such teams. Kovaleski (2002) summarized factors that have been found to be related to successf ul prereferral intervention programs at the building level. These IA best practices can be conceptualized as either system factors (characteristics of school environments that facilitate IA programs), or process factors (procedural factors that help IA pr ograms to realize meaningful outcomes). Although these
6 conditions are described by Kovaleski in the context of multidisciplinary, building wide IA teams, it can be argued that the presence or absence of these conditions may have an impact on the way schoo l personnel (i.e., general education teachers) individually conceptualize and approach intervention efforts for difficult to teach students. Rationale Although prereferral interventions such as instructional modifications or behavioral management strate gies are predominantly carried out by general education classroom teachers, the concept of prereferral intervention is grounded in collaborative consultation among general and special educators, school psychologists, school counselors, and other relevant p rofessionals (Graden, 1989). In particular, the school psychologist can be instrumental in guiding teachers toward the development of effective interventions to remediate academic and behavioral problems. Their experiences with children with academic an d behavioral problems, as well as their knowledge of child development, learning principles, and educational practices, make them a considerable resource for the design, implementation, and evaluation of classroom based interventions. In order for school psychologists to operate as effective consultants, it is necessary to have a more comprehensive understanding of teachers skills and abilities with regard to classroom based interventions than is currently offered by the extant literature. What interven tion strategies do teachers know to resolve common classroom problems? How specifically are they able to describe these interventions? Furthermore, information is needed about teachers training experiences in this realm: do teachers feel that they are p repared to fulfill this role? What training experiences have led general education teachers to their present levels of intervention knowledge? Such knowledge can lead to specific preservice and inservice training programs that building on teachers exist ing strengths and address identified areas of weakness.
7 Purpose The purpose of the present study was to replicate a portion of the research of Wilson et al. (1998), which conducted interviews with general education teachers to assess their knowledge of c lassroom based interventions. Specifically, this study used the structured interview and vignette portion of the Wilson et al. study. However, the present study was an extension of this approach in that, in this investigation, teachers individual profe ssional characteristics (including training experiences), as well as the IA practices of the schools in which they work, were measured to determine their degree of relatedness with teachers knowledge of interventions. This study addressed following resea rch questions: 1. What is the average number of interventions teachers offer to address a hypothetical classroom behavior problem? 2. How specific are teachers in descriptions of interventions/strategies they would use in their classroom (average specificity rat ing per teacher)? 3. What is the likelihood that a teacher will suggest a given type of intervention (e.g., instructional, behavioral, etc.)? 3a. What 2 or more intervention categories, if any, are likely to be suggested by the same teacher (i.e., what is th e probability that a given teacher will suggest both intervention type x and intervention type y ?) 4. What is the relationship between years of teaching experience and number of interventions/strategies suggested, specificity of interventions/strategies descr iptions, and the likelihood that a teacher will suggest a given type of intervention (e.g., instructional, behavioral, etc.)? 5. What is the relationship between the number of times the teacher has participated in IA meetings and the number of interventions/s trategies
8 suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? 6. What is the relationship between teachers referral to eligibility rate and number of interventions/stra tegies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? 7. What is the relationship between training experiences and number of interventions/strategies suggested, spe cificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? 8. What is the relationship between intervention assistance (IA) practices of the participants school and number of interventions/ strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? Definitions Prereferral intervention : A general education teachers modification of instruction or clas sroom management to better accommodate a difficult to teach pupil without disabilities (Fuchs, Fuchs, & Bahr, 1990). Note that this definition is generally not used to discuss multidisciplinary, consultative services provided to assist teachers in the dev elopment of classroom based interventions. Rather, prereferral intervention is used to refer to the actual interventions or strategies used to accommodate difficult to teach students within the general education setting, prior to any referral for suspect ed disability Knowledge : Operationally defined by Wilson et al. (1998) as the number of interventions offered to solve a problem and the specificity with which these interventions are de s cribed Thus, many highly specific intervention descriptions would indicate more intervention knowledge, while fewer, less specific interventions descriptions would be indicative of less intervention knowledge. It is important to note, however, that the measures proposed
9 for use in this study are not designed to be a com prehensive assessment of teachers intervention knowledge. The interview/vignette method used in the present study can be conceptualized as a brief measure that taps into teachers intervention knowledge, much in the way that a one minute curriculum based reading probe taps into a students reading skills. Just as a single curriculum based measure of reading is not used to make broad statements about a students reading strengths and weaknesses, the interview/vignette method of this study should not be ex pected to completely reveal teachers intervention knowledge. Results from Wilson et al. (1998) suggest, however, that the information obtained from the interview/vignette method can be used to make valid statements regarding teachers general knowledge a nd comfort level with various classroom based interventions. Specificity : Operationally defined by Wilson et al. (1998) as the amount of detail or precision included in teachers descriptions of intervention strategies Procedures for coding specificity o f descriptions were adapted from the work of Gresham (1989), and were used by Wilson et al. (1998). Three levels of specificity may be observed in teachers responses to the interview/vignette: o low specificity : descriptions consist of nonspecific or vagu e recommendations. Intervention could not be implemented based on current description alone. Example: I could change the workload o moderate specificity : description contains some, but not complete, detail. Intervention could be implemented if some addi tional details were to be provided. Example : I could sho rten his daily math assignments. o high specificity : descriptions demonstrate a detailed plan for assisting the hypothetical student. Intervention could be implemented on the basis of this descriptio n alone, and there
10 should be no questions about how the intervention would be implemented. Example: I would take Johns math worksheets and cut them into strips of five problems each. When he finishes one strip, he will come up to my desk, and I will tel l him hes doing a good job and give him another strip. This will break down his work into smaller chunks and allow him to get a brief rest and some praise in between sets of problems. Intervention Nature/Type of Intervention : To code the nature of inter ventions, a modified version of a scale developed by Ysseldyke, Pianta, Christenson, Wang, and Algozzine (1983) was used. This scale was also used to analyze responses in the Wilson et al. (1998) study ; modifications to the Wilson et al. version were made following a pilot study of the research measures and materials. Interventions were categorized along the following types: o I nstructional : A change in the teachers approach to instructing the child Examples: Providing individualized instruction or assist ance with classroom work restating direction s, or modifying length, content, or modality of academic task. o B ehavioral : Consequence oriented approach to changing identified behavior using positive or negative reinforcement, removal from reinforcement, or application of punishment. Examples: Differential reinforcement of alternative behaviors (target student or other students), time out (removal from reinforcement, or other removal from classroom, or positive reinforcement in the form of praise, stickers/t okens/points. o Classroom structure : Changes in the amount of the structure provided for student within the classroom contex t. Not limited to instructional tasks may include changes to students
11 responsibilities or duties that impact level of structure of change to the classroom environment as a whole. Examples: M oving students sea t, assigning a peer tutor to assist with in class work and/or behavior, or assigning student duties to allow appropriate opportunities to be out of seat or talking. Note that as signing a peer tutor is considered classroom structure rather than emotional/social support because it is intended to provide greater structure for the students academic performance and behavior in the classroom, not to promote friendships. o Interdiscipl inary S upport : Additional specialized assistance student receives directly from other school personnel Examples: P re taught vocabulary with the resource teacher, counseling with the school counselor. o Information Gathering : Teacher requested or teacher g athered additional information regarding the student Examples: C heck the students cumulative file, called parents to ask if there is anything going on at home, or refer to the child study team for further evaluation. Note that calling home in this contex t is considered information gathering rather than communication parents because its purpose is to get more information, not to make changes in student behavior. o Materials : Specifically identified materials used to supplement instruction or remediation Examples: A udio visual tapes, manipulatives o Communication, with student, class, or family : Conversations, comments, or nonverbal cues directed at the student, class or parent(s) that are intended to change student(s) behavior
12 Examples: T elling student a bout the importance of not calling out, discussing how it disrupts others thinking; alerting the whole class to raise their hands before speaking; conference with parents to come up with a plan to change behav ior at school and at home o Emotional/social su pport : Efforts on teachers part to provide emotional support to the student, increase students self esteem, or create/enhance student friendships. Example s : W orking on building him up, achieving small successes; pairing student up with someone who can se rve as a peer buddy to promote friendships. o Compound : An intervention which consists of more than one code above. The intervention must be described in a way that it is clear the multiple components are intended to be delivered simultaneously. Example : De veloping a behavioral contract ( Behavioral ), which is monitored by the guidance counselor ( Interdisciplinary Support ), and which is sent home to parents as a means of communication about his behavior ( Communication Parents ). Intervention Assistance (IA) team : The term intervention assistance (IA) is used to describe formalized, data based consultation services used to generate classroom based interventions for students who are difficult to teach (Safran & Safran, 1996). The term IA is consistent with consultation and interventions literature; however, many other names have been used for this practice, such as Mainstream Assistance Teams, Pupil Assistance Teams, Child Study Teams and I Teams.
13 Chapter II Review of the Literature Historical Backgr ound The passage of the Education for All Handicapped Children Act (PL 94 142) in 1975 represent ed a critical turning point in both the institution of public education and the profession of school psychology. By mandating that all students have a right to a free and appropriate public education (FAPE), the law forever altered the way educators and administrators conceptualized education for students with special needs. As a result of this change, the decade after PL 94 142 saw special education referral r ates increase 16% nationally as teachers recognized that some students needs were not being appropriately met in general education classrooms (U.S. Department of Education, 1988 as cited in Fuchs, Fuchs, Bahr, Fernstrom, & Stecker, 1990). The demand for school psychologists also was amplified, as comprehensive psychoeducational evaluations became increasingly necessary for eligibility determination. In the early 1980s, roughly 4 6% of the school age population each year was referred for special educatio n evaluation, and 67 75% of those students were identified as disabled and placed into various special education programs (U.S. Department of Education, 1988 as cited in Fuchs, Fuchs, Bahr, Fernstrom, & Stecker, 1990). In the early days of special educatio n, there was little emphasis on designing classroom strategies that could prevent students from being labeled and placed in pull out or self contained programs. Special education was considered an intervention in its own right. The dramatic increase in need for assessment and special education services resulted in disillusionment in Louisiana, where a class action suit was filed ( Luke S. and Hans S. versus Nix et al. 1981 in Safran and Safran, 1996). The suit e stimated that as many as 10,000 children w ere left in the limbo state of
14 postreferral waiting for multifactored evaluations that were months overdue and charged that the state was unprepared to address the massive quantities of referrals in a timely manner (Safran & Safran, 1996). Furthermore, the lawsuit stated that g eneral educators were unable to determine what constituted appropriate referral to special education, indicating that some referrals were unnecessary and could have been resolved in general education classrooms. The ruling in this case stated that schools n eeded to provide a quick, effective evaluation system for students with special needs. From this edict, one of the first intervention assistance systems was created and implemented by classroom teachers in Louisiana. The new ap proach reportedly strengthened the whole system, increased collaboration among teachers, and nearly eliminated inappropriate referrals. However, this statement of improvement came from within the Louisiana education system, and an independent evaluation m ay have found differently (Safran & Safran, 1996). Unfortunately, early modifications such as these did not entirely solve the problems of special education, and the promise of PL 94 142 was not entirely fulfilled. Besides logistical problems of managing referrals, research has determined that the evaluation process can result in inappropriate placement into special education, which is considered largely undesirable. Students may be unnecessarily separated from their peers, stigmatized and labeled, and disrupted in their current educational progress. Further, the financial ramifications of unnecessary placements in special education are considerable (Fuchs, Fuchs, Bahr, Fernstrom, & Stecker, 1990). Even when educational modifications are warranted, r e search supports neither the traditional refer test place model of psychoeducational assessment in particular nor special education services in general as methods of promoting positive student outcomes (Reschly, 1988). According to the U.S. Department of Education in 1995, over 2.4 million students with learning disabilities age 6 21 receive special education services, but as few as 2 8% will return to general education over the course of the school year (Powell Smith & Stewart, 1999).
15 Another report est imated that 80% of students placed in special education remain there three years after initial placement (Clarizio & Halgren, 1993) This statistic implies that for students whose performance is significantly below that of their general education peers, i t is extremely difficult to reduce that discrepancy through services received in special education. Widespread discontent with the special education system led to a reform movement in the late 1980s, which advocated for less emphasis on the refer test p lace pattern that had developed in recent years. Many of the suggested changes in special education focused on including students with mild disabilities in regular education classrooms, citing the detrimental effects of irrevocably removing students from general education. These initiatives capitalized on portions of PL 94 142 and IDEA Part B stipulating that special education and related services must be provided in a setting that is the least restrictive environment (LRE) appropriate for the child. Ac cording to Jacob Timm and Hartshorne (1998), the rationale behind the LRE clause came from a realization in Congress that integration of children with disabilities into general education classrooms was not likely to occur without a legal mandate. This re cognition turned out to be an astute one. Although LRE mandates had been in place since the 1970s, it took years of litigation and struggle for LRE to shift to the forefront of the education consciousness. A nationwide movement toward inclusion ensued, i n which students are educated in the least restrictive environment, making accommodations for students in the general education classroom whenever possible. The Regular Education Initiative urged for the restructuring of both general and special education and some advocates suggested a complete merger between the two systems (Lloyd & Gambatese, 1990) The debate over partial inclusion versus total abandonment of the special education system continues today. A position paper issued in 1995 by the National Association for School Psychologists ( Rights Without Labels ; NASP, 1995) outlined several recommendations for schools and programs considering a noncategorical
16 approach to education. NASP provide d directives on a variety of issues relevant for both educa tors and administrators attempting to overcome the inherent problems of the special education system by focusing efforts in the general education classroom. In particular, the paper states that the resources and materials (including teachers and aides) ty pically used in special education must be made available in general education classrooms in order to reverse the practice of moving handicapped students to special education situations outside regular classes and schools (NASP, 1995). The NASP statement also addresse d the need to remediate educational difficulties before a referral is initiated in order to avoid unnecessary placements in special education and engender a respect for students rights (NASP, 1995). For this purpose, NASP recommend ed preref erral screening and intervention conducted by general education personnel and supported by other service providers within the school (school psychologists, special education teachers, social workers, administrators). Prereferral screening/intervention in t he general education environment, supported with special education resources such as personnel, strategies, and materials, was perceived by the educational establishment as a viable means for supporting the instruction of diverse groups of learners before a referral wa s made Originally known as p rereferral intervention (Graden, Casey, & Christenson, 1985) this service delivery model use d problem solving and consultation among educators and other professionals to develop hypotheses regarding potential ca uses for students academic and behavioral difficulties, formulate a systematic plan for treatment implementation, and empirically evaluate student outcomes to determine if further assessment, intervention, or special education placement are warranted. Cha llenges Facing Todays General Educators As a result of these initiatives and movements, there is today an increased number of difficult to teach students (with and without diagnosed disabilities) who are educated in a least restrictive environment. A var iety of students and needs
17 comprise todays general education classroom. More than two thirds of special education students receive the majority of their schooling in the general education classroom (U.S. Department of Education, 198 5 in Noell & Witt, 199 9). R esearch suggests that there are approximately one to two children with ADHD in the average classroom (Raffaele & Bradley Klug, 2000) and almost 20% of all students have significant difficulty learning to read (Good, Simmons, & Smith, 1998). In one study (Myers & Holland, 2000), general education teachers indicated that one in five of their students exhibited disruptive/off task behaviors and one in 20 exhibited aggressive behaviors to the extent that some form of intervention was necessary for their resolution. While these changes are positive in that they indicate the realization of inclusion movements, the increase of difficult to teach students in general education also poses some serious challenges for teachers. I n response to national reform m ovements, h igher academic standards in education have increased pressure on teachers to move rapidly through complex curricula, leaving difficult to teach students behind (Rathvon, 1999). Although the spirit of the inclusion/LRE movement is laudable, the realities of these ideals may include unintended negative outcomes for difficult to teach students. Intervention Assistance Programs It should be noted that much of the research on preventative problem solving and intervention has used the term prereferr al intervention to describe a teachers modifications of instruction or classroom management to better accommodate difficult to teach students. However, some scholars have suggested that this term perpetuates an attitude of categorical eligibility, makin g prereferral activities appear as another hurdle to overcome in the quest for formal evaluation and placement of students with special needs (Graden, 1989; NASP, 1994). The goal of so called prereferral activities, however, is based on the notion that students should not have to fail or demonstrate a significant deficit before they can receive educational support, and that referral to special education should be avoided if at all possible. Many other names have been
18 proposed for this practice, such as Intervention Assistance (IA), Mainstream Assistance Teams, Pupil Assistance Teams, Child Study Teams and I Teams. Kovaleski (2002) notes that there is presently a lack of professional consensus regarding both the terminology used to describe prereferral intervention and the various definitions or conceptualizations of prereferral intervention practices. For the purposes of this literature review, the terms intervention assistance or IA will be used to describe formalized, data based consultation serv ices used to generate classroom based interventions for students who are difficult to teach (Safran & Safran, 1996), as "intervention assistance best characterizes the goals and purposes of this effort. The term prereferral intervention will be used to refer to the actual interventions or strategies used to accommodate difficult to teach students within the general education setting, prior to any referral for suspected disability The earliest intervention programs were Teacher Assistance Teams (TATs), introduced as an alternative to traditional teacher inservice training (Chalfant, Pysh, & Moultrie, 1979). TATs primarily consisted of general education teachers, although other professionals (e.g., school psychologists, special educators, principals, etc .) were involved as needed. These groups were formed to boost teachers skills in educating difficult to teach students, taking traditional inservice one step further by emphasizing teacher accountability, communication, and effective decision making (Sin delar, Griffin, Smith, & Wanatabe, 1992). TATs are characterized in the educational literature as a self help approach available for teachers who felt they needed assistance with managing challenging students. Less emphasis was placed on the consultative and collaborative aspects of these teams, as the members typically were all teachers. Intervention assistance (IA) teams, or multidisciplinary collaborative teams, were developed in response to the overidentification of students with mild disabilities i n the early 1980s (Graden, Casey, & Bonstrom, 1983, 1985; Graden et al. 1985). These teams offer a more formalized, consultative approach to
19 prereferral intervention, with specific consultation procedures and goals specified prior to team formation. Like TATs, IA programs also work to increase teachers ability to deal effectively with students in general education but place primary emphasis on the development and evaluation of intervention plans to prevent inappropriate special education referrals and pl acements rather than on teacher empowerment (Rathvon, 1999; Safran & Safran, 1996; Sindelar et al., 1992). Although IA teams vary across schools and states, they typically share some common characteristics, including the following (Rathvon, 1999): Collabo rative, consultative approach : IA utilizes indirect services from other professionals. As a result, decisions are based on the input of several individuals. Further more the team approach is intended to prevent the workload of interventionary efforts fro m falling on a single person (e.g., the general education teacher), thus expediting the process and allowing more students to be helped simultaneously Facilitate compliance with LRE of PL 94 142/IDEA 97 : By strengthening teachers capacity to meet the in creasing diversity of student needs, more students will be able to be educated in general education. The IA process should lead to a reduction in referrals as teachers resolve more minor problems within the classroom, as well as an increase in referral pl acement accuracy as only the most critical referrals will be evaluated formally. Ecological perspective/ problem solving model : The IA process is not driven by a search for student pathology, nor by discrepancy models that require considerable assessment. Rather, teachers and other professionals look at the students environment (classroom and teacher factors), as well as variables in the curriculum, to locate prospective intervention targets. This approach may be empowering for teachers because many pot ential factors affecting student performance are under their direct control (classroom materials, peers, curriculum, time/practice allotted, and even their own instructional/behavior management practices).
20 Conversely, this ecological perspective may lead teachers to feel that they are solely responsible for a childs problematic behavior and might result in resistance to participate in the IA process. Emphasis on finding solutions rather than diagnosing problems : In the traditional refer test place paradi gm, professionals tend to search for answers to the questions: Is this student handicapped? If so, by what condition? However, in keeping with the ecological/problem solving model, IA programs ask, What can be done to help teachers improve the performa nce of this student in regular education? Environmental factors are considered in the definition and analysis of problems and the IA teams focus is on students outcomes before and after intervention. Rather than requiring a lengthy referral and assess ment process during which both student and teacher must wait to obtain support IA provides direct and immediate assistance to teachers. To date, both forms of building level intervention programs (TATs and multidisciplinary teams) are in use around the c ountry, although multidisciplinary teams are more commonly discussed in school psychology literature. Team names, formats, and goals vary according to district and state standards for prereferral practices, making summative research on prereferral interve ntion somewhat problematic (Kovaleski, 2002). Some reviews of the literature (e.g., Lloyd, Crowley, Kohler, & Strain, 1988; Nelson, Smith, Taylor, Dodd, & Reavis, 1991), as well as numerous empirical studies (e.g., Fuchs, Fuchs, Bahr, Fernstrom, & Stecker 1990; Harrington & Gibson, 1986; Kovaleski, Gickling, Morrow, Swank, 1999; Wilson, Gutkin, Hagen, & Oats, 1998) do not make a distinction between TATs and collaborative multidisciplinary teams in their discussion of prereferral intervention. Others (Saf ran & Safran, 1996; Sindelar et al., 1992) describe teacher based and multidisciplinary programs separately. This paper will primarily discuss prereferral intervention in the context of IA programs, whose collaborative, systems approach is most consistent with
21 practices in the School District of Hillsborough County (SDHC), in Tampa, Florida where the present study was conducted Teacher Knowledge and Choice of Interventions There is a wealth of research on the efficacy of IA programs (see Fuchs, Fuchs, B ahr, Fernstrom, & Stecker, 1990; Fuchs, Fuchs, & Bahr, 1990; Nelson et al., 1991; and Safran & Safran, 1996 for comprehensive reviews on this topic). Nelson et al. (1991) concluded that IA programs have a positive impact on special education service deli very systems and can serve an educative role in building teachers intervention development, implementation, and evaluation skills. In discussing the quality and impact of IA programs, however, it is important to consider what teachers know about interven tions, as well as the types of interventions typically used in general education classrooms. According to a survey of special education administrators regarding intervention assistance practices, multidisciplinary teams and general education teachers were the agents responsible for the design and evaluation of interventions roughly half of the time, while general education teachers alone were almost always the individuals implementing the intervention (92%; Bahr, 1990). Although IA teams are designed to r educe the burden of interventions on general education classroom teachers, Bahrs findings suggest that these professionals are still the primary individuals responsible for providing assistance to difficult to teach students in the general education class room. Teacher knowledge about interventions has been linked with their use of classroom interventions and self efficacy for teaching challenging students (Wilson et al., 1998). Furthermore, both Harrington and Gibson (1986) and Inman and Tolefson (1988) reported that teachers were unsure if intervention assistance teams provided them with new, untried classroom intervention ideas and if a sufficient variety of intervention options had been explored by the team. Thus, it would seem that educational researc hers and school personnel alike would benefit from a baseline measure of teachers knowledge and use of classroom interventions, to serve as a foundation on which IA team suggestions
22 may build. Additionally, an understanding of teachers acquaintance with interventions can impact research studies that aim to evaluate overall efficacy of IA programs. Unfortunately, there is minimal research available that quantifies teachers intervention skills and/or knowledge. This may be due to the difficulty inheren t in defining and adequately assessing the whole of teachers knowledge of intervention strategies. Of those studies conducted in this area, many focus on teachers understanding or recognition of interventions for specific disorders such as ADHD (e.g., H awkins, Martin, Blanchard, & Brady, 1991), or on broad teaching strategies and instructional modifications not necessarily related to the individualized nature of the IA process (e.g., use of modeling, visual aids, brainstorming, etc; Kling, 1997). A pauc ity of literature on teachers intervention skills means that administrators, school psychologists and other educational professionals can only operate on the assumption that teachers are prepared to design, implement, and evaluate interventions, and such an assumption may mean diminished benefits for children targeted by IA teams. Pugach (1985) interviewed 39 elementary and junior high school teachers to describe the day to day practices that influenced their decisions to refer difficult to teach students to special education, including intervention attempts prior to referral. In this study, prereferral interventions were not conducted in the multidisciplinary or teacher assistance team format traditionally described in IA literature. Rather, prereferral interventions were literally interventions implemented by the general education teacher prior to making a decision to refer a student for special education evaluations. Prereferral interventions described by teachers in Pugachs study (1985) were coded a s major (numerous/persistent attempts to remediate a problem, involving a specific intervention plan in which effects of intervention were evaluated) or minor (nonspecific, casual, passive, or nonsystematic attempts at remediation). In her analysis of qua litative data, Pugach found that the majority of elementary school teachers attempted major interventions prior to referring
23 students for special education evaluation. These teachers verbalizations in the interview session indicated that they supported a n ecological approach to prereferral intervention; they assumed that students were typically trying their hardest and took professional responsibility for recognizing student problems and rearranging the environment to address them accordingly. Although P ugachs data (1985) indicate that teachers can and often do implement intensive and exhaustive interventions considering environmental aspects of students problems, more recent research suggests otherwise. In fact, much of the recent literature on preref erral intervention and intervention assistance expresses concern with teachers abilities to effectively remediate students problems in the general education classroom (Nelson et al., 1991; Wilson et al., 1998). In a more specific investigation of teach ers intervention practices, Myers and Holland (2000) analyzed general and special educators choices of interventions for a hypothetical problem to determine whether the function of problematic behavior was considered. Functional behavioral assessment (F BA) attempts to generate hypotheses as to why a child is engaging in a given behavior, in order to develop more appropriate, individually oriented interventions. The FBA approach to intervening with behavior is consistent with professional best practices (Upah & Tilly, 2002), and is required by the 2004 reauthorization of the Individuals with Disabilities Education Improvement Act for some types of disability evaluations (IDE I A, 2004 ). To assess teachers consideration of behavioral function, Myers and Hol land (2000) sent surveys to 177 general education and 32 special education teachers. The survey consisted of three vignettes of children displaying problem behavior, and respondents were asked to supply intervention suggestions for each vignette. Each of the three vignettes implied a different behavioral function: seeking teacher attention, seeking peer attention, and escaping from an aversive activity. Teachers responses were rated by the authors as appropriate (i.e., the intervention addressed the imp lied function of problem behavior), inappropriate (i.e., the intervention did not address the implied function of problem behavior), or
24 vague (i.e., it was unclear whether the intervention suggestion addressed the behavioral function). Myers and Holland f ound that few general education or special education teachers suggested intervention strategies that were appropriate for the implied behavioral function. Teachers were more likely to suggest appropriate strategies for teacher attention seeking behaviors, than for escape motivated or peer attention seeking behaviors. The authors noted that it is not clear whether teachers were able to accurately recognize and intervene with teacher attention seeking behaviors, or if the most commonly used intervention str ategies just happened to address the function of seeking teacher attention. Although FBA may be conceptualized as more of a way of viewing behavior problems, rather than a specific requisite step of intervention development, the fact that teachers often do not give consideration to behavioral function is some cause for concern, especially given its recent inclusion in the reauthorization of IDE I A ( 2004 ). Furthermore, Myers and Holland suggest that teachers may not be deliberate in their choices of interv entions, taking a cookbook approach where they merely select from a list of commonly accepted interventions for a particular problem. Thus, interventions might not be appropriately individualized and may be less than effective. To understand both teache rs knowledge and use of classroom interventions in combination, Wilson et al. (1998) used two structured interviews with 20 general education teachers. Participants were given a standardized vignette describing a disruptive student and asked to offer as m any intervention s uggestions as possible to effectively manage the students behavior. Interviewers also encouraged teachers to describe strategies in as much detail as they could, and teachers were given examples of specific and nonspecific descriptions. In the second interview, teachers were asked to recall a student they had taught who eventually was identified as mildly handicapped. Again, teachers were asked to recall all intervention strategies they had employed as specifically as possible, from th e time they first noticed the problem (prereferral)
25 through prereferral intervention team, referral, and postreferral stages of the process. Participants were to describe all classroom interventions, data collection and documentation methods, and types of people with whom they consulted. It should be noted that the authors of this study explained their decision to use an interview methodology by suggesting that survey, checklist, and rating scale procedures might impose a priori assumptions on data and li mit or distort teachers responses. The interview method was selected in order to allow teachers the greatest freedom in response possible, while still allowing for quantitative analysis of data. In the standardized vignette portion, Wilson et al. found t hat teachers generated an average of 9.6 interventions each ( SD = 3.7). Over half of the intervention suggestions were behavioral (54%), 23% were instructional, and 13% involved manipulation of the classroom environment. Teachers responding to the vignet te interview generally did not use specific language to describe their intervention suggestions; only 10% of all intervention suggestions were considered highly specific. In the referral case, teachers reported an average of 9.2 interventions each ( SD = 2.58), 81% of which were implemented by the teachers themselves. Types of intervention used were similar to the case vignette, with behavioral strategies most frequently reported (34%). Intervention descriptions for the referral case also were nonspecifi c. Fifty nine percent of teacher mediated interventions were described in general terms, and only 13% were considered highly specific. Descriptions of interventions implemented by other individuals (e.g., support staff, parents, etc.) were even more vagu e; 71% were considered low in specificity. In detailing data collection procedures, 79% were rated as low in descriptive specificity, and almost all (94%) were considered haphazard in their approach. Teachers frequently commented that they felt under tra ined in the area of data collection and other areas of IA team functioning. Most importantly, teachers reported that the majority of the intervention efforts they
26 made occurred before a referral was ever initiated and that they rarely consulted with othe r professionals in this preliminary stage of the process. Wilson et al. (1998) concluded that teachers limited knowledge of interventions may hinder intervention plan development at IA meetings, and may lead teachers to have low expectations about teams abilities to generate viable intervention options. Furthermore, this paucity could impact teachers implementation and evaluation of interventions. The authors stated that school psychologists and other consultants in IA programs may need to function in an educative capacity, in order to enhance teachers knowledge of potential strategies available to them. The interview format of this study also may have impacted these results. Teachers may not have given an exhaustive list of interventions or descript ions of adequate specificity due to the demands of the interview situation. However, the findings described in Wilson et al. (1998) suggest a powerful limitation in the IA process that must be considered by all involved professionals. In addition, it is i mportant to consider the finding that teachers most often implemented interventions prior to referring the student to the IA team Although IA processes have been developed to support teachers implementation of interventions and to render educational dec isions in a team format, Wilson et al.s data suggest that teachers still function independently in intervention development and implementation for a considerable portion of this process. This is consistent with Bahrs finding (1994) that teachers are ofte n the de signers and almost always the implementers of classroom based interventions. In light of this finding, Wilson et al.s conclusion that teachers are lacking in intervention knowledge becomes even more significant. Although the supportive, educativ e functions of IA teams may assist teachers in the intervention process once referral is initiated, data from Wilson et al. beg the question: Do teachers have the necessary and sufficient intervention skills to successfully remediate student problems on th eir own prior to the point of referral? Furthermore are IA teams
27 receiving referrals that may have been unnecessary if teachers had stronger intervention skills? Flugum and Reschly (1995) investigated the typical prereferral intervention and concluded that professional best practices for intervention design and implementation are frequently overlooked. The authors found that behavioral definition of the referral problem in objective, measurable terms, direct measure of student outcome, systematic and d etailed intervention plan, graphic representation of results, or comparison of student progress to baseline levels were often lacking in prereferral interventions. Flugum and Reschly also found that interventions that did contain these elements were perce ived as being more successful, although actual student data were not reviewed. Findings from Flugum and Reschly suggest that intervention attempts are not systematic or well planned, and that teachers may be more random or capricious in their selection of intervention strategies than professional standards would dictate. This study directly contradicts the conclusions made by Pugach (1985), and further questions teachers abilities to effectively intervene with student problems. A literature review by Gre sham (1989) addressed the specific issue of treatment integrity, or the degree to which an intervention plan is implemented as intended, and argued that integrity of interventions is a critical component that is often overlooked in the empirical literature Those few studies reviewed indicated that interventions without operational definitions of behavior or precise intervention plans were less likely to be implemented with high levels of integrity. Among the technical issues that support treatment integr ity, Gresham cites the specification of all intervention components in exact, behavioral language as crucial to the success of an intervention. Adequate definition of intervention components allows them to be measured accurately, which facilitates both fo rmative and summative evaluation processes. Gresham distinguished between three potential levels of specification (global, intermediate, and molecular), and concluded that intermediate specificity is the optimal level for which interventionist should aim when designing treatment plans. Although molecular
28 descriptions of intervention plans are ideal for determining a functional relationship between intervention and behavior change, Gresham notes that interventions at this level can be met with resistance b y those who are required to carry them out (i.e., teachers). Thus, intermediate specificity provides adequate information at a depth that is reasonable to all participants. Several studies have been conducted to determine which classroom interventions tea chers frequently use to assist difficult to teach students. A qualitative study by Mamlin and Harris (1998) identified the most common prereferral interventions used by several teachers at one suburban school in Maryland. Among the most common were behav ioral management (e.g., individual behavioral contracts, daily report cards), academic/instructional modifications (e.g., extra practice or manipulatives, individualized instruction or explanation, change of instructional grouping), and help from others (e .g., outside counseling, parent assistance, consultation with intervention assistance team). Although data were only collected from teachers at one school, limiting external validity, Mamlin and Harriss findings are representative of those from studies w ith larger and more diverse samples. Sevcik and Ysseldykes (1986) survey of 105 general elementary educators yielded a list of more than 90 interventions to assist students with behavior problems. Approximately one third were specific behavioral interven tions. Other commonly described strategies included discussion/conference with other professionals or parents and instructional modifications. Two thirds of respondents supported the use of teacher mediated interventions, as opposed to those implemented by other professionals. Sevcik and Ysseldyke determined that general educators were, by and large, willing to try interventions in their own classrooms, although consultation may be necessary for the most effective implementation. A similar study by Brow n, Gable, Hendrickson, and Algozzine (1991) surveyed 201 teachers regarding their intervention practices. They found that consulting with other professionals was most common, followed by parent conferences, behavior management
29 techniques, and individual i nstruction. Cooperative learning and peer tutoring were used least frequently. Teachers in this study reported they were willing to work with teams and outside consultants when necessary. Determinants of Teachers Intervention Knowledge: Training Although IA programs were initially conceptualized as a mechanism for helping teachers perfect their interventionary skills in lieu of more formal inservice training (Rathvon, 1999), recent best practices recommendations for conducting intervention assistance team s suggest that teachers need preliminary, preservice training in designing, implementing and evaluating interventions prior to participating in IA teams and programs (Kovaleski, 2002). Some academics have also suggested that IA should be supplemented with inservice training (Logan & Stein, 2001; Nelson et al., 1991), to provide teachers with a constant source of information on the latest empirically supported practices in classroom based interventions. Unfortunately, there is presently very limited inform ation available on the extent to which general education teachers are trained in intervention strategies for difficult to teach students. Numerous studies have bemoaned the lack of preparation av ailable to teachers in training with regard to classroom bas ed interventions for students with a variety of problems (e.g., Newman, 1999; Wilson et al., 1998; Worthington, Wortham, Smith, & Patterson, 1997). Within the relevant teacher training literature, few studies directly address the goal of preparing teacher s to design and conduct interventions within the general education classroom. For example, Newman (1999) described a supervised intervention training experience in which teaching interns went into a classroom to work with individual students, targeting sp ecific behavioral or academic interventions. These interns outcomes were compared with those of a control group, who were not exposed to the training experience. However, interns skills in basic intervention elements (e.g., intervention planning, use o f reinforcement strategies, gathering of baseline and intervention data, etc.) were not included as outcome variables in the Newman (1999) study; rather, the investigation
30 examined the effect of the intervention exercise on interns own perceptions of self efficacy and locus of control. Although this study is innovative in its experimental design and applied approach to this topic matter, it does not provide useful information about how teachers skills and abilities actually changed as a result of partici pation in the intervention experience. With regard to preservice training, Wood et al. (1990) found that although roughly 25% of states (13) require or recommend intervention assistance programs, only three reported that training in this area was provide d at the preservice level by universities and colleges. Kovaleski (2002) also state d that the complex skills necessary for effective prereferral intervention are rarely included in typical teacher training programs. Although many training programs may br iefly introduce concepts of observation and data collection to teachers in training, it is rare that teachers have the opportunity to attempt, under supervised instruction, a targeted intervention in the naturalistic setting of the classroom (Newman, 1999) Newmans research (1999) provides one notable exception, but does not offer results that are useful in describing how developing teachers intervention knowledge might be facilitated by such a hands on experience. In addition to formal and informal trai ning experiences, teachers may learn about recent innovations in intervention strategies via research and professional literature available in the form of journals and texts. Unfortunately, teachers do not often read research studies about classroom manag ement; when they do so, they report that the strategies used in research investigations do not often appear to be feasible for use in their own classroom situations (Malouf & Schiller, 1995; Viadero, 1994). In an innovative attempt to inform teachers about empirically supported classroom strategies, Logan and Stein (2001) developed the Research Lead Teacher Model. The goal of the 3 year program was to bring the research based methods of instruction and classroom management often found in special education, such as positive behavior support and applied behavior analysis, into general education classrooms of one school. Teachers participated in building wide staff development groups about behavior
31 management and positive behavior support, in which they had an opportunity to design positive interventions, collect progress monitoring data, and discuss the implementation of their interventions. These groups were facilitated by a Research Lead Teacher (RLT), a full time teacher at the participating school who had extensive familiarity with research on behavioral interventions, special and general education classrooms, and mentoring teachers in a consultative relationship. In addition to staff development groups, teachers could also request observations and i ndivi dual sessions with the RLT to develop more specific intervention recommendations and plans for ongoing classroom issues. Teachers reported a wide range of improvements in student behavior, rating 89% of all interventions developed through the RLT as succe ssful. Additional qualitative information regarding teachers perceptions of the RLT program indicated a generally positive response. Similarly to the Newman (1999) study, however, teachers skills in intervention development and implementation were not assessed in either pre or post RLT phases, and follow up data on teachers long term intervention practices were not collected. As a result, it is unclear to what extent teachers knowledge of interventions was improved through their association with thi s program. Given the importance of this role for teachers, especially in states in which IA processes are mandated, it is surprising that teacher knowledge of interventions has garnered so little attention in the extant literature. Furthermore, it is int eresting that standards for training teachers to provide classroom modifications for difficult to teach students have not been better delineated in either theoretical or empirical treatments of this topic. Determinants of Teachers Intervention Knowledge: Building level IA Practices Although an emphasis on determining teachers intervention practices is warranted, especially in light of the aforementioned literature, it is also imperative that teachers intervention efforts are considered within the greate r context of the school in which they work. Teachers do not operate in a vacuum and, as part of a complex educational system, their efforts on behalf of students may be
32 impacted by numerous variables external to their own knowledge and skills. Kovaleski (2002) notes that psychologists skills in consultation and intervention development cannot be analyzed independently of the system factors that contribute to successful prereferral intervention programs. Similarly, any investigation of teachers skills m ust consider the impact that existing prereferral programs within their schools may have on intervention knowledge and practices. In a controlled experimental design, Pugach and Johnson (1988) investigated the effect of participation in a collaborative, p roblem solving process on the tolerance, accuracy of problem identification, and effectiveness of the prereferral interventions of teachers. Results from this study indicated that participation in an IA like program increased teachers tolerance for a bro ad range of cognitive abilities, improved 91% of their target behavior definitions, and generated apparently successful interventions (teachers perceptions of effectiveness were reported in lieu of data on actual behavior change). These findings suggest that teachers intervention skills and perceptions regarding IA programs can be impacted merely by participating in such teams. Kovaleski (2002) summarized factors that have been found to be related to successful prereferral intervention programs at the b uilding level. These IA best practices can be conceptualized as either system factors (characteristics of school environments that facilitate IA programs), or process factors (procedural factors that help IA programs to realize meaningful outcomes). Alth ough these conditions are described by Kovaleski in the context of multidisciplinary, building wide IA teams, it can be argued that the presence or absence of these conditions may have an impact on the way school personnel (i.e., general education teachers ) individually conceptualize and approach intervention efforts for difficult to teach students. System factors. Team format : Research has demonstrated that when a general education teacher refers a student for intervention assistance, teams of school personnel are best able to successfully support
33 the teacher (Rathvon, 1999). In addition, creating building level IA teams aids in encouraging a mission and sense of team enthusiasm (Kovaleski, 2002). Principal leadership : Although initial publications i n the area of IA recommended against the inclusion of principals on building level teams (Chalfant et al., 1979), more recent research has found that administrator involvement is a crucial component in gaining teachers acceptance of IA teams and their act ivities (Kruger, Struzziero, Watts, & Vacca, 1995). Mandating prereferral intervention : Kovaleski (2002) noted that system level adoption of IA programs is facilitated by state or district mandates, as administrators and personnel are essentially forced to direct their energy toward IA activities. Furthermore, resources necessary to conduct such programs are more readily available in areas where IA is required. Assignment of staff : IA teams are, by definition, multidisciplinary in nature, requiring the e fforts of a variety of school personnel in order to be successful. However, teams are optimally effective when one or more school staff are assigned either part or full time to facilitating IA activities. In Hillsborough County, FL, this role is often f ulfilled by the Exceptional Student Education (ESE) coordinator, although research indicates that school psychologists and guidance counselors are often asked to devote their time to this professional responsibility. Ensuring accountability : At the micro ( student) level, accountability of IA teams can be indexed by examining formative and summative data from interventions developed for identified students. However, accountability at the macro (system) level is equally important to ensuring that the IA team is accomplishing broader administrative and educational goals.
34 School wide indicators such as number of students served by the team, number of students referred for special education evaluations, and number of students retained can be useful indices of t he IA teams overall impact on the school as a system. Training : Kovaleski (2002) underscored the importance of providing preservice and inservice training for all members of IA teams, especially teachers, in the fundamental skills of collaboration/consult ation, curriculum based assessment, behavioral assessment, and relevant instructional strategies or intervention strategies available to assist students. In addition, IA teams should go beyond inservice trainings to offer team members in vivo practice and professional mentoring in conducting these activities. Process factors. Creating a data based practice : As mentioned previously, collecting data on the actual interventions recommended and implemented by the team is crucial to monitoring the efficacy of both the individual intervention and the IA team as a whole. This need is supported by the research of Flugum & Reschly (1994), who found that data collection (both baseline and ongoing intervention data) contributed to the perceived success of prereferr al interventions. Selecting research based strategies : Given the increasing professional impetus to link assessment data to intervention strategies, selection of such interventions cannot be an arbitrary process (Batsche & Knoff, 1995). As indicated by My ers and Holland (2000) and Logan and Stein (2001) demonstrated, teachers cookbook approach to selecting interventions may be to the detriment of difficult to teach students. Successful IA
35 teams select interventions that have been demonstrated in resear ch to show quick and effective results, before considering more lengthy intervention or assessment activities (Kovaleski, 2002). Establishing the intervention : The multidisciplinary nature of IA teams should not be limited to intervention development. Rat her, team members should go a step further to actually work with the student either in a group format or individually, assisting the teacher in the classroom implementation of the intervention. The involvement of additional team personnel works to ensure treatment fidelity and to model correct implementation to the general education teacher (Kovaleski, 2002). Incorporating the intervention : One of the most common refrains of teachers with regard to classroom based intervention activities is, I dont have time for all this! Teachers are fiercely protective of their already limited time, and intervention acceptance has been found to be related to the demands the strategy places on the teacher (Gresham, 1989; Inman & Tolefson, 1988). One way to ensure tha t teachers implement the intervention to the greatest extent possible is to design it such that it fits easily within the teachers day to day routine, rather than simply recommending a strategy and leaving to the teacher any specific plans for incorporati ng it into the classroom. Albin, Lucyshyn, Horner and Flannery (1996) describe this premise in the positive behavioral support literature as contextual fit which is defined as the congruence between intervention plan features and those variables that seriously affects the development and implementation, and therefore the effectiveness, of those plans.
36 Involving parents : The intensive involvement of parents in IA activities has been found to be a characteristic of highly successful IA programs (Kovalesk i, 2002). Parent input can greatly inform the development of interventions, and can allow for school or classroom based interventions to be extended into the students home environment. Screening for further evaluation : One of the IDEA requirements for disability determination is to rule out the possibility of lacking instruction. Kovaleski (2002) argues that IA teams can be useful in testing this hypothesis by generating intervention plans that include instructional strategies or accommodations that a re feasible in the general education classroom. However, in instances in which interventions do not successfully address student problems, a lack of instruction is ruled out and screening/evaluation is required to determine both the precise needs of the s tudent and the various resources available to assist the student. Given that IA programs are capable of functioning in an educative capacity (Nelson et al., 1991; Safran & Safran, 1996; Wilson et al., 1998), it can be hypothesized from the preceding list that the extent to which schools adhere to these IA best practices may have a positive effect on the ways teachers conceptualize and attempt interventions. When IA teams model effective, research supported supports for intervention development, teachers m ay be more likely to have greater knowledge in interventions. Purpose of Study The purpose of the present study was to replicate a portion of the research of Wilson et al. (1998), which conducted interviews with general education teachers to assess their knowledge of classroom based interventions. Specifically, this study used the structured interview and vignette portion of the
37 Wilson et al. study to examine the interventions teachers suggest in response to a multifaceted, hypothetical student problem However, the present study extende d their approach in that, in this investigation, teachers individual professional characteristics (including training experiences), as well as the IA practices of the schools in which they work, were measured to determin e their degree of relatedness with teachers knowledge of interventions. Research questions to be addressed in this study were as follows: 1. What is the average number of interventions teachers offer to address a hypothetical classroom behavior problem? 2. Ho w specific are teachers in descriptions of interventions/strategies they would use in their classroom (average specificity rating per teacher)? 3. What is the likelihood that a teacher will suggest a given type of intervention (e.g., instructional, behavioral etc.)? 3a. What 2 or more intervention categories, if any, are likely to be suggested by the same teacher (i.e., what is the probability that a given teacher will suggest both intervention type x and intervention type y ?) 4. What is the relationship betwee n years of teaching experience and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention (e.g., instructional, behavioral, etc.)? 5. What is the relationship between the number of times the teacher has participated in IA meetings and the number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? 6. What is the relationship between teachers referral to eligibility rate and number of interventions/strategies suggested, specificity of
38 interventions/strategies descriptions, and the likelihood that a teacher will suggest a given typ e of intervention? 7. What is the relationship between training experiences and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? 8. Wh at is the relationship between intervention assistance (IA) practices of the participants school and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? This study offered two major contributions to the present intervention literature base. First, by refining a method by which school psychologists might assess teachers knowledge of classroom based interventions, this study helps l ay the groundwork for further research examining elements of teachers intervention knowledge (i.e., types of interventions with which they are most familiar, average levels of familiarity/comfort with interventions). Secondly, characteristics that may be related to teachers intervention knowledge (i.e., years of teaching experience, exposure to training activities, schools adherence to IA best practices, number of children referred for evaluation, number of children found eligible for service) are sugges ted. Such teacher/school characteristics can be considered in the development of constructive recommendations for changes in preservice/inservice training, building level approaches to IA activities, and future theoretical or empirical analysis of teacher s involvement in IA programs. In addition, it is hoped that this research may have some level of practical impact, shaping the ways present teachers assist difficult to teach students in the general education classroom.
39 Chapter III Research Methods P articipants and Setting Twenty nine second and third grade general education teachers from six elementary schools in Hillsborough County, Florida were recruited to participate in this study. Second and third grade teachers were selected as participan ts in this study because rates of referral for suspected disability tend to be highest in these years of education (Ysseldyke et al., 1983 ). In earlier years (i.e., kindergarten/first grade), students academic and behavioral skills are still developing a nd teachers are often encouraged to wait until potential problems become more pronounced. In later years (i.e., fourth/fifth grade), most of the students with relevant problems have either been placed in special education classes, had Individualized Educa tion Plans (IEPs) developed for their general education programs, or had their needs addressed through intervention assistance (IA) programs in their schools. Sample size was limited a priori roughly 30 participants, in consideration of the time and effor t necessary for the researcher to contact participants, conduct individual interviews, and transcribe/analyze interview sessions. Sampling procedure. In the proposal for this project, s chools were to be selected for inclusion using a stratified random s ampling procedure. All elementary schools in the Hillsborough County School District were ranked on the basis of a risk index, which was operationally defined in this study as an equation that averages each schools percentages of 1) students receiving free or reduced cost lunch, 2) students receiving special education services, and 3) teachers without advanced degrees These three demographic variables, obtained from the Florida School Indicators Report (available online at http://info.doe.state.fl.us/fsir/ ) were selected because they provide a measure of
40 student socioeconomic background, incidence of student disability within the enrolled population at a school site and level of teacher training that can be detrimental to student outcomes. After ranking schools based on the risk index, all Hillsborough County schools were divided into five quintiles and one school was randomly selected for participation from each quintile, using a random number tab le. This was done to allow the sample of schools to roughly approximate the variability seen in the population of Hillsborough County Schools and to avoid the random variability seen in very small samples. The researcher contacted the assistant principals of each of the selected schools by telephone to discuss the proposed study and solicit initial support for data collection in their school buildings. If assistant principals declined to allow their school to participate, the original plan called for anot her school to be randomly selected from the same quintile to take its place. Unfortunately, the sampling procedure had to be modified due to unforeseen complications in obtaining administrator support for data collection in the randomly selected schools. The researcher followed the proposed sampling procedure and spent several months making phone calls to assistant principals, often without receiving a return phone call. The researcher used a general rule of three attempts at initial contacts before aba ndoning efforts at a given school and selecting another school from the same quintile, and several schools were ruled out as participating sites through this procedure. In addition, several assistant principals declined to participate, citing low teacher interest or too many conflicting demands on teacher time (e.g., other research projects or school initiatives, statewide assessment, etc.). These difficulties in obtaining building level support for data collection led to a change to a convenience sampli ng method, which was approved by the supervising committee. Three elementary schools in Hillsborough County where first year school psychology students were participating in an observational practicum were recruited as data collection sites. These sites were recruited because administrators and staff were familiar with the USF School Psychology
41 Program and were hoped to be more receptive to the research project. Furthermore, first year practicum students volunteered to serve as data collectors to complet e a research training requirement, and using their current practicum sites considerably facilitated the process of contacting and scheduling teachers for interviews. In addition to the practicum sites, three other local schools were recruited on the basis of previous contact with the USF School Psychology program or proximity to the university. This resulted in a total of six school sites from which teachers were recruited. Despite the need to use convenience sampling, demographic characteristics of the resulting sample of schools were compared using the original risk index to assess variability among participating schools (see Table 1) It should be noted that risk index rankings are available for only 5 of the 6 schools participating in this study. S c hool B opened in the 2003 2004 school year and because the Florida School Indicators Report used 2002 2003 data, no data were available to indicate school Bs ranking among the participating schools. Of the schools participating in this study, risk index rankings ranged from a high of 15 th out of 129 elementary schools in the county (school A) to a low of 91 st out of 129 (school D). After dividing ranked schools into quintiles, it was found that schools represented the first (schools A and E), third (scho ol C) and fourth (schools D and F) quintiles, indicating a broad range of risk as measured by percentage of students with disabilities, students receiving free or reduced price lunch, and teachers without advanced degrees. No schools from the second and f ifth quintiles participated in this study. Although equal representation from each of the quintiles was not achieved, the participating schools do reflect the variability seen within Hillsborough County schools overall demographics. Limitations to exter nal validity created by this change in sampling procedure are addressed in Chapter 5. Once approval was obtained from each assistant principal, the researcher acquired a list of all second and third grade teachers in each school. From this list, six part icipants (3 second grade and 3 third grade teachers) were selected at
42 random using a random number table. Teachers were contacted in one of two ways: (1) a flyer describing the purpose and basic method of the research study, along with lines for providin g contact information were placed in teachers mailboxes (see Appendix A), or (2) teachers were approached personally and verbally told the same information as on the flyer by a data collector or the main researcher. Both forms of contact were used to sch edule a specific interview time with the teacher. As a form of incentive to participate, teachers were offered a $15 gift card to Staples to compensate them for their time, and this was stated both in the flyer and in verbal contacts with data collectors. If a teacher declined to participate in the study, another teacher from the same grade level was selected from the list provided by the principal until the appropriate number of participants from each school and grade level had been recruited. If less t han six teachers from a given school were available to participate, additional teachers were recruited from another participating school until the target number of participants ( N =30) was reached. Due to a last minute cancellation, however, the total samp le size for this study was 29. Table 1 Demographic Characteristics of Participating Schools School a % Students with Disabilities % Students Receiving Free/Reduced Price Lunch % Teachers Without Advanced Degrees Risk Index County Rank b Quintile A 14.2 21. 3 70.6 35.4 15 1 C 22.2 46.2 71.4 46.6 56 3 D 18.5 75.2 70.4 54.7 91 4 E 14.2 29.7 63.5 35.8 19 1 F 19.3 68.8 71.1 53.1 79 4 a School B was a new school in 2003 2004 Because risk index rankings were developed using 2002 2003 school year data, no indi cators were available for school B. b Rank is out of 129 elementary schools in the School District of Hillsborough County, FL Setting. Teachers were individually interviewed in a private location at their school at a time most convenient to them. When ever possible, interviews
43 took place in a conference room or private office; however, when teachers were interviewed in their own classroom, the session took place at a time when he/she was not responsible for supervising or instructing students. Regardle ss of exact location, all interviews took place behind closed doors to ensure privacy and confidentiality. Materials and Measures Prior to initiating data collection, a pilot study was conducted to assess the validity and utility of the following measur es. Although this phase of the study is described in greater detail under Preliminary Data Collection in the Procedure section of this chapter, modifications made to the measures and materials adapted from Wilson et al. (1998) are mentioned in the descr iptions below to clarify how study instruments were improved to better meet the needs of the current research objectives. A summary of outcomes from the pilot phase can be found in Appendix B. Demographic questionnaire. An 18 item demographic questionnair e was administered to each participant (see Appendix C). This instrument was used to obtain information about the participants age, gender, and race, as well as grade level currently taught, years of teaching experience, participation in IA teams, referr al history, training experiences, and schools adherence to IA best practices. This information was used to describe the sample as well as to analyze intervention descriptions according to relevant teacher characteristics (i.e., grade level taught, years of experience, IA participation, etc.). To ensure confidentiality, participants were not identified by name on this form. Modifications to the questionnaire resulting from the pilot study were minor and primarily consisted of wording changes to improve pa rticipants understanding of items. For example, pilot participants responses suggested that not all schools, whether public or private, have building level problem solving teams. This was problematic, as many questionnaire items referred to such teams in a way that assumed all elementary schools used these teams with regularity. As such, it was decided to add an item that specifically asked if such
44 a team exists. All references to such a team were modified to school based problem solving team, rathe r than Child Study Team or intervention team, to overcome the variability in terminology that may exist from school to school in Hillsborough County. Several other minor changes were made to make items clearer or to make the questionnaire less demandi ng for teachers (e.g., asking them to recall the past two years of referrals, rather than three, as teachers tended to struggle to recall that far back). Interview instructions and standardized vignette. Prior to administration of the vignette, the researc her read a set of instructions to the participant explaining the task at hand (see Appendix D). The instructions informed the teachers that they would be asked to read a vignette that described a hypothetical student related behavior problem in a general education classroom and subsequently would be asked to describe all the ways they knew to help the student achieve the goals presented. Before they began responding, teachers were informed that their intervention descriptions should be as descriptive and specific as possible. The vignette, describing a classroom based problem exhibited by a third grade student (John), was taken from the Wilson et al. (1998) interview protocol with permission of the first author (see Appendix E). It consisted of three pa ragraphs describing Johns academic difficulties in math and reading, behavioral problems exhibited in the classroom, and peer responses to Johns behavior. The vignette also listed two goals that the participant has hypothetically set for John: 1) to st op talking out in class and 2) to stay in his seat. All teachers received the same vignette, which was printed on an 8 x 11 sheet of paper and handed to participants to read individually. Data from the pilot phase of the study revealed the need for two modifications to the script in order to maximize participants responses. The first was a procedural change that clarified the number of interventions described to the data collector and/or transcriptionist. Although pilot participants had many inter vention ideas in response to the vignette, it was difficult to determine where one intervention ended and another began. This complicated the process of
45 entering, coding, and counting each individual intervention. To facilitate responding in a way that d esignated clearly distinct interventions, a modification was made in the script to prompt teachers to hold a small token while describing each intervention idea and to drop it into a plastic cup when they were finished describing it. When listening to aud iotapes of interviews for transcription, the sound of the token dropping into the cup provided a cue that marked the break between intervention ideas. This change in procedure facilitated data entry and analysis considerably and served as an additional pr ompt for providing detailed descriptions due to the statement unfortunately, this means you cant go back and add to an idea once youve dropped it into the cup, so try to describe your ideas as completely as possible before you drop it. The second mod ification to the script was made to enhance the specificity of participants responses. The original version of the script used in the pilot phase provided a brief prompt and example of specific responses. Pilot study data, however, indicated that partic ipants responses were typically brief and low in specificity. The script text was analyzed to determine if there might be a better way to elicit highly detailed responses from participants. The original script text used the following prompt: Before we begin, I need to ask that you try to be as specific in your descriptions as you can. For example, if I asked you to describe the types of things you might do to help John to succeed in the classroom, and you said, I could change the mode of instruction that would be too general and would not give us the type of information we need. However, if you said, I could shorten his daily math assignments by cutting them in half , and so on, that would give us more of the kind of information we need. Again, g ive us as much detail as you can. Upon further inspection, it became clear that the above example of a more specific response would actually have only received a rating of a moderate specificity. As such, examples of low medium and high specificity r esponses
46 were incorporated in the script with additional emphasis on the importance of providing highly detailed descriptions. Note that the text below also includes bracketed prompts to data collectors instructing them on precisely when and how to model the use of the tokens, referred to here as poker chips. I also want to remind you to be as specific as possible in your descriptions. Give as much detail as you can Try to describe what you would do in a way that is so clear that I, as another educa tional professional, would know exactly how to implement your idea just from hearing your description. Let me give you some example responses that provide low, medium and high levels of detail. While I give these examples, I will show you how to use the poker chips like I just described. If I asked you to describe the types of things you might do to help John succeed in the classroom, and you said, [pick up a chip] I could change the workload, [drop chip into Goal 1 cup and pause for 2 seconds] that w ould be a low detail response. That is too general and doesnt tell me exactly how you are planning to help John. If you said, [pick up a chip] I could shorten his daily math assignments, [drop chip into Goal 1 cup and pause for 2 seconds] that would give a medium amount of detail I have a better idea of what you want to do, but Im still not completely sure how you would do it. Finally, if you said, [pick up a chip] I would take Johns math worksheets and cut them into strips of five problems each When he finishes one strip, he will come up to my desk, and I will tell him hes doing a good job and give him another strip. This will break down his work into smaller chunks and allow him to get a brief rest and some praise in between sets of problem s. [drop chip into Goal 1 cup and pause for 2 seconds], this would be a highly detailed response. I would know exactly how
47 to implement this idea based on your description. This is the kind of response were looking for. Beyond these two modifications, no other changes to the script were made. The vignette text also remained unchanged after the pilot phase. Interview coding form. Data obtained from the interview was typed into a form that facilitated coding of the salient characteristics of teachers r esponses (see Appendix F). Each distinct intervention suggestion was typed into a separate text area on the form, and below each text area were categories for the type of intervention described (e.g., instructional, behavioral, etc.) and the degree of its specificity (e.g., low, medium, high). More information about the definition of each of the codes for intervention type and specificity is provided below. Finally, the number of intervention suggestions was counted by summing the text boxes used to desc ribe each discrete intervention. No changes were made to this form following the pilot study with the exception of adding new intervention type codes as described below. However, in the data entry phase, a summary page was added to the coding form, which provided all relevant information to be entered for analysis (number of interventions described, presence and frequency of hypotheses, frequency of intervention types, specificity ratings for each individual suggestion, mean specificity rating by intervent ion type, and overall mean specificity for the participant). This summary sheet is the last page in Appendix F. Code definitions A copy of the code definitions, including examples of suggestions for each dimension, can be found in Appendix G. Procedures for coding specificity of descriptions were originally adapted from the work of Gresham (1989) and were used by Wilson et al. (1998). The original coding procedures used in the pilot phase included the same descriptions of low moderate and high speci ficity responses as those used by Wilson et al. (1998) They were as follows: (1) low specificity : descriptions consisting of nonspecific or vague recommendations (e.g., I could use one of those B Mod things, or I could write stuff down); (2) moderate specificity : description contains some, but
48 not complete, detail (e.g., A volunteer could help him with reading in the afternoons, in the library, or He could earn chips if he stays in his seat for the whole lesson); and (3) high specificity : descripti on demonstrates a detailed plan for assisting the hypothetical student (e.g., During the recess period every other day, John and a paraprofessional would sit in the Reading Corner of the classroom and John would read aloud for 20 minutes. The para could keep track of errors and words read correctly per minute, and she and John could chart his progress on a special graph). Based on the changes to the interview script as previously described, however, the coding procedures were modified after the pilot pha se to give clear examples and non examples of specificity. Specificity of responses was operationally defined as a description that provided enough detail that another educational professional could implement the intervention idea without further informat ion. Following the pilot phase, each intervention recommended by teachers was coded as follows: (1) low specificity if it consists of nonspecific or vague recommendations, and/or responses where the intervention suggested could not be implemented based on the current description alone because more information is needed (e.g., I could change his workload); (2) moderate specificity if the description contains some, but not complete, detail, and/or the intervention suggested could be implemented but addi tional details would need to be provided; (e.g., I could shorten his math assignments); or (3) high specificity if descriptions demonstrate a detailed plan for assisting the hypothetical student that could be implemented on the basis of this description alone (e.g., I would take Johns math worksheets and cut them into strips of five problems each. When he finishes one strip, he will come up to my desk, and I will tell him hes doing a good job and give him another strip. This will break down his work into smaller chunks and allow him to get a brief rest and some praise in between sets of problems). To code the nature of interventions, a scale adapted from Ysseldyke, Pianta, Christenson, Wang, and Algozzine (1983) was used. A similar version of
49 the scale was also used to analyze responses in Wilson et al.s (1998) study along the following types: (a) instructional (e.g., individual help, restating directions), (b) behavioral (e.g., behavior modification principles), (c) classroom structure (e.g., mo ving students seat), (d) interdisciplinary support (e.g., pre taught vocabulary with the resource teacher, counseling with the school counselor), (e) information gathering (e.g., checked the students cumulative file, called parents, and (f) materials (e. g., audio visual tapes, manipulatives). Data from the pilot study indicated the need for three additional categories because several intervention ideas could not be coded using the existing codes described above. The three new categories, (g) communicati on with student, whole class, or parent/family (e.g., discussing with the student, class, or parents about the importance of not calling out in class), (h) emotional/social support (e.g., work on building the students self esteem and achieving success), a nd (i) compound (e.g., developing a behavioral contract that is monitored by the guidance counselor behavioral and interdisciplinary support), made it possible to code all responses from the pilot study. Two existing intervention types, behavioral and cla ssroom structure also were modified on the basis of pilot data. Some intervention strategies consisted of prompts or cues (verbal and nonverbal) from teacher to student, and there was some confusion as to whether these could be classified as form of teac her student communication or as an antecedent behavioral cue. As such, the behavioral category was redefined to be consequence oriented responses to behavior, such as reinforcement, punishment, extinction, time out, behavioral contracts, etc. All cues f rom teacher to students were coded as communication The definition for classroom structure was modified to allow for interventions that included changes to students responsibilities/duties that impacted the level of structure in noninstructional activit ies (e.g., providing the student with opportunities to run errands as a way to be able to walk around more frequently) or changes to the classroom environment, including rules or policies (e.g., allowing the student to stand at his chair if he can demonstr ate that he is working productively on his assignment).
50 Finally, an additional note was added to the coding form about noting the presence of hypotheses about potential causes of behavior in teachers responses. Hypotheses were occasionally apparent in pi lot participants responses, and it was decided that they would be highlighted in the body of the text and counted for later analysis. Procedure Protections of confidentiality and informed consent. Prior to data collection, approval to conduct this stud y was sought and obtained from both the Department of Assessment, Accountability, and Evaluation of the School District of Hillsborough County (SDHC) and the Institutional Review Board (IRB) of the University of South Florida. In additional, a letter of s upport for contacting teachers and conducting research on school grounds was obtained from school principals or assistant principals. At the outset of each participants interview session, the data collector used the interview script in Appendix D to thoro ughly describe the purposes of the study and provided assurances of confidentiality. The teachers were informed both orally and in writing that they were being audiotaped for the purposes of recording, transcribing, and analyzing their responses. Teacher s also were told that only the data collectors (including the primary researcher) and the major professor would have access to these tapes, that they would not be labeled/identified using participants names (rather using number/letter combinations), and t hat the tapes would be destroyed upon completion of the study. Each teacher was given a consent form containing all of this information, which was signed in the presence of a data collector prior to his/her participation in the study. Teachers also were a ssigned a participant code that included an individual participant number, a school code, and a grade level code. This code was used for the purposes of identifying teacher responses on audiotapes, interview transcripts, and all other research records. F or example, a hypothetical participant Mrs. Smith from Apple Elementary was noted on all relevant
51 documentation as Participant 4A2, indicating that she was Participant #4 from School A, teaching grade 2. By using this coding system, no participants n ame or school appeared on any research documentation or records. Preliminary data collection (pilot study phase). To gather more information about the measures proposed for use in this study, a pilot study was conducted. In particular, the pilot study sought to address the following research questions: 1. Do the questions in the interview and questionnaire elicit the appropriate/desired responses from teachers? 2. Are there any aspects of the study that are confusing or unclear to teachers and which might req uire further explanation or changes in the research materials? 3. How much time does the interview session require of teachers (including informed consent procedures and administration of questionnaire)? 4. Are the proposed procedures for coding interview data s ensitive enough to detect salient features of teacher responses (e.g., number of interventions, specificity of descriptions, and intervention type)? IRB approval for the pilot study was sought in conjunction with approval for the main study in one comprehe nsive proposal. Several teachers were recruited from local schools, via convenience sampling, to participate in this pilot investigation. Identical procedures for obtaining principal consent for participation were followed in this phase of the study, wit h the only deviation being the explicit delineation of the goals of the pilot study. A total of 3 teachers participated in the pilot phase: one second grade teacher and one third grade teacher from a private school in Pasco County and one pre kindergarten teacher from a preschool in Hillsborough County. A fourth participant, an advanced graduate student in school psychology and instructor in classroom management strategies for preservice teachers, also participated in the pilot to confirm the utility of the procedure using tokens to
52 represent discrete interventions as they were being described. This participant was added because the pre kindergarten teacher, the first to use this new procedure, had a total of only five intervention ideas with a n overall mean specificity of 1.8. She attributed her responses to a lack of knowledge about interventions and limited teaching experience (less than two years), but there was concern that the token description procedure might have interfered with her ability to pr oduce numerous intervention ideas. To address this concern, the fourth participant was recruited because of his advanced knowledge of intervention strategies. It was assumed that he would have many intervention ideas in response to the vignette, and if h e also produced few ideas, then the token description procedure might indeed be inhibiting responding. The fourth participant, however, offered a total of 13 strategies for working with John, incl uding six hypotheses, and had an overall mean specificity o f 2.5. Based on the data from the fourth participant, it was concluded that the token description procedure did not significant ly inhibit responding and it remained in the interview script for use with primary study participants. Procedures for conducting teacher interviews, as well as for transcribing and coding the interviews, were nearly identical to the final phase of data collection and are described below. Pilot participants were also compensated with the $15 gift card to Staples. One important exce ption to the protocol was that, upon the conclusion of the interview session, pilot participants were asked additional open ended questions that addressed the pilot study research goals previously listed. Through these questions, participants provided qual itative feedback regarding the validity of proposed measures, which led to the previously described changes in measures and procedures. Answers to these questions can be found in Appendix B, which summarizes data from this phase of the study. Because dat a gathered in this phase of the research were preliminary, correlations were not calculated between teacher demographic information from the questionnaire and teacher responses to the interview. After analyzing data from the pilot phase, preliminary resul ts (Appendix B) were
53 approved by the members of the supervising committee via e mail, and necessary changes were made to the research protocol and instruments. A secondary goal of this initial investigation was to develop necessary materials for training a n advanced graduate student to serve as an independent rater for reliability checks. Blank copies of the transcripts resulting from the pilot study were developed for use in training exercises for the independent rater. In the interview sessions of the pi lot study, teachers were informed that transcripts of their interviews were intended to be used as future materials for training. As no one had been trained to serve as an independent rater during the pilot phase, interrater reliability was not quantified for the pilot participants data. Rather, to determine the sensitivity and reliability of proposed coding procedures for the pilot participants, the researcher and major professor both coded tapes from pilot sessions, compared coding forms, and discussed discrepancies until they were resolved. Main study data collection. Once changes to the research protocol/instruments were approved by the supervising committee, seven first year school psychology graduate students were trained by the researcher to serve as data collectors. Because data collectors had previously completed University of South Floridas IRB Human Subjects Training requirements, they were authorized to administer informed consent, as well as the demographic questionnaire, interview script, a nd vignette. Including the researcher, a total of eight people served as data collectors for this study; six first year students worked in pairs at three schools (schools A, B, and E), while the researcher paired with the seventh data collector at a fourt h school (school D) and collected data independently at the fifth and sixth schools (school C and F). Data collectors were trained in the administration of study materials and audiotaping of participants responses to the vignette, but only the researcher transcribed and coded data. Data collection in this phase began with the selection of schools and teachers for participation as previously described. During the teacher interview
54 session, informed consent procedures were followed as described previously. After written informed consent was obtained, each teacher was asked to complete the demographic questionnaire. The data collector then read the interview script, including a set of instructions describing the research task, and teachers subsequently were given the vignette to read themselves. All participants received the same vignette and the exact same set of instructions. Teachers were given as much time as they needed to read and think about the vignette and were prompted to take notes if necessary. When they indicated that they were ready to begin, the data collector described how to use tokens to describe separate interventions (referred to in the script as poker chips; see Appendix D for complete text). Participants were told that each token re presented an individual idea for helping the student in the vignette, and just as there are many, many things we can think of to help students, there are many chips in the bag. You do not have to use them all. Participants were then instructed to hold one token in their hand while describing an intervention, to be as specific and detailed as possible, and then to drop the chip in the cup when finished describing the idea. After reading this description, the data collector provided sample low medium and high specificity responses for the participant and modeled the use of the token procedure for marking discrete interventions as he/she gave the examples. After describing and modeling appropriate responses, the data collector asked teachers if they ha d any questions and then started tape recording the session at this time. Teachers then were prompted to begin with the first behavioral goal, stop talking out in class, and describe all the ways they could help the child achieve that goal. The data col lector provided one final prompt to be specific and use the tokens as modeled and then allowed teachers to provide their intervention ideas while tape recording their comments. When a participant stopped offering intervention suggestions, the data collect or asked, Is there anything else you can think of that could help John achieve this goal? This prompt was offered each time teachers stopped suggesting interventions until
55 each teacher indicated that he or she had exhausted ideas for the first goal. Th en, the procedure was repeated for the second behavioral goal, stay in his seat, until the participant again indicated that he or she could not think of any additional intervention strategies. At that time, the teacher was thanked for his/her participat ion, and the session was concluded by giving him/her the gift card. Data entry Subsequent to the interview session, all tapes of teacher interviews were transcribed by the researcher onto the interview coding form and analyzed for number of suggestions offered, specificity of descriptions (Gresham, 1989), and intervention type (Ysseldyke et al., 1983). Because each participants coding form was at least three pages long, it is not feasible to include all participants intervention descriptions. A repr esentative sample of five completed coding forms, showing the variability in the number of interventions, as well as the breadth (number of types suggested) and depth ( overall mean specificity), are provided in Appendi ces H L When transcribing interventio ns, the researcher used the sound of the token dropping into the cup as a cue to move to a new text box on the coding form for a new idea for the majority of intervention suggestions. Occasionally, however, teachers responses suggested that they did not understand the token procedure because they did not drop tokens as modeled by the data collector. One teacher (4A3), for example, described five discrete interventions before dropping the token in the cup. After her first response, the data collector pro mpted her to drop the token and she responded, I know, but did not drop the token in the cup and proceeded to describe additional ideas. She finally dropped the token when she indicated she had exhausted her ideas for helping the student achieve the fir st goal. This suggests perhaps this teacher misunderstood the purpose of the tokens to serve as a cue that she was ready to move on to the next behavioral goal in the vignette, rather than to indicate separate intervention ideas all targeting the same beh avioral goal. In this and similar such situations, data collectors typically provided a single reminder about how to use the tokens,
56 and if the participant continued to use them incorrectly, he or she was allowed to continue so as to prevent any inhibitio n of response by stopping to retrain the token procedure. When entering these data, a decision rule was used to determine how to break intervention ideas. First, if the participant provided verbal cues of discrete ideas such as, Another thing I would d o or On the other hand, these were used to signal the break between intervention ideas. If no such cues were available, the content of the teachers ideas was used to guide the breaking of intervention ideas. Specifically, if the intervention was des cribed in a way that clearly indicated the simultaneous use of two or more intervention ideas, it was kept together and coded as compound. If however, a participant switched from describing one type of intervention (e.g., behavioral ) to describing a separ ate idea that would be coded as a different intervention type (e.g., instructional ) and was not to be implemented in conjunction with previously mentioned strategy, this also served as a cue to break intervention ideas. In the text boxes of a coding form, the word (CHIP) is shown to indicate that a chip was dropped, and for situations such as participant 4A3, the parenthetical notation (continues on) at the end of an intervention description indicates that the participant did not drop a chip and the interv ention was broken by the researcher. On one occasion, a teacher (5B2) demonstrated incorrect use of the token description procedure in the opposite manner, dropping chips arbitrarily between ideas where there was no apparent shift in content or interventio n strategy. A portion of a single intervention idea is provided below to illustrate this phenomenon: We have a desk inspector who goes around and checks the desks (CHIP) to make sure theyre organized and keeps their papers where they should and if it s not then they have their color moved and at the end of Friday, if theyve had their color moved many times during the week then they dont get to eat lunch in the classroom with me on Friday. So its kind of like a good job, kudos,
57 yeah you get somethin g, or a pull back, not a punitive thing, you know you have to make them aware that youre not doing the right thing so youre not going to get the reward. (CHIP) You have to set up something in your room thats going to let them know the boundaries and alw ays make it so that when they internalize that its not you punishing them, that they are making you have to take this away, and Im sorry to have to take this away from you because I was really hoping that this week youd get to have lunch with me, and w ere going to have popcorn and watch a movie during lunchtime so Im sorry youre going to miss that, so make them internalize that its something they are making you do, rather than something you are doing to them. Because you dont want them to resent or see you as the Grim Reaper, its more that they need to understand that if I do these things then I get to have all of this so thats sort of building. And then they love you and they cant wait to do those things, and they understand that when they are punished, they dont get those things, and its something they need to work on so the next week its like, I know Im going to do it right, Im going to get to eat lunch and its something they realize they can control. (CHIP) Despite the fact that several tokens were dropped during the description, this intervention was transcribed and coded as a single behavioral intervention because everything this teacher described referred to the use of positive and negative consequences to impact student behav ior. Participant 5B2 was the only teacher to use the chips in this fashion. Following transcription, each participants data from the demographic questionnaire and interview were entered into an SPSS statistical database for analysis. Interrater reliabi lity. Following the data entry phase of the study, an advanced graduate student was recruited to serve as an independent rater for
58 coded data. To establish the reliability of results obtained from analyzing interview transcripts along specificity, and in tervention nature codes, an independent rater reviewed approximately 50% of the participant transcripts ( N =14). The independent rater was familiar with the purposes and procedures of the study, and was trained on the definitions used to code teacher respo nses in terms of specificity of description, and type of interventions suggested (see Appendix G). For the purposes of training, the independent rater was required to complete a coding form from one of the four pilot study transcripts, on which responses had been transcribed into individual text boxes and counted but not yet analyzed for intervention type or specificity. The transcript used for training had been previously coded on a different form by the researcher and major professor prior to independent rater training, and the results obtained by the researcher/major professor served as the standard by which the trainee was evaluated. To check the reliability of coding for specificity and intervention nature during the training phase, the independent ra ters codes for the training transcript were then compared to those obtained by the researcher/major professor, using the formula: (# of Agreements) (Total Agreements + Disagreements) x 100. This formula was used separately on specificit y and intervention results to calculate the accuracy of coding on each coding variable. The independent rater was required to achieve 80% agreement with the researcher/major professors results on each coding variable to pass the training exercise. This occurred on the first try, so no further training was provided. Following transcription/initial coding by the researcher, 14 interviews (135 total intervention suggestions) were randomly selected to be reviewed for interrater agreement. The independent ra ter was given blank copies of the coding form for each participant with text only, void of any coding marks or notes, and was then asked to determine specificity and type of each of the intervention suggestions and identify any hypotheses generated by the participants. The
59 formula used for these calculations was the same as that which was used in the training exercise, and a standard of 80% agreement was set. If the interrater reliability fell below the 80% criterion, then coding for all transcripts would be reviewed and disagreements would be resolved. Final interrater reliability for specificity was 81.5% and 85.2% for intervention type, which met the standard levels of acceptable interrater reliability, so only 50% of transcripts were analyzed an d reso lved for discrepancies. Data Analyses Variables investigated in this study are listed in Table 2 and independent and dependent variables are discussed separately below. Table 2 Independent, Dependent and Types of Variables Independent Variables Ty pe Dependent Variables Type Years of teaching experience continuous Number of interventions offered continuous IA participation continuous Overall m ean specificity continuous Referral to eligibility percentage continuous Presence of intervention typ es (each type is a separate dependent variable) Instructional Behavioral Classroom Structure Interdisciplinary Support Information Gathering Materials Communication Emotional/Social Support Compound dichotomous 1= present 0= absent Independent variables Independent variables were those teacher characteristics obtained from demographic questionnaire data. To determine the reliability of the instrument, an internal consistency analysis was conducted by
60 calculating Cronbachs alpha. In particular, Cronb achs alpha values were calculated for questions on which sub items were averaged to create a composite score to determine if any sub items were answered in inconsistent ways that might affect the reliability and meaningfulness of the composite scores. Th ese reliability values are highlighted below as they apply to individual variables. The variables of teaching experience and IA participation were taken directly from the demographic questionnaire and required no calculation. Referral to eligibility rate also an independent variable, is reported as a percentage and was calculated with data from obtained from the demographic questionnaire, using the following formula: ( Mean students eligible for special education 2002 2004) (Mean referrals for problem solving 2002 2004) x 100. In this way, referral to eligibility rate can be conceptualized as a hit rate or a measure of the accuracy with which teachers refer students to special education, providing the percentage of cases referred which result ed in the development of an IEP and/or change in educational placement ( McNamara & Hollinger, 2003; Sindelar et al., 1992 ) On the questionnaire, teachers reported their referrals to problem solving teams, referrals to school psychologists, and children f ound eligible for ESE for two academic years: the previous school year (2003 2004) and two years ago (2002 2003). This was done to improve the accuracy and reliability of this rate for teachers, given the potential for annual fluctuations in the number of children referred for special education. It was assumed that these rates should be reasonably similar for teachers across years, though not necessarily identical. To test this assumption, three Pearson product moment correlations ( r ) were calculated bet ween the 2002 2003 and 2003 2004 rates for each of these variables to determine their degree of relatedness (see Table 3). Since all correlations were positive and statistically significant, 2002 2003 and 2003 2004 rates were then averaged to create a mea n value for each variable. The referral
61 to eligibility rate for each teacher was calculated using these mean values and thus reflects two years of teacher referrals. Table 3 Pearson P roduct M oment C orrelations (r) B etween the 2002 2003 and 2003 2004 R at es of R eferrals to P roblem S olving T eams, R eferrals to S chool P sychologists, and C hildren F ound E ligible for ESE Services Questionnaire Items Correlation (p value) 10. How many children have you referred to your schools problem solving team in each of th e following years? Last Year (2003 2004) Two Years Ago (2002 2003) r =.784 ( p < 0.001) 11. Of those above referred children, how many were eventually referred to the school psychologist or other personnel for evaluation for suspected disabilit y in each of the following years? Last Year (2003 2004) Two Years Ago (2002 2003) r =.694 ( p = 0.001) 12. Of those above children you referred for suspected disability, how many were eventually found to be eligible for ESE services in each of the following years? Last Year (2003 2004) Two Years Ago (2002 2003) r =.737 ( p < 0.001) There were two items on the demographic questionnaire that measured the independent variables of training experiences and IA practices of schools each of which cont ained several sub items scored on a 5 or 6 point Likert scale (see Appendix C questions 15 & 16). For each of these questions, scores from the Likert scaled items were averaged to yield a Composite T raining Experiences score and a Composite IA Practices of School score that reflects the whole of teachers responses to each item. On the Composite IA Practices of School variable, items on which teachers responded Dont Know were not counted when calculating the mean. For example, if a teacher responded Dont Know to one of the eight subitems on Question 15 ( IA practices of school ), then that teachers score was calculated as an average of seven possible items, rather than eight. For responses to item 15 ( IA practices of schools ), Cronbachs
62 coefficien t alpha was 0.90, indicating a relatively high degree of reliability. It should be noted that this calculation is based on a sample of N =12, because 17 cases were eliminated due to missing data or responses of Dont Know on one or more of the sub items. With regard to responses on item 16 ( training experiences ), Cronbachs coefficient alpha was 0.69 ( N =29), indicating a moderate degree of reliability. Although this is a lower value than desired, this level of internal consistency was viewed as appropria te for the training experiences question, because there is expected to be more variability among teachers individual experiences with learning about interventions due to the college and timeframe in which they were trained, the schools in which they work, and their own motivation to seek out training on intervention strategies. Although there is no firm standard for interpreting alpha values, the psychological literature tends to view alpha values greater than .70 as acceptable (Cortina, 1993). Because bo th coefficient alpha values were close to or above this generally accepted standard for internal consistency reliability, the decision was made to proceed with creating composites for the IA practices of schools and training experiences variables in the ab ove described manner. Furthermore, Cortina (1993) suggests that the level of reliability that is deemed adequate for a given scale depends upon the decisions to be made with that scale. For a high stakes assessment such as the Scholastic Aptitude Test (SAT), for example, high alpha values would be necessary since scores are often used for making important educational decisions such as college admissions. In this case, both IA practices of schools and training experiences variables are used as a self re ported indicator of teacher experiences at the personal and building level and are not believed to be a perfect measure of either construct. As such, a more modest measure of reliability such as that obtained on the training experiences variable was deeme d acceptable. Dependent variables Dependent variables consisted of the salient features of teacher responses to interview/vignette. The first dependent variable, number of interventions was taken directly from the Intervention Coding Form,
63 on which th e number of interventions described for each goal was reported. This variable required no calculation. T he outcome variable type of interventions was treated as nine separate dichotomous variables in order to offer descriptive information regarding the p roportions of teachers who described each of the intervention categories (e.g., percent of teachers described one or more behavioral interventions). In addition, treating type of interventions in this dichotomous fashion allowed calculation of correl ation coefficients that index the degree of relationship both between intervention types (e.g., the likelihood that teachers will suggest both behavioral and classroom structure interventions) and among intervention types and teacher characteristics (e.g., relating years of teaching experience to a participants likelihood of suggesting instructional interventions). For each type of intervention variable, teachers responses were scored as 0 (indicating the teacher did not suggest that type of strategy) or 1 (indicating the teacher did suggest that type of strategy). Finally, overall mean specificity of descriptions was calculated for each participant by taking the mean specificity rating from all of each teachers intervention descriptions. In order to a ppropriately combine teachers specificity ratings in this way, it was necessary to determine reliability across all participants specificity ratings. Cronbachs alpha, commonly used for calculating internal consistency reliability was not a n appropriate metric for use with specificity ratings as it would likely be impacted by differences in the number of intervention suggestions between types (e.g., instructional interventions were suggested a total of 9 times across all 29 teachers while behavioral inte rventions were suggested significantly more often 90 times ). To overcome this obstacle, individual participants intervention suggestions were entered from the coding form by the presence or absence of an intervention type (1 for presence, 0 for absence, as described above), intervention type frequency (e.g., how many times a participant suggested behavioral interventions), and the mean specificity of intervention descriptions by type (e.g., the mean specificity of a teachers behavioral intervention sugge stions). T o assess the reliability of specificity
64 ratings, t he mean specificity ratings per type values were correlated, to determine if there was support for the notion of combining specificity ratings into an overall mean specificity value that represen ts all of a participants intervention suggestions. Only mean specificity ratings of intervention types with an N greater than 10 (i.e., were suggested more than 10 times across the total sample) were included in the correlation matrix, as types with fewe r than 10 observations were not likely to be reliable enough for analysis. Intervention types with frequencies greater than 10 were behavioral, classroom structure, communication, and compound. The correlation matrix in Appendix M depicts findings from t his analysis that these four intervention types were indeed positively correlated indicat ing that participants who tended to be specific in describing one intervention tended to be specific in describing other interventions, while those who were less spec ific in their descriptions tended to be less specific across interventions. These data support the decision to collaps e specificity ratings into an overall mean specificity metric based on the average specificity ratings across intervention descriptions D esign. This study employed a correlational/nonexperimental design which involved collecting quantitative data to describe teachers professional characteristics (i.e., years of experience, IA team participation, training, and school IA practices) and int ervention ideas in response to the interview/vignette (i.e., number, specificity, and nature of interventions suggested), as well as to determine the degree of relationships that may exist between teacher demographic variables and intervention responses. Interestingly, terminology used to describe research of this type is the subject of some controversy, as recently scholars have suggested that the term correlational research is outdated and places undue focus on a given statistical analysis rather than o n a given research technique. Contemporary educational researchers suggest that this form of research instead be termed nonexperimental research or
65 systematic empirical inquiry in which the scientist does not have direct control of independent variables because their manifestations have already occurred or because they are inherently not manipulable. Inferences about relations among variables are made, without direct intervention, from concomitant variation of independent and dependent variables (Kerlin ger, 1986, p. 348 in original, as quoted in Johnson, 2001, p.7). The proposed category of nonexperimental research encompasses both causal comparative and correlational designs, which determine the relationships among categorical or continuous independent variables (respectively) and relevant dependent variables. Furthermore, this study can be classified as descriptive nonexperimental research because the primary objective of the proposed is to describe the phenomenon of teachers knowledge of classroom ba sed interventions, providing characteristics and potentially related factors where relevant (Johnson, 2001). Statistical prediction, characteristic of predictive nonexperimental research is not a primary goal of this study. Statistical analysis. Analysi s of data in this study involved the use of various descriptive statistics to examine dependent variables, namely the number of interventions teachers offered, the specificity of their descriptions, and the nature of the interventions they suggested. In a ddition, correlation coefficients were calculated to examine the degree of relationship between independent variables obtained from teacher demographic information and dependent variables that resulted from teachers responses to the interview and vignette Specifically, analyses sought to answer the following research questions: 1. What is the average number of interventions teachers offer to address a hypothetical classroom behavior problem? Analysis : Descriptive statistics
66 2. How specific are teachers in des criptions of interventions/strategies they would use in their classroom (average specificity rating per teacher)? Analysis : Descriptive statistics 3. What percentage of participating teachers suggests a given type of intervention (i.e., instructional, behavio ral, etc.)? Analysis : Descriptive statistics 3a What 2 or more intervention categories, if any, are likely to be suggested by the same teacher (i.e., what is the probability that a given teacher will suggest both intervention type x and intervention type y ?) Phi coefficient ( r ) 4. What is the relationship betwee n years of teaching experience and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of in tervention (i.e., instructional, behavioral, etc.)? Analysis : Pearson product moment correlation ( r ) for number and specificity p oint biserial c orrelation ( r pb ) for type 5. What is the relationship between frequency of participation in IA meetings and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? Analysis : Pearson product moment correlation ( r ) for number and specificity, p oint bi serial correlation ( r pb ) for type
67 6. What is the relationship between teachers referral to eligibility rate and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will sugges t a given type of intervention? Analysis : Pearson product moment correlation ( r ) for number and specificity, p oint biserial correlation ( r pb ) for type 7. What is the relationship between training experiences and number of interventions/strategies suggested, s pecificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? Analysis : Pearson product moment correlation ( r ) for number and specificity, p oint biserial correlation ( r pb ) for type 8. What is the relationship between intervention assistance (IA) practices of the participants school and number of interventions/strategies suggested, specificity of interventions/strategies descriptions, and the likelihood that a teacher will suggest a given type of intervention? Analysis : Pearson product moment correlation ( r ) for number and specificity, p oint biserial correlation ( r pb ) for type Pearson Product Moment correlation coefficients ( r) were used to determine if relationships existed between independent and continuous dependent variables. A series of p oint b serial correlation coefficients ( r pb ) were calculated to examine the relationships between continuous independent variables and the dichotomous type of intervention variables. Finally, any correlati ons among types of interventions were calculated using a Phi coefficient ( r ) since both variables are dichotomous.
68 Chapter IV Results Descriptive Analyses A series of descriptive analyses were employed to address the first three research questi ons of the study. Frequencies/percentages of responses and measures of central tendency (mean, range, standard deviation) are provided to illustrate the demographic characteristics of the participants, as well as their training and school based experience s with regard to working with IA teams, consulting with other educational professionals, and developing interventions for referring difficult to teach students. Salient characteristics of teachers responses to the vignette/structure interview are also pr esented, highlighting the number of interventions suggested, the overall mean specificity and types of interventions described Finally, p ost hoc descriptive analys e s provide information about the degree to which teachers felt that their training in inter ventions was adequate, as well as the frequency with which they offered hypotheses about the causes of student behavior in the context of possibilities for support ing the difficult to teach student in the classroom. Questionnaire Participant demographics. A total of 29 second and third grade teachers from 6 elementary schools in one large metropolitan school district in southwest Florida participated in this study. Of those teachers, 15 (51.7%) taught second grade and 14 (48.3%) taught third grade. The s ample was 96.6% female ( N =28), with only 1 male participant (3.4%). The race/ethnicity of the teachers included 1 African American (3.4%), 1 multiracial (3.4%) and 27 Caucasian (93.1%) participants. Teachers ages ranged from 23 to 57 years with a mean a ge of 34.5 ( N =28, 1 not reporting; SD =9.71). The majority of teachers ( N = 24; 82.8%) held a bachelors degree and 5 (17.2%) held masters degrees. All participants
69 held a Florida Elementary Education Teacher Certificate (N=29, 10 0% ), and additional certi fications among participants included English for Speakers of Other Languages (ESOL ; N = 14 ; 41.4%), Early Childhood Education ( N = 2; 6.9%), and Exceptional Student Education (ESE; N = 1; 3.4%). The average number of years of teaching experience was 9.67 yea rs ( SD =9.84), with a range of 0.5 35 years. Table 4 summarizes participant demographic characteristics. Table 4 Demographic Characteristics of Participant Sample Gender Highest Degree Earned Race/Ethnicity Mean Years Teaching (SD) Male Female B.A/ B. S. M.A./ M.S./ M.Ed. Caucasian African American Multi racial Grade 2 ( N =15, 51.7%) 10.33 (10.20) 3.4% 48.3% 48.3% 3.4% 48.3% 3.4% 0% Grade 3 ( N =14, 48.3%) 8.96 (9.78) 0% 48.3% 34.5% 13.8% 44.8% 0% 3.4% Total Sample ( N =29) 9.67 (9.85) 3.4% 96.6% 82.8% 1 7.2% 93.1% 3.4% 3.4% Participant problem solving characteristics. Teachers responded to items on the questionnaire that reflected their perceptions of school based problem solving practices, as well as their own experiences with referring stud ents to problem solving teams, consulting with other educational professionals, and receiving training on intervention strategies. Twenty three teachers (79.3%) indicated that their school had a problem solving team that met regularly to discuss teachers concerns about students academic or behavioral performance, while six (20.7%) indicated that such a team did not exist at their school. Accordingly, a majority of teachers ( N =24; 83%) indicated that they were required
70 to refer students to a problem solv ing team before they could be referred for a suspected disability Notably, on several occasions teachers within the same school did not agree as to whether such a team existed. For example, at school B, four teachers responded that there was not a schoo l based problem solving team, while two teachers indicated that there was a team that met on an as needed basis. Overall, of all teachers who indicated the presence of a problem solving team at their school, 10 (34.5%) reported that the team met on a we ekly basis, four (13.8%) reported that the team met once a month, and nine (31.0%) reported that the team meet on an as needed basis. Again, teachers within schools disagreed on how often problem solving teams convened. For example, at school F, three te achers responded that the team met on a weekly basis, two teachers responded that the team met on a monthly basis, two teachers responded that the team met on an as needed basis, and one teacher responded that there was no such problem solving team. A sum mary of teachers responses to these items, broken down by response rates at each school, is displayed in Table 5 (see next page)
71 Table 5 Item Responses b y School: Existence, Schedules, and Requirements of Schoolwide Problem S olving Teams. Response Frequency (% of participants from school) Questionnair e Item and Response Options School A ( N =4) School B ( N =6) School C ( N =1) School D ( N =6) School E ( N =4) School F ( N =8) Total All Schools ( N =29) Does the school have a problem solving team that meets o n a regular basis? Yes 4 (100%) 2 (33.3%) 1 (100%) 6 (100%) 3 (75.0%) 7 (87.5%) 23 (79.3%) No 0 (0%) 4 (66.7%) 0 (0%) 0 (0%) 1 (25.0%) 1 (12.5%) 6 (20.7%) Problem solving team meets on the following schedule: Weekly 2 (50.0%) 0 (0%) 0 (0%) 4 (66.7%) 1 (25.0%) 3 (37.5%) 10 (34.5%) Monthly 0 (0%) 2 (33.3%) 1 (100%) 0 (0%) 2 (50.0%) 2 (25.0%) 4 (13.8%) As Needed 2 (50.0%) 0 (0%) 0 (0%) 2 (33.3%) 0 (0%) 2 (25.0%) 9 (31.0%) Are you required to refer students to the problem solving team prior to refer ring to the school psychologist for testing? Yes 4 (100%) 4 (66.7%) 1 (100%) 6 (100%) 2 (50.0%) 7 (87.5%) 24 (82.8%) No 0 (0%) 2 (33.3%) 0 (0%) 0 (0%) 2 (50.0%) 1 (12.5%) 5 (17.2%) If teachers indicated on the questionnaire that their school did have a building level problem solving team to assist teachers with addressing student concerns, they also answered an 8 part question about the degree to which the team utilized several best practices (Kovaleski, 2002). Participants rated the frequency of ea ch practice on a 5 point Likert scale, ranging from 1 (Not at all) to 5 (Always); teachers indicated DK if they did not know about a particular item. The mean of the sub items from this question was used as the participants Composite IA Practices of Sch ool score, which was later correlated with
72 responses to the vignette. Table 6 (following page) summarizes means for each sub item of this question, including an overall mean Composite IA Practices of School score for all participants of 3.81 ( SD =0.71) sug gesting that teams IA practices are somewhat to usually consistent with best practices. Receiving the highest ratings were items asking about the degree to which empirically based interventions are used ( M =4.52; SD =0.59) and whether someone on the te am (teacher or other member) is required to collect data on the intervention ( M =4.50; SD =0.89).
73 Table 6 Sub Item Means: Utilization of Best Practices among Schoolwide Problem S olving Teams Questionnaire Item Mean* SD N (%) Resp. Dont Know (a) Does t he principal or assistant principal participate in team meetings? 3.50 1.10 1 ( 3 %) (b) Does the team look at schoolwide indicators (e.g., number of students served by the team, number of students referred for special education, number of students retained ) to determine the teams impact on the school as a whole? 3.13 1.06 1 0 ( 34 %) (c) Does your school provide other opportunities to get information about interventions for students with academic/behavioral problems from inservice trainings, case studies, re ading groups, etc.? 4.00 0.89 1 ( 3 %) (d) Are you (or is someone else) required to collect data on the intervention you implement? 4.50 0.86 1 ( 3 %) (e) Does the team attempt to use intervention strategies with demonstrated research support? 4.52 0.59 2 ( 7 %) (f) Does someone on the team assist you in getting interventions started in your classroom? 3.36 1.11 0 ( 0 %) (g) Does the team develop a plan to incorporate the intervention into your day to day instructional routine? 3.76 1.17 0 ( 0 %) (h) Does the te am invite parents to participate in selecting interventions for their children? 3.61 1.41 2 ( 7 %) Composite IA Practices of School Score** 3.81 0.71 Note Response choices were as follows: DK) Dont Know; 1) Not at all; 2) Rarely; 3) Somewhat; 4) Usuall y; 5) Always. DK responses were not calculated in the IA Practices of School Score. N =25 (87% of total sample) Number and percent of participants is reported because participants who indicated that there was not a problem solving team at their schoo l did not respond to these items. ** Score was calculated using the following formula: Sum of items (a) (h) Number of items with response of 1 5
74 Items with a response of Dont Know were not included in the calculation of the Composite IA Pract ices of School score but are reported descriptively (percent of individuals responding Dont Know) to indicate the relative degrees of certainty of each item. Most items were consistently answered with a rating of 1 5, although over one third of partici pants ( N =10; 34%) responded Dont Know to the item Does the team look at schoolwide indicators to determine the teams impact on the school as a whole? It should be noted that only 25 participants (87% of sample) responded to this 8 part item; partici pants who indicated that their school did not have a problem solving team did not respond to this question. Questionnaire items also included teachers individual behaviors with regard to referring students to problem solving teams and school psychologis ts, as well as the frequency with which those same referred students were found eligible for ESE services As mentioned previously, six participants indicated their current school s ites did not have problem solving teams ; however, o nly four of these indiv iduals reported that they referred zero students to the problem solving team in the past two years; two teachers had been at a different school site during the previous two years and were able to indicate referral patterns at those sites Five of these si x teachers had referred students to the school psychologist, despite the lack of a problem solving team at their site, and those responses are included in the results below. A total of five teachers did not provide any da ta about their referral behaviors across the past two years Four of these participants were new teachers who had not had prior experiences with referrals ; it is not clear why the fifth teacher failed to report these data as she had six years of teaching experience Due to these irregula rities in reporting, the following results about referral behaviors are described in reference to the total number of participants responding to each item. In response to item 10 on the questionnaire, t eachers indicated the frequency with which they refe rred difficult to teach students to problem solving teams during the 2002 2003 and 2003 2004 school years, and the mean of these values across both year s was calculated for each teacher On average, each of
75 the 22 responding teacher s referred approximatel y three difficult to teach students to their schools problem solving team each year ( M =3.3; range 0 9; SD = 2.02). Item 11 asked how many of those referred students from both academic years who had been referred to the problem solving team were subsequent ly referred to the school psychologist or other personnel to conduct a psychoeducational evaluation for a suspected disability Again, these data were collapsed across the two years reported. O f the 24 teacher s responding to this item, an average of appr oximately two difficult to teach students were referred to the school psychologist each year ( M = 1.8 ; range 0 6 ; SD = 1.42 ). Finally, item 13 asked how many of the students from both academic years referred to the school psychologist were found eligible for ESE services. On average, each of the 24 responding teacher s reported that a mean of 1.6 difficult to teach students were found eligible for ESE services (range 0 5; SD = 1.40). These data were then transformed to provide a referral to eligbility percen tage. As mentioned previously, five teachers did not provide information about their referrals in the 2002 2003 and 2003 2004 school years. Thus, the mean referral to eligibility rate was calculated with an N of 24 participants. For teachers who did pro vide referral information but did not have a problem solving at their building ( N =7) an alternative method of calculating referral to eligibility rate was necessary to eliminate the possibility of having an undefined value (e.g., 3/0= 3 students eligible out of 0 problem solving team referrals). For participants who had missing data or values of zero for the variable of mean referrals for problem solving from 2002 to 2004, the referral to eligibility rate was established by using the mean referrals to a s chool psychologist from 2002 to 2004. Across the 24 teachers for whom these data were available, the mean referral to eligibility accuracy was 52.1% (range 0 100, SD = 34.5), indicating that approximately half of all students referred either to the problem solving team or school psychologist were found to have significant problems warranting some form of ESE services.
76 In addition to formal referral processes, teachers were also asked to indicate all the educational professionals with whom they were likely t o consult about difficult to teach students. A list of educational professionals was provided that including school psychologists and counselors, teachers, specialists, and administrators ; teachers were permitted to select mor e than one person from the lis t Figure 1 (following page) i ndicates the percentage of teachers who indicated they consult with various educational professionals. Most teachers reported consulting with a different grade teacher ( N= 22; 75.9%); in particular, they consulted with teach ers from the grade prior to their own (e.g., third grade teachers consulted with second grade teachers). A majority of teachers also reported seeking the advice of school counselors ( N= 19; 65.5%), while school psychologists and same grade teachers were u tilized as consultants by a minority of teachers ( N= 10 or 34.5%, and N= 11 or 37.9%, respectively).
77 0 10 20 30 40 50 60 70 80 90 100 School Psychologist School Counselor Same Grade Teacher Different Grade Teacher ESE Teacher ESE Specialist Specialist Principal Other Ed. Professional Educational Professionals Serving as Consultants Percentage of Teachers Indicating Consultation w/ Ea. Professional 34.5% 65.5% 37.9% 75.9% 6.9% 17.2% 17.2% 17.2% 6.9% Figure 1 Educational Professionals w ith Whom Teachers Consult Abo ut Difficult to Teach Students Teachers were asked to estimate the number of times the y had consulted with any individual educational professional about a difficult to teach student (rather than a problem solving team) during both the 2002 2003 and 2003 2004 academic years, in order to yield a more reliable estimate. As with referral to eli gibility rate, it was assumed that rates of consultation should be reasonably similar for teachers across years, though not necessarily identical. To test this assumption, a Pearson product moment correlation ( r ) was calculated to determine the degree of relatedness between 2002 2003 and 2003 2004 rates of consultation. Since the correlation was positive ( r = .471, p < .05), these numbers were then averaged to create a combined estimate of consultation frequency for each teacher that would reflect two years of experience. Overall, teachers
78 indicated that they had consulted with one or more of these educational professionals an average of 4.43 times (range 0 20, SD = 4.84) in an academic year. To provide information about the degree to which they had been tr ained in classroom interventions for difficult to teach students, teachers completed a 5 part questionnaire item about several specific training experiences such as undergraduate/graduate coursework, CEUs and inservice workshops, participating in school ba sed IA teams, and supervision or teaching experiences in intervention development. Participants rated the frequency with which they had participated in each of the training experiences on a 5 point Likert scale, ranging from 1 (Not at all) to 5 (Extensive ly). Similarly to the Composite IA Practices of School score, a Composite Training Experiences score was calculated for each teacher by averaging the rating for each of the sub items for the training question, and was later correlated with responses to the vignette. The mean Composite Training Experiences score was 2.56 ( SD =0.71), indicating an overall training experience of rarely to somewhat participating in relevant intervention training experiences. Table 7 (following page) summarizes means for eac h sub item of this question, as well as the overall composite.
79 Table 7 Sub Item Means: Teachers Training Experiences in Classroom based Interventions Questionnaire Item Mean SD (a) Classes taken in college or graduate school 3.03 1.21 (b) Inservice wo rkshops 3.03 1.21 (c) Continuing Education Unites (CEUs) obtained at non school workshops/professional conferences 1.69 0.93 (d) Participation in problem solving teams or similar consultative groups 2.79 1.11 (e) Supervised practice in developing and im plementing interventions 2.48 1.21 (f) Have taught/mentored others in developing and implementing interventions 2.45 1.30 Composite Training Experiences Score* 2.56 .071 Note Response choices were as follows: 1) Not at all; 2) Rarely; 3) Somewhat; 4) Often; 5) Extensively. Score was calculated using the following formula: Sum of items (a) (f) 6 Vignette and Structured Interview The first two research questions of this study addressed the outcomes from the vignette and structured i nterview : What is the average number of interventions teachers offer to address a hypothetical classroom behavior problem ? and How specific are teachers in descriptions o f interventions/strategies they would use in their classroom (average specificity rat ing per teacher)? Across all 29 participating teachers, a total of 282 interventions were described in response to the vignette and structured interview, with a mean of 9.72 (range= 3 18; SD = 3.75) and a mode of 11 interventions described per teacher. T he average of the overall mean specificity score across participants was 2.18 ( SD = 0.43), commensurate with a rating of moderate specificity on a 3 point Likert scale of low (1) to high (3). No significant relationship was identified between number and specificity of interventions (r=.054, p<.779), suggesting that overall specificity rating was not a function of the number of interventions offered (i.e., teachers who offered many
80 interventions were not less likely to be specific than those who only offer ed a few intervention ideas). With regard to the third research question, What percentage of participating teachers suggests a given type of intervention (i.e., instructional, behavioral, etc.)? Figure 2 (following page) shows the percentage of teachers suggesting each of the intervention types. Nearly all teachers (96.6%) suggested at least one behavioral intervention, with a majority of teachers also offering one or more classroom structure (79.3%), communication (75.9%) and compound (62.1%) interventi ons. Because compound interventions were not coded to denote their constituent types (e.g., behavioral and communication), it is not clear from these data what types of interventions the compound interventions comprise. The breadth of teachers intervent ion suggestions was analyzed by counting the number of intervention types each teacher offered. For example, one teacher might have suggested only one or two intervention types (e.g., behavioral and classroom structure), while another teacher might have s uggested a variety of intervention types. The mean number of intervention types suggested was 4.14 (range= 2 6; SD = 1.4), indicating that teachers offered an average of approximately four of the nine intervention categories.
81 Figure 2 Percent of Total Sample Suggesting Each Intervention Type 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Instructional Behavioral Classroom Structure Interdisciplinary Support Information Gathering Materials Communication Emotional/Social Support Compound Intervention Type Percent of Teachers Suggesting 27.6% 96.6% 79.3% 6.9% 31.0% 20.1% 75.9% 20.1% 62.1% Intervention categories also were analyzed in reference to the mean specificity rating by type. Figure 3 illustrates the relative frequency and mean specificity rating for each intervention type, with the dashed line indicating the overall mean specificity across all intervention descriptions. Again, the most commonly offered interventions were behavioral ( N =90), classroom structure ( N =58), communication ( N =47), and compound ( N =45); these four types comprised app roximately 85% of all intervention suggestions ( N =240). Notably, not all of the most frequently described intervention types were among the most specific. Instructional ( M = 2.50, SD = 0.53), behavioral ( M = 2.18, SD = 0.50), classroom structure, and interdi sciplinary support ( M = 3.0, SD = 0.0) were described in the most specific language. It should be noted, however, that a total of only two interdisciplinary support interventions were offered by two separate teachers, and this low frequency explains the lac k of variability and high rating of this particular intervention type. A summary of all relevant descriptive data for
82 intervention suggestions, including frequency, percent reporting, mean specificity, can be found in Table 8 (see next page) Figure 3 T otal Number of Suggestions and Mean Specificity Rating by Intervention Type 0 10 20 30 40 50 60 70 80 90 100 Instructional Behavioral Classroom Structure Interdisciplinary Support Information Gathering Materials Communication Emotional/Social Support Compound Intervention Type Number of Times Suggested 0 0.5 1 1.5 2 2.5 3 Mean Specificity Rating # of Times Suggested Mean Specificity Rating 9 90 58 2 13 10 47 8 45 mean specificity = 2.18 ( SD =0.43) Post Hoc Descriptive Analys e s Perception of training adequacy Item 17 on the questionnaire asked teachers if they felt that they were adequately trained in classroom based int erventions, using a similar 5 point Likert scale that ranged from 1 (Not at all) to 5 (Definitely). Though this variable was not included in the proposed analyses, a summary of teachers responses to this item may provide useful information for educationa l professionals. Overall, teachers reported an average rating of 3.72 ( SD =0.96), indicating a perception of somewhat to mostly adequate training in classroom interventions.
83 Table 8 Intervention Suggestions in Response to the Vignette and Structu red Interview by Number of Teachers Suggesting, Relative Frequency, and Mean Specificity Intervention Type Number (%) of Teachers with Type Present Frequency of Suggestion Mean (SD) Specificity Rating Instructional 21 (27.6%) 9 2.50 (0.53) Behavioral 28 (96.6%) 90 2.18 (0.50) Classroom Structure 23 (79.3%) 58 2.17 (0.53) Interdisciplinary Support 2 (6.9%) 2 3.00 (0.0) Information Gathering 9 (31.0%) 13 2.00 (0.69) Materials 6 (20.1%) 10 1.88 (0.87) Communication 22 (75.9%) 47 2.09 (0.67) Emotional/ Social Support 6 (20.1%) 8 2.08 (0.66) Compound 18 (62.1%) 45 2.15 (0.64) Total 282 2.18 (0.43) Presence and frequency of hypotheses Data from the pilot study (Appendix B) indicated that in addition to describing ideas for helping the student in the standardized vignette, teachers occasionally offered hypotheses about the cause of student behavior. A prompt for noting and quantifying hypotheses was subsequently added to the code definitions and coding procedures (Appendix G) to allow for post hoc ana lysis. A total of 18 teachers (62.1%) suggested one or more hypotheses with a mean of 2.2 hypotheses offered per teacher (range 1 10 ; SD = 2.16 ). A closer examination of these data revealed that the distribution of hypothesis frequency per teacher was p ositively skewed (skewness=3.09) and
84 considerably leptokurtic (kurtosis=10.67), with an extreme value of 10 hypotheses for a single participant (8F3 ; see Figure 4 for a boxplot depiction of this variables distribution ). Because this single observation wa s having such a strong impact on this variable, the extreme value was removed temporarily from the data set to better understand the distribution of hypothesis frequency. Without 8F3 ( n =17), the mean hypothesis frequency was reduced to 1.76 (range 1 4; S D = 0.97); skewness and kurtosis values were within acceptable limits (less than 1). Given the impact of this single observation, it was decided to exclude participant 8F3s data when computing any correlations that involved the hypothesis frequency variab le. Figure 4 Boxplot Depicting Distribution for Hypothesis Frequency 29 N = Hypothesis Frequency 12 10 8 6 4 2 0 -2 29
85 Correlational Analyses A series of correlational analyses were performed to answer the last five research questions of the study. T he relationship between teacher demographic characteristics (years teaching experience, frequency of participation in IA teams, referral to eligibility rates, composite training experiences, and composite IA practices of school) and primary outcomes on the structured int erview (number of interventions offered, overall mean specificity, and intervention types employed ) were investigated. The degree of relationships between intervention types also was analyzed. Finally, a series of post hoc correlational analyses were con ducted to examine additional questions about the data that arose after the formal proposal of this study about the degree of relationship between the presence and frequency of hypotheses, perception of training adequacy, and other teacher characteristics Prior to calculating correlation coefficients, scatterplots were examined for nonnormal distribution (e.g., curvilinear relationships) or outlying/extreme observations that would significant impacted correlation values. Unless otherwise noted, the data se t remained intact for computing correlations and the analyses that follow should be understood to reflect all participants data. Rationale for Exploratory Correlational Analysis All correlations described in the following pages are exploratory in nature and should be interpreted only as a means of further describing characteristics of the sample. Inferences about correlations in the population are not warranted by these data for two important reasons. First, the small sample size limits the precision o f the correlation coefficients, creating large confidence intervals in which the population correlation coefficient ( ? ) may fall. Second, and perhaps more importantly, assessing the statistical significance of these data is further complicated by the high number of correlations conducted. Sample size. T o provide informati on on the reliability of results for a sample of n =29 95% confidence interval s were calculated for three Pearson
86 product moment correlation coefficients ( r) obtained in this study by applying Fishers Z transformation to r statistics and using the following formula: CI Z = Z r (1.96)( s Z ), where s Z = __1___ v ( n 3) The result of the above formula is a confidence interval using Fishers Z; the final step is to transform confidence intervals back to an r statistic in order to be meaningfully reported and interpreted. It should be not ed that Fishers Z is not as accurate of a calculation for point biserial and phi coefficients, so correlations among intervention types (dichotomous variable) were not tested in this way. Confidence intervals were calculated for three correlation coeffic ients that represent the magnitude of correlations obtained from this data set. As Table 9 (following page) illustrates, a 95% confidence interval for a moderate to high magnitude positive correlation ( e.g., r 1 = .416) indicates that 95% of the time, ? will be found between .055 and .680. Th e large span of this confidence interval suggests that the true magnitude of r 1 may be relatively low, indicating very little relationship, or considerably high, indicating a high degree of relatedness between vari ables. As such, very little certainty exists when interpreting such a statistic. Somewhat more definitive statements can be made about high magnitude correlations (e.g., r 2 = .714). When applying the 95% confidence interval to r 2 it appears that ? fall s somewhere between .860 and .470. Both limits of the confidence interval indicate a fairly strong negative relationship between the variables, but the precise magnitude of the relationship in the population cannot be ascertained. Furthermore, correlat ions of this magnitude are rare in both the psychological literature in general (Cohen, 1992) and in this study in particular. Finally, examining the confidence interval of a low magnitude correlation indicating minimal relationship between variables (e.g ., r 3 = .097) revealed that variables in the population may actually have anywhere from a low to moderate magnitude negative relationship ( .280) to a moderate to high magnitude positive relationship (.445). These analyses demonstrate that
87 inferential stat ements about correlations obtained with this small sample would not be precise Table 9 Sample Confidence Intervals for Obtained Pearson Product Moment Correlations (r) CI r (95%) = 1 1 Obtained r statistic Z r (1.96) v n 3 Z r + (1.96) v n 3 r 1 = .416 .055 .680 r 2 = 714 .860 .470 r 3 = .097 .280 .445 Number of correlations Reporting the presence of intervention type s as nine separate dichotomous variable s greatly increased the number of correlations to be calculated C omparisons of the five main independent variables against the 11 dependent variables (number of interventions, overall mean specificity, and nine intervention types), as well as inter correlating the intervention types themselves, yielded a total of 1 0 6 si ngle correlation coefficients. To determine the amount of power required for each individual correlation to reach achieve significance, a n experimentwise alpha level of .05 would have to be divided by 1 0 6 As such, p values of correlations are not report ed and the magnitude of relevant correlation coefficients is discussed only as a means of describing relationships observed in the sample of the present study and developing hypotheses for future research exploration. Relationships Between Teacher Charac teristics and Interview Outcomes The final five research questions were concerned with the degree of relationship between five teacher characteristics (years teaching experience,
88 frequency of participation in IA teams, referral to eligibility rates, compo site training experiences, and composite IA practices of school) and the key interview outcomes (total number of interventions, overall mean specificity and types of interventions ) A series of exploratory Pearson product moment correlations ( r ) were con ducted among each of the teacher characteristics and the two continuous interview variables (total number of interventions and overall mean specificity ) The correlation matrix in Table 1 0 (following page) i llustrates the correlations found between these v ariables. Low magnitude positive relationships were identified between years of teaching experience and the total number of interventions teachers discussed ( r =.254), and between frequency of participating in IA teams and overall mean specificity ( r =.312) These data suggest that, for this sample, greater teacher experience was associated with a higher number of intervention ideas in response to the vignette, and that a history of more frequent participation on IA teams was associated with greater specifi city in intervention descriptions.
89 Table 10 Pearson Product Moment Correlations ( r ) Between Selected Teacher Characteristics Total Number of Interventions, and Overall Mean Specificity Teacher Characteristics Total Number of Interventions Overall Mean Specificity Years of Teaching Experience ( N =29) .254 .152 Frequency of P articipating in IA Teams ( N =22) .164 .097 Referral to E ligibility R ate ( N =24) .094 .312 Composite T raining E xperience s Score ( N =29) .181 .018 Composite IA P ractices of S chool Sc ore ( N =25) .063 .136 N values for each correlation are reported due to the irregularities in responses to questions about IA teams. Point biserial correlations ( r pb ) were used to calculate relationships between the continuous teacher characteristic vari ables and the dichotomous variables of presence of intervention types (0= absent 1= present ). The majority of correlations were positive but small in magnitude. These relationships are summarized by the correlation matrix in Appendix N and several of the correlations from this matrix are highlighted below. Years of teaching experience was related to several intervention types. Positive associations between years of teaching experience and interdisciplinary support ( r pb =.305) and compound interventions ( r pb =.488) suggest that teachers who have been teaching for a longer period of time are more likely to recruit support from their fellow educators and support staff or combine intervention strategies to assist difficult to teach students. A note of caution is warranted regarding the correlation between years of teaching experience and interdisciplinary support ; o nly two teachers (6.9% of the sample) suggested
90 interdis ciplinary support interventions and this restriction of variability is likely to affect the results of any correlation coefficients resulting from this variable. Two moderate ly strong positive relationships were identified between frequency of participation in IA teams and intervention types suggested, specifically interdisciplinary support ( r pb =.434) and materials ( r pb =.396). Again, t he correlation between frequency of participation in IA teams and interdisciplinary support should be viewed in light of low variability in the interdisciplinary support variable. The correlation between frequen cy of participation in IA teams and materials is more likely an accurate estimate of relationships in the sample, however, because materials interventions were offered by a greater proportion of participants (20% of the sample; N =6). Referral to eligibilit y percentage was negatively associated with both behavioral ( r pb = .295) and communication ( r pb = .517) interventions, indicating that teachers who were more accurate in their referrals were less likely to suggest these interventions. Similarly to the int erdisciplinary support variable the correlation between referral to eligibility percentage and behavioral interventions shou ld be interpreted with caution as only one teacher in the sample (3B3, who also had a very high referral to eligibility percentage of 1.0 or 100%) did not suggest a behavioral intervention. Limited variability in the behavioral intervention variable, coupled with a high referral to eligibility rate in the single observation without a behavioral intervention likely reduced the relia bility of this particular estimate The negative correlation between referral to eligibility and communication, however, is believed to be a more stable measure because it reflects a greater degree of variability in the communication variable; 75.9% ( N =22) of the sample suggested communication interventions while 24.1% ( N =7) did not. The c omposite training experiences score was somewhat positively associated with the presence of three intervention types. These data indicate that teachers in the sample wi th higher composite training scores were more likely to suggest classroom structure ( r pb =.346), materials ( r pb =.263), and
91 compound ( r pb =.217) interventions. Finally, the c omposite IA practices of school score was not found be meaningfully related with the presence or absence of any of the intervention types. Relationships Among Presence of Intervention Types A corollary to the third research question asked: What 2 or more intervention categories, if any, are likely to be suggested by the same teacher (i .e., what is the probability that a given teacher will suggest both intervention type x and intervention type y )? To address this question the nine dichotomous variables of intervention type were inter correlated using Phi coefficients ( r ) to determine if two or more intervention types were likely to be present together. These correlations are summarized in Appendix O. Again, because the dichotomous variable of intervention type indicating presence/absence of each strategy was used in this analysis, co rrelations including behavioral and interdisciplinary support types should be interpreted in light of their limited variability as described above. Although many of the correlations were of low magnitude, several noteworthy relationships emerged from this analysis. First and not surprisingly, compound interventions were positively associated with three other intervention types: behavioral ( r =.242), interdisciplinary support ( r =.213), and information gathering ( r =.214). Somewhat more interesting, howev er, was the finding that compound interventions were negatively associated with instructional ( r = .313) and classroom structure ( r = .224) interventions, indicating that as compound interventions are increasingly suggested, instructional and classroom s tructure are less likely to be offered. Other intervention types found be positively associated included information gathering and communication ( r =.247), behavioral and communication ( r =.306), classroom structure and materials ( r =.261), and emotional/s ocial support and instructional ( r =.256). All of these correlations were of a low to moderate magnitude, indicating that the strength of the relationship between variables was not especially strong in the sample. One
92 moderately strong negative relations hip was observed between instructional and behavioral intervention types ( r = .306), indicating that the likelihood of suggesting behavioral interventions was associated with a decreased likelihood of suggesting instructional interventions. Post Hoc Corre lational Analyses Perception of training adequacy. Teacher s professional characteristics were correlated with responses to the questionnaire item asking how adequately trained they believed they were with regard to classroom based interventions. Results in Appendix O indicate that several of the teacher characteristics were positively associated with a higher satisfaction with training as rated on the questionnaire, including years of teaching experience ( r =.275), composite training score ( r =.433), and co mposite IA practices of school score ( r =.323). As would be expected increases in composite training experiences scores were accompanied by increases in the degree to which teachers felt well trained, and participants who had been teaching for longer peri ods of time also were likely to rate them selves as adequately trained. Additionally, t eachers who rated their schools IA teams higher on use of best practices also were more likely to feel that they were adequately trained in classroom based intervention strategies. Presence and f requency of h ypotheses. Hypotheses, which were added to coding procedures following the pilot study, were also inspected for relationships with the studys five main independent variables describing teacher characteristics ( year s teaching experience, frequency of participation in IA teams, referral to eligibility rates, composite training experiences, and composite IA practices of school ). Table 11 (following page) summarizes the relationships among hypothesis variables and teac her characteristics. Again, s everal of the teacher characteristics were found to be positively associated with the presence of hypotheses within teachers responses to the vignette, including years of teaching experience ( r pb =.286), composite training sc ore ( r pb =.369), and composite IA practices of school score ( r pb = .279) These data suggest that teachers in the sample who were more likely to suggest hypotheses were those
93 who had more years of teaching experience, more exposure to training activities, and perceived their schools IA teams to be consistent with best practice recommendations. Table 11 Pearson Product Moment ( r ) and Point Biserial ( r pb ) Correlations Between Selected Teacher Characteristics Perception of Training Adequacy, Hypothesis P resence, and Frequency Teacher Characteristics Perception of Training Adequacy ( r ) Hypotheses Present ( r pb ) Hypothesis Frequency ( r ) Years of Teaching Experience ( N =29) .275 .286 .326 Frequency of P articipating in IA Teams ( N =22) .096 .125 .033 Referr al to E ligibility R ate ( N =24) .052 .020 .144 Composite T raining E xperience s Score ( N =29) .433 .369 .127 Composite IA P ractices of S chool Score ( N =25) .323 .279 .416 N values for each correlation are reported due to the irregularities in responses to questions about IA teams. Hypothesis frequency was also examined as a dependent variable, to determine what (if any) teacher characteristics were associated with a high number of hypotheses within intervention descriptions. As discussed under descriptiv e analyses, however, inspection of skewness/kurtosis values and visual analysis of a boxplot for the hypothesis frequency variable indicated a nonnormal distribution (Figure 4) led to the decision to exclude an extreme observation (10 hypotheses, participa nt 8F3) from the data set. As such, correlations involving hypothesis frequency reported in Table 11 do not include participant 8F3.
94 Results of this analysis suggest that two main variables were associated with high hypothesis frequencies in the sample: years of teaching experience ( r =.326) and composite IA practices of school score ( r =.416). More experienced teachers as well as those who reported perceptions of best practices within their IA teams were more likely to generate high number of hypothese s about student behavior than less experienced teachers or those who did not feel that their IA teams operated in a manner consistent with best practices.
95 Chapter V Discussion As a result of laws such as EHA (1975), IDEA (1997) and IDEIA (2004), move ment s toward inclusion of students with disabilities in general education classrooms ( Lloyd & Gambatese, 1990 ), and changes in psychoeducational service delivery emphasizing prereferral intervention for difficult to teach students (Graden, Casey, & Christe nson, 1985; NASP, 1995), todays teachers must be capable of differentiating instruction for a heterogeneous group of students and responding proactively to a wide variety of academic and behavioral needs. School based multidisciplinary IA teams have beco me a common mechanism for problem solving and supporting teachers efforts with difficult to teach students. Research suggests that most of teachers intervention efforts for difficult to teach students occur prior to referring the students to the IA team ( Wi lson et al., 1998). If teachers often function independently in intervention development and implementation (Bahr, 1994; Wilson et al., 1998), then a baseline measure of teachers knowledge and use of classroom interventions may serve as a foundation o n which IA team suggestions may build. Additionally, an understanding of factors associated with teachers intervention knowledge may provide insight as to the individual and building level attributes that maximize a teachers ability to respond to the ne eds of difficult to teach students. The present study had two primary objectives : (a) replicate a portion of Wilson et al. (1998) to describe teachers intervention ideas in response to a standardized vignette of a hypothetical student with academic and behavioral difficulties and (b) expand upon the work of Wilson et al. by examining trends and relationships among teachers professional characteristics and their responses to the vignette The following discussion addre sses the findings of this study, d raws comparisons to the findings of Wilson et al. where relevant, and
96 considers implications with regard to how IA teams in general and school psychologists in particular can better support teachers efforts to develop interventions for difficult to teach students. Limitations of the present study are discussed and implications for future research, including a follow up investigation of the present data, are offered. Teachers Self Reported Intervention Knowledge As in Wilson et al., the present study util ized three variables to assess the self reported knowledge base and behavioral regularities of second and third grade general education teachers when working with difficult to teach children in their classroom. In response to a vignette depicting a typic al student problem, teachers were asked to provide information of how to help a target student achieve two goals ( Stay in his seat and Stop talking out in class ). Subsequently, the number, specificity, and nature of their intervention ideas were analyz ed descriptively. Tables 1 2 and 1 3 provide a side by side comparison of the findings of the present study to that of Wilson et al. Wilson et al. arrived at a relatively pessimistic characterization of teachers intervention knowledge and speculated th at teachers limited knowledge of intervention strategies likely impeded brainstorming prior to and during IA team meetings (p. 56), o bserving that teachers may have many ideas for assisting difficult to teach students ( M =9.6; SD =3.6) but that their ideas were often described in relatively vague terms (mean specificity rating=1.63) Results from the present research share some of Wilson et al. findings. Specifically, these data support Wilson et al.s findings with regard to the number of intervention idea s generated and mostly concur with their findings on the most common types of interventions. However, participants in this study were able to offer ideas that were more specific than previously reported. Wilson et al.s finding that teachers were largely nonspecific or vague in their intervention ideas is widely cited as evidence that teachers may lack sufficient intervention knowledge. Given the present studys findings, there may be cause to reassess this widely held belief.
97 Table 12 Comparison of Fi ndings between Wilson et al. (1998) and Present Study with Regard to Number and Type of Interventions Wilson et al. (1998) N =20 Present Study N =29 Intervention Type Frequency of Suggestion % of Total Interventions Frequency of Suggestion % of Total Int erventions Instructional 43 23 9 3 Behavioral 103 54 90 32 Classroom Structure 24 13 58 21 Interdisciplinary Support 4 2 2 1 Information Gathering 17 9 13 5 Materials 0 0 10 4 Communication N/A N/A 47 17 Emotional/Social Support N/A N/A 8 3 Compound N/A N/A 45 16 Total 191 ( M =9.6; SD =3.6) 100% 282 ( M =9.72; SD =3.75) 100% Table 13 Comparison of Findings b etween Wilson et al. (1998) and Present Study w ith Regard to Specificity of Intervention Suggestions Wilson et al. (1998) Present Stu dy N % N % Low 90 47 49 17 Moderate 82 43 145 51 High 19 10 88 32 Mean (SD) Specificity M =1.63 (SD not reported) M =2.18 ( SD =0.43)
98 Number of interventions Notably, teachers in both the current research and Wilson et al. offered nearly iden tical mean number of interventions ( M =9.6; SD =3.6 in Wilson et al. and M =9.72; SD =3.75 in this study). An important caution must be offered here. The number of interventions described in response to the vignette and structure interview does not necessari ly translate to number of interventions a teacher would try if they were actually working with the target student (John). In Wilson et al., although teachers listed a mean of 9.6 intervention strategies in response to the hypothetical standardized case, when reporting on actual cases with difficult to teach students they indicated that they had attempt ed only roughly six ideas befor e w orking with an intervention assistance team. Because an actual referral case was not used in the present s tudy, conclusi ons about teachers actual use of intervention strategies cannot be inferred. Type of interventions Similar patterns in utilization of the intervention types can be seen across both studies, with behavioral interventions as the most often recommended int ervention strategy (54% of interventions in Wilson et al.; 32% of interventions in the present study). This finding is consistent with past survey research on the most prevalent types of interventions (Brown et al, 1991; Sevcik & Ysseldyke, 1986; Mamlin & Harris, 1998). Classroom structure interventions also featured prominently in the ideas of teachers in both studies (13% of interventions in Wilson et al.; 21% of interventions in the present study). However, teachers utilized instructional interventions (23% of interventions in Wilson et al.; 3% of interventions in the present study) and materials interventions to a lesser degree (0% of interventions in Wilson et al; 4% of interventions in the present study). A greater breadth of intervention types was observed in the current study as evidenced by the expansion of the codes to include three new categories (communication, emotional/social support, and compound). Combined, these three categories constituted over one third (36%) of all intervention suggest ions. It is unclear if similar types of interventions were observed in Wilson et al. and
99 coded as one of the original six categories, or if these new codes represent an expansion of intervention efforts into previously unseen areas. Furthermore, the addi tion of a compound intervention category permits a better understanding of how teachers utilize intervention strategies simultaneously to address the needs of difficult to teach students. Because compound interventions were not unpacked in data entry an d coding to reveal their constituent types, however, it is not clear exactly what comprises them. Intervention types were also examined to determine if certain types were likely to be suggested together. In other words, the probability that a given teach er would suggest both intervention type x and intervention type y was estimated by inter correlating intervention types. Findings from this analysis reveal some interesting patterns in intervention type utilization. First, and rather expectedly, the comp ound intervention type was found to correlate with behavioral, interdisciplinary support, and information gathering interventions. Given the nature of the compound type, containing two or more discrete intervention types in a single action, one possible e xplanation of this finding is that compound interventions are more likely to be suggested in conjunction with behavioral, interdisciplinary support, and information gathering interventions because compound interventions themselves often contain these compo nents. Conversely, negative relationships were identified between compound interventions and instructional and classroom structure interventions. A similar explanation may also be applied to these negative relationships; if compound interventions freque ntly include instructional and classroom structure components, there may be less need to provide additional instructional/classroom structure interventions (thus decreasing their frequency). These contradictory findings underscore the need to further exam ine the compound interventions to determine what intervention types they comprise and possible reveal the conditions under which teachers prefer to combine efforts rather than attempt them individually.
100 Beyond relationships with compound interventions, a n egative relationship between instructional and behavioral was also discovered. This finding might indicate two different takes on the target students behavior described in the vignette : t he instructional approach associated with hypotheses suggesting t hat academic issues are impacting behavior, and the behavioral approach suggesting that behavioral issues are either the primary concern or are impacting academic success. The following quote from participant 3B3, who offered one instructional and zero be havioral interventions (5 interventions total), might be viewed as an example of a more academic or instructional approach to Johns problem. Again, going along with this, I still feel hes bored. So not necessarily to give him more work, but along the same lines of giving him his work in chunks, such as his math, reading, and writing. Give it to him in a portion, so that he raises his hand, tells me hes done, so he can look forward to me giving him more work so that he will say, Im done with this p oint, and I will go over, check his work, give praise, and then I say, OK, heres the next portion. Do this and when youre done, raise your hand, wait for me to come, dont come to me just constantly reinforcing that I will come to him. Another quote from participant 5B2, who offered seven behavioral and zero instructional interventions (14 interventions total), is illustrative of a behavioral approach to the same problem. If youre training him to say in his seat, you could also say, If I notice th at youve been staying in your seat until we go to lunch, and you dont get out of your seat except when you give me a signal like if he needs to go sharpen his pencil, its a 1, if he needs to go to the restroom its a 2, if he needs a drink of water it s a 3, set up some sort of symbol for a movement out of his chair, and then you say, OK, once you use that symbol, you cant use it
101 again. So if you need to go sharpen your pencil, you may go sharpen your pencil and then you make a tally mark on the boa rd, OK? And after youve gone to the restroom, youve gone to the restroom and you can put a 2 up there for that, and after youve had your drink of water, thats your 3, so OK, before the first part of the day, youve used your three symbols, and once you ve used them you cant use them anymore, you cant get out of your seat. If you do, then there will be a consequence for it, because youve used up your movement options. (CHIP) And then just keep a tally of that so hes more aware of the times hes ge tting up. At present, these correlational data preclude any firm conclusions about why some interventions were found to be related while others were not. Content analysis of both instructional and behavioral intervention descriptions, as well as hypothes es indicating how teachers choose to employ certain interventions over other could further elucidate this phenomenon. Specificity of intervention descriptions An important finding revolves around the degree to which teachers described their intervention ideas in clear, replicable terms (i.e., another educational professional could implement the intervention based on the teachers description). As illustrated in Table 15, teachers in the present study were able to offer fully 30% more moderate or high sp ecificity intervention ideas than in Wilson et al. This discrepancy is attributed to the changes in the structured interview procedures for eliciting specific responses. Interview script text read by data collectors was revised to heavily emphasize the im portance of providing specific responses. By providing clearer examples and nonexamples of specific responses, teachers had a better idea of what types of responses would qualify as high specificity. In addition, the token description method, added aft er the pilot study, served as an additional prompt for providing detailed descriptions for a single intervention. This procedure included the statement unfortunately, this means you cant go back and add to an idea once youve dropped it into the cup, so try to describe your
102 ideas as completely as possible before you drop it. Such a directive prompted teachers to focus their attention to one intervention at a time. Anecdotal observations of participants indicated that some teachers reached to drop a tok en into the cup and hesitated or made comments like, Was that it for that one? and sometimes subsequently elaborated on interventions before finally deciding to drop a token and move on to a new idea. During the pilot phase, it was hypothesized that a heavy emphasis on specificity in the interview might result in a trade off where participants reduced the number of interventions they offered in order to be more detailed and comprehensive. If this was the case, one would expect to find a negative corr elation between number of interventions and specificity; the lack of a meaningful relationship between these two variables as reported in Chapter 4 provides preliminary evidence that no such trade off existed. Teachers were allowed to write down their i ntervention ideas prior to responding to the vignette; this might have promoted high specificity in spite of a high number of interventions by reducing the need to respond primarily from memory. R esults from this study suggest that when sufficiently p rompted, teachers can offer many specific suggestions. It is not clear, however, to what degree the heavy emphasis on specificity might have prompted teachers to be more specific than they would be in real life consultation or prereferral intervention or whether they may have included details that they might not have otherwise considered For example, participant 1F2 offered the following high specificity idea for helping John to stop talking out in class: I would give him a personal behavior chart so that, by class period, he would see how hes doing each period. For some kids, I use a cup with the little teddy bear counters, but he would probably play with them. So we probably wouldnt do that. For some kids I have a little piece of paper (basica lly a 3x5 index card) and have it copied on it M, T, W, Th, F and have the class periods listed top to bottom, and then either put like a smiley face when hes doing well so it
103 could be like a chart that I use to monitor his behavior and he sees his progr ess or it could be a chart that he does. Ive also done it with students where when they do something, when they talk out of turn, they make a tally mark on how many times they do that. So depending on how John feels about that, and how the other kids ta ke it (because if they see me going over and making a mark, that may make them wonder Oh, hes being bad, or hes being good), so depending on how my conversation with John goes, that would make the decision of who does the chart. There is no way to k now whether this participant added details to her description (e.g., the use of the index card marked by day and class period allowing John to self monitor his behavior ) to fulfill the requirements for specificity that were not a part of her typical teach ing /intervention practice. Modifications to the interview script that emphasized specificity might have been instructional in nature, teaching participants through examples and nonexamples how to describe their ideas in detail when they might not otherwis e have known how to do so. This is a marked deviation from the Wilson et al. protocol, which provided more neutral prompts for specificity, and comparisons between the Wilson et al. study and the present research must be viewed in light of this change. W hether intervention ideas are consistent with actual practice or not it is clear that teachers responses to the vignette do reflect their own intervention knowledge. The above example of a teacher and/or self monitored behavior chart was not modeled in the interview script; thus, it represents a teacher generated idea for working with John based on either past experience, observation of fellow colleagues, or training in behavior change strategies. As previously mentioned, Wilson et al. concluded th at teachers in their investigation described interventions in vague or unclear terms and suggested that classroom based interventions for difficult to teach students implemented by general education teachers may be of low quality. This assumption, however might be premature. Gresham (1989) distinguished between three potential
104 levels of intervention specification (global, intermediate, and molecular) on which specificity ratings of low moderate and high specificity in this study and Wilson et al. wer e based. In examining the role of specificity of treatment plans in resistance to consultation, Gresham concluded that intermediate specificity is the optimal level for which an interventionist should aim when designing treatment plans. Although molecula r descriptions of intervention plans are ideal for determining a functional relationship between intervention and behavior change, Gresham notes that interventions at this level can be met with resistance by those who are required to carry them out (i.e., teachers). Thus, intermediate specificity provides adequate information at a depth that is reasonable to all participants. Although the discussion in Gresham (1989) was aimed primarily at school psychologists and other individuals developing intervention s for difficult to teach students, it is not unreasonable to presume that teachers might operate in similar ways. If applying Greshams standards to teachers intervention suggestions obtained in this study, an overall mean specificity rating across all p articipants of 2.18 (corresponding to the moderate specificity rating) indicates an appropriate level of detail in intervention suggestions. Fully 83% of intervention ideas reported in this study received a rating of moderate or high specificity, su ggesting that when teachers are properly prompted and have specific responses modeled for them, the vast majority were able to communicate their ideas with sufficient detail such that they could be implemented by another educational professional based on t he description alone. Perhaps a more salient issue is whether specificity, as a dimension of teacher responses to the vignette and structured interview, is a valid indicator of teachers intervention knowledge. This study employed an open ended response f ormat using specificity ratings rather than a more traditional Likert scale to further quantify responses and provide a more sensitive index of teacher knowledge of interventions than previous studies using checklists or Likert scaled surveys. As stated i n Chapter 3, checklists or surveys may impose an a priori structure on the data; however, specificity ratings likewise may unintentionally
105 distort or misrepresent the construct of teacher knowledge. Specificity of response may provide some insight as to p recisely how a teacher plans to implement a given intervention, but a rating of a high specificity should not be interpreted as a good intervention, or one that is necessarily likely to achieve the goals stated in the vignette. With regard to the curre nt study, it became clear during discussion and resolution of disagreements between the researcher and the independent rater that an intervention could have a high specificity rating because it is clearly described and replicable, even though the intervent ion itself might be simplistic in nature or even inappropriate for the target behavior. For example, to address the goal of Stay in his seat participant 1C2 offered the following idea: Another thing I might try to do is if hes having a hard time sitti ng at math, saying OK John, were doing math. Why dont you go sit at my desk and do your math? It might make him feel special and the other kids dont give a hoot where they sit, theyre not having this problem. But if he gets to sit, not really in a rewarding area, just in a different area to take his mind off of things or if my desk has too many things that might distract him, I might let him sit at the guided reading table or another area where its just a different scenario. Maybe Ill let him si t in my teachers chair, so that its just something different to keep him staying in his seat. Even if it is a reward, if its a motivating reward, then its probably going to be worth it for him. Although this idea was coded as a classroom structure intervention because it modifies the amount of structure available to the student, from a behavioral perspective (one which many teachers described in their interventions), this intervention appears to be a misapplication of the Premack Principle (using p referred activities or events as reinforcers contingent upon completion of an unpreferred task; Cooper, Heward, & Heron, 1986). Allowing John to sit at the teachers desk at a time when he is demonstrating out of seat
106 behavior might actually serve as a reinforcer and could possibly increase the future rate of the behavior. One might suspect that if the goal is to get John to sit in his seat with greater frequency across the day, allowing him to complete his work in other areas of the classroom might no t help achieve that goal. Regardless of theoretical orientation or perspective, however, this idea should receive a specificity rating of high because there is no information missing about how to implement the intervention. It could be argued that a m ore knowledgeable teacher would anticipate the possibly reinforcing nature of this intervention and take a different approach to modifying the students behavior, but if specificity is used as a hallmark of knowledge then such problem solving skills as g auging the appropriateness of certain interventions for a given problem are overlooked. By definition, problem solving teams at the building level should promote a process in which educators identify relevant characteristics of student academic and behavi or problems, generate hypotheses about what might be causing student issues, and finally, implement and monitor evidence based intervention strategies in classroom settings to decrease academic/behavioral difficulty (Graden, Casey & Christenson, 1985). Sp ecificity might be representative of teachers understanding of how to implement various intervention types, but might not be an appropriate indicator of overall intervention functioning. Using specificity of description as a primary indicator of teacher k nowledge without quantification of related skills such as hypothesis development, problem solving ability, and progress monitoring, places undue focus on a single, isolated step of an otherwise complex and dynamic process. Alternative methods similar to that employed here have been developed to assess teachers problem solving skills. The neutral interview (Curtis & Watson, 1980) asks teachers to describe a problem they are having with a current student. Following transcription, the Consultation Verbal Analysis System (CVAS; Curtis & Zins, 1988) is used to code the teachers verbalizations to obtain a measure of problem solving skills. Prior research has shown that the degree to which teachers can clearly state and define a problem in the problem
107 identi fication stage of consultation accounts for 60% of the variance in plan implementation ( r =.776) and that plan implementation accounted for 95% of the variance in problem solution ( r =.977; Bergan & Tombari, 1977). The neutral interview has been used in sev eral empirical studies as a measure of teachers problem solving skills (Baker, 1997; Curtis & Watson, 1980; Durda, 2000; Grier, 2000, 2001). Perhaps a combination of approaches, including quantification of problem identification skills via the neutral in terview and intervention specific knowledge via the specificity rating, could provide a more holistic assessment of teachers overall functioning across all phases of the problem solving process. Teacher Problem Solving Characteristics The present study is unique in that it assessed teachers individual characteristics (e.g., years of teaching experience, referral/consultation history, training experiences, and perceptions of training) and their perceptions of building level practices regarding problem s olving (e.g., presence/absence of IA teams, use of best practices within the IA team). Although research regarding teachers perceptions of IA practices is relatively common, the majority of research to date has focused on educators attitudes about the pr ocess (e.g., Harrington & Gibson, 1986; Hawkins et al., 1991; Inman & Tolefson, 1988 ). Up to date information on how teachers perceive actual functioning of IA teams, including requirements for referral, frequency of meetings, and adherence with best pract ice recommendations, is currently unavailable. Though both Bahr (1994) and Carter and Sugai (1989) published reports from a survey of state education administrations regarding their requirements for IA and prereferral intervention, their findings are no l onger current and likely do not reflect current requirements. More importantly, survey responses from high level administrators removed from the day to day realities of education might not be consistent with perceptions of general education teachers worki ng on the front lines to support difficult to teach students. Information on IA team functioning from teachers perspectives was sought in the present study to better understand how teachers perceive IA teams to operate in the context of intervention de velopment.
108 Existence and practices of IA teams S tate and district ESE regulations require that teachers demonstrate efforts to provide interventions with difficult to teach students prior to referring them for a suspected disability The majority of tea chers in this study ( n =23; 79%) indicated that their school had an IA team that met regularly to assist teachers with these cases; 24 teachers (83%) reported that they were required to refer difficult to teach students the IA team prior to initiating a ref erral to the school psychologist. Five teachers, therefore, indicated that they are not required to go through a problem solving team before requesting an evaluation. Given state and local requirements to attempt prereferral interventions, this finding su ggests that some teachers may be developing prereferral interventions on their own and without the multidisciplinary support that an IA team can provide. Most IA models expect that teachers will take some independent actions to assist difficult to teach s tudents before consulting problem solving teams (e.g., Tilly, 2002 ); h owever, teachers without any access to problem solving teams are forced to go through the entire prereferral intervention process alone I t seems likely that these teachers may run out o f ideas more quickly and fewer interventions would be attempted before a referral for suspected disability is initiated, though no data are available to support this contention. Interestingly, teachers within the same school disagreed about (a) whether t he school had an IA team, (b) the frequency with which the team met, and (c) whether they were required to refer students to the IA team prior to referring to the school psychologist (see Table 5 in Ch. 4). This finding may be due to variability across te achers understanding of their schoo ls policies related to classroom based interventions. Alternatively, the school itself may not have a clear and consistent IA system in place and this is reflected in teachers variable responses. Another possible expl anation for this finding might be that ambiguous questionnaire wording about IA teams failed to elicit the appropriate response from teachers For example, p articipants from school E answered one of two
109 ways for questionnaire item #8 ( Does your school ha ve a problem solving team that regularly meets to discuss teachers concerns about students academic or behavioral performance ): weekly or as needed. Some participants provided anecdotal information on the questionnaire that revealed that teachers parti cipate in weekly grade level planning meetings and as needed administration/guidance meetings. Thus, the term problem solving team might have been too general and may require further clarification (e.g., multidisciplinary problem solving team with one or more administrators, guidance counselor, ESE personnel, etc.) to elicit a consistent and accurate response from teachers. Teachers who affirmed that their school did have a building wide problem solving team were also asked about the degree to which th e team engaged in several system and process level best practices associated with high quality IA programs (Kovaleski, 2002). Responses to this 8 part item indicate that, across all schools and participants, each of the practices is utilized at least so mewhat (corresponding to a rating of 3 or higher on a 5 point Likert scale; see Table 5 in Ch. 4). Teachers indicated that their IA teams were most likely to (a) use empirically supported research strategies and (b) require that data be collected data on the intervention. These findings are encouraging, given the recent emphasis on evidence based interventions in the field of education (Kratochwill & Shernoff, 2004) and prior research suggesting that progress monitoring data in prereferral intervention h as been lacking (Flugum & Reschly, 1993; Wilson et al., 1998). It is important to underscore that these responses are merely perceptions of teachers who have participated in IA teams. It is not clear to what degree these perceptions reflect the reality of IA team practices or functioning at the school level, and variability among teachers at the same school indicates similar patterns of inconsistency as previously described items. For example, composite IA practices scores of participants at school B ra nged from 1.8 (not at all to rarely) to 4.0 (usually), with two participants not responding because they had indicated that their school did not have problem solving team. The fact that
110 teachers within one school have different ideas about what their IA team does or does not require in terms of interventions may suggest that IA teams themselves are inconsistent in their practices and teachers individual perceptions are indicative of their own personal experience with their schools team. Data from th is study are inconclusive on this point and more empirical investigation is necessary to determine the cause of this phenomenon. Self reported consultation behaviors In addition to seeking assistance through multidisciplinary teams, often teachers reques t more individualized and immediate guidance in the form of one on one consultation. Behavioral consultation, or more generically called problem solving consultation (Kratochwill, Elliott, & Stoiber, 2002) has been a successful method for addressing a v ariety of educational issues including academic, emotional, and behavioral problems through the collaboration of a teacher and a school psychologist, guidance counselor, or other educational professional (Feldman & Kratochwill, 2004). Teachers in the prese nt study confirmed that they engaged in individualized consultation with one or more school based consultants an average of 4.4 times each year. Interestingly, though, teachers most often sought assistance from other teachers both at their own grade level and one grade below that which they taught ( N =22; 75.9% of sample). For example, third grade teachers reported consulting with second grade teachers, typically asking their predecessors what strategies had been successful for a specific difficult to teac h student in the previous year. Such a strategy for consultation is consistent with recommendations for multi level service delivery models in the literature in which the amount of resources (time, money, personnel) increases with the intensity of a give n problem ( Tilly, 2002 ). Consultation among teachers represents an appropriate entry level strategy for brainstorming on difficult cases; only those cases which cannot be adequately addressed with peer to peer consultation should be taken to the next leve l of team based problem solving. Anecdotal comments from teachers in this study suggested that consultation with same or different grade teachers might be less systematic and
111 might involve only a single instance of consultation rather than an ongoing, col laborative process as typically described in the literature. This form of consultation is somewhat consistent with the TAT model introduced in the late 1970s (Chalfant et al., 1979) and may be seen as qualitatively different from the behavioral/problem so lving consultation described in the literature by Kratochwill and colleagues, which typically entails a more prescriptive approach to problem description/analysis, intervention development, and progress monitoring. By contrast, TATs served more of a self help function for teachers struggling to assist difficult to teach students. Less emphasis was placed on the consultative and collaborative aspects of these teams, as the members typically were all teachers. Teachers who reported seeking assistance from other non instructional professionals most often reported consultation with the school counselor ( n =19; 65.5% of participants); only approximately one third of the sample ( n =10; 34.5%) indicated that they had consulted with a school psychologist in the la st two academic years. Consultati on has long been identified as a preferred professional activity by school psychologists (Meacham & Peckham, 1978) but these data indicate that school psychologists do not appear to function as the primary educational cons ultant in schools. Although this finding runs counter to the growing literature base on the role of school psychologist as consultant, recent data may explain this phenomenon. Numerous studies in the school psychology literature indicate that, even with recent movements towards IA and problem solving service delivery school psychologists continue to spend the majority of their time engaged in assessments related to special education (Curtis, Hunley, Walker, & Baker, 1999). In a recent survey of school p sychologists, teachers, and administrators, both teachers and administrators reported a greater desire for school psychologists to engage in teacher consultation than did school psychologists themselves ( Gilman & Gabriel, 2004). While 62% of teachers and 63% of administrators indicated that they would like to see more teacher consultation
112 from school psychologists, only 41% of school psychologists wanted more involvement in teacher consultation and 59% wished to remain their present level. Given that exp ectations for special education testing in recent years have not abated school psychologists may have reached a ceiling with regard to the amount of time they can spend in consultation It may be telling that 62% of school psychologists responding to the survey by Gilman and Gabriel (2004) indicated that they wished to maintain their current level of special education assessment. Thus, if most school psychologists do not want to decrease their assessment caseload, it is not surprising that they generally do not wish to further add to it by increasing their consultation with teachers. The data from the present study, however, are certainly not sufficient to draw such a conclusion and further examination of r ates and activities of consultation with school psychologists, school counselors, and other teachers is warranted. Training experiences Although IA teams were originally conceptualized to provide multidisciplinary support for teachers in vivo intervention skill development (Rathvon, 1999), recent bes t practices recommendations for IA teams suggest that teachers may require specific preservice and inservice training to prepare them for this important role (Kovaleski, 2002). A major question of this investigation was how teachers training experiences might influence their ability to respond to classroom behavior problems. A brief assessment of training experiences revealed that, as a group, teachers in the sample had limited to moderate exposure to training experiences that might prepare them for inte rvention development, including preservice coursework, inservice/CEU/workshop credits, supervised practice, or an opportunity to mentor others. Not surprisingly, the primary mechanism s for training teachers were preservice and inservice education. It has been frequently observed in the IA literature that traditional teacher education programs lack direct training in and exposure to intervention strategies for difficult to teach students (Newman, 1999; Worthington et al., 1997); thus, although teachers in this sample report learning
113 about interventions in this venue, the intensity or quality of that training cannot be assessed. Teachers in the present study indicated feeling that their training in classroom based interventions was somewhat to mostly a dequate, though teachers with greater professional experience were more likely to report high levels of satisfaction with their training than newer teachers. A relationship was identified between training experiences and perception of training adequacy, s uggesting that teachers who are more trained on classroom based are more likely to feel that their training is adequate. Perhaps more interesting, however, was the finding that teachers at schools where IA teams frequently followed best practices were als o more likely to report satisfaction with their intervention training. Although th ese findings can only be applied to the present sample, if this same finding were reported in the general population it would offer support for the educative role of IA team s often espoused in the literature ( Nelson et al., 1991; Rathvon, 1999; Safran & Safran, 1996 ) Relationships Between Selected Teacher Characteristics and Interview Outcomes Given teachers prominence in the IA process, it is essential that teachers have a solid foundation of ideas and understand from which they can draw when working with difficult to teach students ( Harrington & Gibson, 1986; Inman & Tolefson, 1988 ). An understanding of teachers familiarity with various interventions, and any profession al characteristics associated with such knowledge, may provide valuable information with which educators can evaluate the overall efficacy of IA programs. Knowledge of prereferral interventions, as conceptualized in the current study, is a complicated con struct comprised of more than one variable (number, specificity, and type of interventions presence and frequency of hypotheses ). Several variables were explored to determine what, if any, relationship they might have to the intervention variables under investigation. Unfortunately, t he present study found that teachers intervention ideas are not strongly related to the majority of independent variables of interest
114 ( years of teaching experience, IA team participation, referral to eligibility rate, train ing experiences, and IA practices of schools ) In general, small sample size and variability among participants responses to various items made correlations difficult to interpret and inferences about the population were not possible. Relationships amon g teacher characteristics and primary intervention outcomes were negligible, with two exceptions. Based on low to moderate correlations, years of teaching experience appeared to be somewhat related to number of interventions and IA participation appeared to be positively related to specificity of descriptions. The magnitude of these correlations were both low, but these findings raise the possibility that the number of interventions a teacher can generate in response to a typical classroom problem may inc rease with years teaching That years of teaching experience was the only variable to be associated with number of intervention ideas may suggest that personal experience and exposure to years of students with diverse needs may be a more powerful way to l earn about interventions than other opportunities assessed in this study (participation on IA teams, preservice and inservice training, and schools building wide intervention practices). Similarly, teachers may become better able to elaborate on interven tion ideas with increasing exposure to the IA process. Given that one of the primary goals of IA teams is to work with the teacher to establish a detailed plan for intervention (Kovaleski, 2002), this finding seems to suggest that teachers can indeed attr ibute some of their intervention knowledge to experiences with their schools IA team. Assorted teacher characteristics were found to be minimally associated with the choice of various intervention types, but a clear and consistent pattern among these var iables could not be identified. Furthermore, limited variability on the interdisciplinary support and behavioral types precludes any firm conclusions about these data. These findings may indicate spurious correlations resulting from an unknown third vari able (Vogt, 1999) and may be better understood through further investigation.
115 In addition to having numerous and detailed intervention ideas, the construct of intervention knowledge suggests that teachers should also be able to generate ideas about why the problem might be happening. Problem analysis, one step in the traditional problem solving model, prompts educators to consider what variables in a childs ecology might be causally related to manifestations in behavior and academics (Batsche & Knoff, 199 5 ). Furthermore, hypothesis development is central to the functional behavioral assessment (FBA) process; teachers must not only learn to recognize behavioral contingencies maintaining students behaviors (e.g., attention, escape/avoidance, tangible), but must be able to effectively modify antecedents and consequences in the environment to alter those contingencies (Myers & Holland, 2000). The majority of participants su ggested at least one hypothesis, and the number of hypotheses teachers suggested w as moderately related to years teaching experience and IA practices of schools. Though preliminary, these data suggest that teachers may be more likely to offer ideas about the causes of student behavior as they gain experience working in schools, or as t hey participate in IA teams that consistently use best practices for intervention development (Kovaleski, 2002). This finding is somewhat encouraging to the degree that it suggests teachers ability to develop hypotheses might be influenced by the practic es of other educators. Limitations N onrandom sampling of participating schools, homogeneity of the participants (i.e., second and third grade teachers from only one county) methodological constraints and small sample size of this study limit the externa l validity of these results. Each of these issues is discussed in greater detail in the following sections. Nonrandom selection of schools As described in Ch. 2, the proposal for this project called for schools to be selected for participation using stra tified random sampling on the basis of districtwide risk index ranking Unfortunately, this approach to sampling proved too logistically difficult to carry out for this thesis project. A comparison of participating schools risk indices were compared
116 i n the rank ordered database of all Hillsborough County schools (see Table 1, Ch. 3) This post hoc analysis suggests that participating schools in this study reasonably approximated the variability of schools in the county on the indicators of free/reduce d lunch, rates of disability, and percentage of teachers without advanced degrees. Furthermore, since school level analyses were not conducted (e.g., comparing number, type, and specificity of interventions offered by teachers at each school), school leve l characteristics are not believed to significantly impact the validity of these data. Homogeneity of participants The method of participant selection may have introduce d some bias into the research sample, in that teachers who agreed to participate may be those who have greater knowledge, training, or experience in classroom based interventions. Likewise, those teachers who declined to participate might be those who feel less experienced or knowledgeable in such interventions. For this reason, a $15 gi ft certificate to a local teacher supply store was offered as an incentive, to make participation in this study rewarding to all potential participants rather than exclusively to those who may find discussing interventions intrinsically rewarding. Demogra phic characteristics and descriptive data for this sample appear to refute that possibility, as a range of age, teaching experience, training experience, consultation/referral behaviors were observed among the participants in this study. Furthermore, the variability in the number and specificity of interventions identified in response to the vignette, generally consistent with the findings of Wilson et al. (1998), demonstrates the representativeness of the current sample. Nevertheless, it is possible tha t recruiting teachers from multiple counties or even states might have further diversified the sample and illustrated local differences in IA practices, and recruitment from a single school district limits the generalizability of these findings. Small samp le size. A substantial limitation of the present research is the small sample with which it was conducted ( n =29). Due to the time constraints of conducting interview research, the sample size was limited a priori to between 20
117 and 30 participants Confid ence intervals for obtained r values were calculated in order to predict the level of statistical precision analyses in this study would yield. Unfortunately, this exercise illustrated that confidence intervals at three potential r values ( r 1 =. 416 r 2 = 7 14 r 3 = 097 ) were relatively large (see Table 9, Ch. 4) Descriptive r esults from this study should be interpreted and generalized to larger populations with caution ; correlational results are intended to describe relationships in the sample only and sho uld not be used to infer the characteristics of the population B roader research using a larger and more stringe ntly selected sample could help validate the current findings. Methodological limitations The use of a single, hypothetical classroom situati on as a stimulus for teachers intervention suggestions can be considered both a limitation and strength of this study. Because the situation only described a single problem, it did not tap into all of the knowledge of interventions that a teacher might h ave and may have resulted in an underestimate of teachers intervention knowledge. Only a limited sample of the construct of knowledge can be obtained from the present studys measures. However, the problem described in the vignette was a multifaceted pr oblem that involved both l ow reading/math achievement and off task/disruptive behavior. As described previously, these are some of the most frequently observed problems in elementary level general education classrooms (Good et al., 1998; Myers & Holland, 2000; Raffaele & Bradley Klug, 2000). Thus, the vignette was expected to tap into some of the most necessary and often used intervention skills. This assumption was anecdotally supported by several teachers who made comments about the vignette such as G ood one, this is so common, or This IS a typical classroom behavior problem! With regard to the hypothetical nature of the scenario, it cannot be said to assess teachers actual intervention practices in vivo However, using a standardized vignette ( rather than asking teachers to describe an actual referral case) greatly facilitates comparisons among participants responses to the interview by eliminating numerous extraneous variables. In this way, use of a
118 single situation hypothetical vignette in t his study can be likened to the use of a single curriculum based measurement (CBM) probe to measure students reading skills; both measures provide an index of performance on a given domain, but do not purport to measure the whole of the domain. Finally the repeated prompts for specific responses might have elicited intervention ideas that were more detailed than they would have been in real life. Teachers may have added components to intervention ideas that they might not otherwise have considered. R esults from this study, therefore, should be interpreted as an indication of what teachers are capable of with regard to classroom based interventions and not necessarily what they do A potential threat to internal validity is the use of an interview me thod. Wilson et al. (1998) explained their use of this procedure as a way to maximize teachers ability to respond to questions, without the a priori limitations that surveys and checklists tend to impose on potential responses. However, demand character istics of the interview may have impact ed teachers responses in several ways. First, teachers may have suggested interventions that they would not typically use or describe interventions in ways they would not actually conduct them, as a result of social desirability effects. This was observed in one participant (3F3), who used a book on prereferral interventions to generate ideas for responding to the vignette ( McCarney, Cummins Wunderlich & Bauer, 1993) Although she stated verbally that she would us e the book in a real life situation to guide her prereferral intervention strategies, there is no way to determine the veracity of this claim. The researcher decided to permit the participant to use the book as the interview script and research protocol did not specify otherwise; she used it for the first two to three intervention ideas only and then closed the book and spoke extemporaneously. Subsequent inspection of this participants data revealed that she was no more detailed or creative than other participants in this sample; if anything, her overall mean specificity rating (1.5) fell below the mean observed across all participants and her number of intervention ideas (11) was not a cause for concern. This participant illustrates that it is not pos sible to state
119 with any certainty that teachers responses to the vignette represent their true response to a similar situation; rather, these data only indicate teachers potential responses to a classroom based problem idea. Secondly, although the in terview was designed to allow participants the greatest possible freedom of response, teachers may have rushed through their answers to complete the task quickly. This was observed in at least one participant (1D3), who indicated she was finished respondi ng to the first behavioral goal after describing only one intervention. When asked if she could think of anything else, she responded, Well, thats what I would try first. If that didnt work, then I would try something else, but thats the first thing I w ould do. Because this comment suggested that the teacher had other ideas that she was withholding, the data collector then prompted her to continue by saying, Well, if that idea didnt work, what else do you know to do to help John? The participant was then able to go on and describe another intervention for the behavioral goal, but still only produced a total of 4 interventions. It is not clear whether other teachers who did not make any revealing comments might have been responding in the same man ner. Anecdotal observations of several participants suggested that they might have rushed through the interview or were not fully invested in it. A dditionally, several teachers reported after the session that they would have liked to have received the vi gnette in advance of the interview so they could think it through. These participants indicated that they often spend a considerable amount of time thinking about how they will address the problems of real life students before they actually begin to act on them. Comments such as these should be considered in future studies using similar methodology, to maximize participants responses to the greatest degree possible. As with all self report data, results from the questionnaire portion of the study must b e interpreted with some caution. Teachers reported on the existence, schedules, and practices of IA teams at their schools, but these data do not necessarily reflect the actual practices of schools. This was clearly illustrated in the question about whet her schools have a building level problem
120 solving team; some teachers indicated that their school did have such a team, while other teachers at the same school reported that no such team existed. As previously discussed, it is not clear what caused these discrepancies in response. Teachers responses to these questionnaire items should be interpreted as their self reported perceptions of school practices, rather than statements about actual policies and practices of schools. Finally, the number of corre lations conducted in the present study significantly increases the likelihood of Type I error (inappropriately rejecting the null hypothesis; Cohen, 1992). Coding the variable of intervention type as nine separate dichotomous variables raised the number o f t otal dependent variables to 11 and the total number of planned correlations to 106. To determine the amount of power required for each individual correlation to achieve significance, an experimentwise alpha level of .05 would have to be divided by 106. Because in this situation ? values are essentially meaningless, statistical significance cannot be reported and characteristics of populations cannot be inferred. Despite this limitation, correlations were reported in this study to provide further descriptive information about the characteristics of participants in the sample and to highlight potential avenues for further investigation. Given these limitations, the results of the proposed study cannot be said to be definitive descriptions but rather a rough guess about the charac teristics and behaviors of this population. This study offers a unique contribution to the literature because it is the first to examine the relationship between teachers intervention knowledge and their professional characteristics. In addition, future studies of this scope and size could corroborate the findings of the present study, and aggregation of results in meta analysis literature could more precisely examine this phenomenon. Implications for School Psychologists This studys primary contributi on to the literature lies in its explication of teachers response patterns to a typical classroom academic/behavioral problem and perceptions of IA team functioning Although the problem presented in the
121 vignette was hypothetical in nature, it contained s everal relevant issues that teachers frequently encounter (Myers & Holland, 2000; Raffaele & Bradley Klug, 2000). Many school psychologists serve as behavioral consultants to teachers, either at the individual or problem solving team level (Kratochwill, E lliott, & Stoiber, 2002) A more comprehensive understanding of general education teachers typical responses to classroom problems may expedite the problem solving process by allowing behavioral consultants to quickly focus in on new intervention ideas t hat expand teachers previous efforts. If patterns observed in this study were to be true in the general teaching population, consultants might consider aiming training efforts at intervention types that were not frequently utilized here. Although some types of interventions may overlap with IA teams efforts, it might be appropriate to emphasize the importance of seeking interdisciplinary support (even if just from ot her teachers) and gathering information about the problem from early on in the process as well as considering ways in which materials and/or emotional/social supports might enhance student functioning. Such training efforts could occur on an individual, building wide, or district wide level. An important implication can be drawn from the fact that added prompts for specificity resulted in 30% more moderate and high specificity responses. To the extent that high levels of specification are beneficial in the problem solving process, particularly in the problem definition stage (Bergan & To mbari, 1977), school based consultants might wish to consider adding prompts (including examples and nonexamples) for specificity as a part of their initial consultation sessions. Such prompting might elicit more detailed description of student behavior a nd might encourage teachers to be more creative as they generate ideas for intervention in consultation. Broader implications for school psychologists stem from findings related to building level IA practices as reported by teachers in this study. To the extent that school psychologists are key players in the IA process (Kovaleski, 2002) these data may provide valuable information about potential weaknesses in the
122 IA service delivery model at the school level. School based practitioners may wish to condu ct an awareness survey of teachers at their school sites to establish what teachers believe the practices, policies, and schedules of IA teams are and to check these self report data against stated or desired roles of the IA team. School psychologists mig ht also be encouraged to note that IA practices do appear to have a somewhat educative function for teachers developing intervention skills. Consideration of how IA teams might be designed to promote professional development, rather than serve a reactiona ry function in response to student problems, might further enhance the efficacy of these teams. Finally, teachers self report data regarding individual consultation with educational professionals suggest ed that school psychologists we re not the dominant consultant in most schools. Given the availability of other administrators and support staff also available to assist teachers, as well as the itinerant status of most school psychologists, this may be an appropriate state of affairs. School psychologist s who seek to increase their involvement in teacher consultation may find this result discouraging; however, data from Gilbert & Gabriel (2004) indicate that teachers generally welcome consultative support from school psychologists. Implications for Futu re Research The present study attempted to describe general education teachers knowledge base with regard to classroom based interventions. As previously discussed, modifications were made to the Wilson et al. (1998) protocol to achieve the highest pos sible specificity of response. Future research using open ended, structured interview methodology must weigh the potential benefits of providing numerous prompts for specificity (i.e., capturing what intervention ideas teachers are capable of developing) against the limitations of such an approach (i.e., leading participants into overly specific responses not necessarily typical in everyday practice). In the present investigation, a primary goal was to understand what teachers know about classroom based i nterventions, so the emphasis on specificity helps elucidate the breadth and depth of teachers
123 intervention knowledge. Studies focused on assessing teachers actual intervention practices, however, should use prompts for specificity with caution as they might create social desirability effects that could skew data. Despite limitations in external validity previously discussed, the present study expands the literature base on general education teachers knowledge of prereferral intervention. Momentum is c ontinuing to gather for IA models of service delivery The Presidents Commission on Excellence in Special Education (2001) resoundingly announced that children placed in special education are general education children first (p. 7). IA processes offer an efficient model of ongoing, multi level consultative support to the teacher while maintaining the students placement in general education for as long as possible. More study of teachers as interventionists is critical to enhancing the effectiveness o f the IA process. While descriptive and nonexperimental research can be helpful in describing current conditions of practice, applied experimental research will be imperative in demonstrating how teachers intervention knowledge may be enhanced through tr aining, supervised practice or mentorship, consultative relationships, and IA teams. An immediate avenue for empirical investigation lies in the current data set. While the quantitative analyses described in this study offer some insights into patterns in teachers responses to typical behavior problems, they stop short of clearly depicting how teachers intervention ideas are or are not consistent with empirically supported, best practice interventions, or are appropriate for the problem in the vignett e. A mixed method c ontent analysis of the descriptions provided by this studys participants can help elucidate some of the more salient features of teachers intervention ideas that might be more closely related to their ultimate effectiveness. As descr ibed previously, several intervention types (e.g., instructional, behavioral, communication, compound) need to be unpacked to better understand what ideas comprised them. Examination of what student characteristics teachers most closely attend to when s uggesting interventions, as well as their hypotheses about potential causes of student behavior, is also
124 warranted. These two variables may shed light on how teachers come to perceive that a problem exists and what generally held beliefs teachers might hav e about student behavior problems. Additionally, a panel of intervention experts could be asked to review participants responses to rate the extent to which they represent high quality, appropriate interventions for targeting behaviors described in the v ignette. Unfortunately, an investigation of this magnitude is beyond the scope of the present study and will require significant co nsultation in mixed methodology. For this reason, the present study remains focused on preliminary descriptive and correlati onal data originally proposed. Conclusion This research study replicate d and extend ed the work of Wilson et al. (1998), which assessed the self reported knowledge based of general education teachers. In the present study, second and third grade general education teachers responded to a hypothetical description of a student behavior problem with many and varied classroom based intervention ideas. The most common types of interventions were behavioral, classroom structure, communication, and compound, but participants displayed a total of nine broad categories of interventions. Furthermore, when given numerous prompts and examples/nonexamples of specific responses, many teachers were able to describe their ideas with high levels of detail. Modifications t o the interview script emphasizing the importance of giving specific responses yielded over 30% more moderate and high specificity responses than reported in Wilson et al. Teachers also reported their perceptions of school practices with regard to IA, as well as their own consultation and referral practices and training in classroom based interventions. Most teachers indicated that their schools have a building wide problem solving team to support teachers in developing interventions for difficult to tea ch student; 83% of participants indicated that they were required to refer students to these teams before requesting an evaluation for a suspected disability. However, teachers within the same school sometimes disagreed about (a) whether the school had an IA team, (b) the frequency with
125 which the team met, and (c) whether they were required to meet with this team before referring for a suspected disability. Participants indicated that they consulted with a variety of educational professionals about diffic ult to teach students, including same and different grade teachers, school counselors, and school psychologists, with an average of approximately four consultative interactions occurring each year. When working with IA teams at their schools, 2), teacher s reported that their schools engaged in each of the eight systems and process level best practices of IA teams (Kovaleski, 2002) at least somewhat; among the most frequently reported of these best practices were use of evidence based interventions and data collection to monitor intervention progress. Finally, teachers indicated that they had limited to moderate exposure to training experiences that might prepare them for intervention development; teachers most often received training in the form of und ergraduate/graduate coursework and inservice training. Overall, teachers felt that their level of training in classroom based interventions was somewhat to mostly adequate. Unfortunately, teachers intervention ideas were not found to be strongly cor related to the majority of these teacher characteristics. A small sample size ( N =29) created large confidence intervals for each correlation and a high number of correlations (106) greatly increased the probability of Type I error. Previous research ( Harr ington & Gibson, 1986; Inman & Tolefson, 1988; Wilson et al., 1998) has suggested that general education teachers are typically lacking in knowledge of classroom based interventions. These results, although exploratory, suggest that teachers may have a gr eater knowledge base than previously thought; however, additional analysis of the present data might be helpful in establishing the quality and appropriateness of teachers intervention ideas, examining hypotheses regarding causes of student behavior, and determining the precise nature of teachers intervention ideas. Given the critical role that general education teachers play in the problem solving process (Tilly, 2002), such research will be important in determining exactly how teachers function in this role and how IA processes can better support them in their efforts.
126 References Albin, R. W., Lucyshyn, J. M., Horner, R. H., & Flannery, K. B. (1996). Contextual fit for behavioral support plan: A model for "goodness of fit ." In L. K. Koegel, & R. L. Koegel, et al. (Eds). Positive behavioral support: Including people with difficult behavior in the community ( pp. 81 98). Baltimore, MD, US: Paul H. Brookes Publishing Co. Bahr, M. W. (1994). The status and impact of prereferral intervention: We need a b etter way to determine success. Psychology in the Schools, 31 (4), 309 318. Baker, A. C. (1997). Relationship between problem solving skills and self efficacy in elementary general education teachers. Unpublished masters thesis, University of South Flor ida, Tampa. Batsche, G. M., & Knoff, H. M. (1995). Best practices in linking assessment and intervention. In A. Thomas and J. Grimes (Eds.), Best practices in school psychology III (pp. 569 585). Washington DC: National Association of School Psychologist s. Bergan J. R., & Tombari, M. ( 1976 ). Consultant skill and efficiency and the implementation and outcomes of consultation. Journal of School Psychology, 14 3 13. Brown, J., Gable, R., Hendrickson, R., & Algozzine, B. (1991). Prereferral intervention pr actices of regular education teachers: Implications for regular and special education preparation. Teacher Education and Special Education, 14 192 197. Chalfant, J. C., Pysh, M. V., & Moultrie, R. (1979). Teacher assistance teams: A model for within buil ding problem solving. Learning Disabilities Quarterly, 2 85 95.
127 Clarizio, H. F., & Halgren, D. W. (1993). Stability of special education placements: Data based policy consultation. Journal of Educational and Psychological Consultation, 4 (1), 49 67. Cort ina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98 104. Cohen, J. (1992). A power primer. Psychological Bulletin, 112 155 159. Cooper, J. O., Heron, T. E. & Heward, W. L. ( 198 7 ). Applied behavior analysis Upper Saddle River, NJ: Prentice Hall. Curtis, M. J., Hunley, S. A., Walker, K. J., Baker, A. C. (1999). Demographic characteristics and professional practices in school psychology School Psychology Review 28, 104 116. Curti s M. J., & Watson K. L. (1980) Changes in consultee problem clarification skills following consultation. Journal of School Psychology, 18 210 221. Curtis M. J., & Zins, J. E. (1988) Effects of training in consultation and instructor feedback on acqui sition of consultation skills. Journal of School Psychology, 26 185 190. Durda, C. M. ( 2000 ). Differences in problem solving skills between general education teachers and special education teachers. Unpublished doctoral dissertation, University of South Florida, Tampa. Education of All Handicapped Children Act of 1975, 20 U.S.C. Sec. 401 (1975). Feldman, E. S., & Kratochwill, T. R. (2004). Problem solving consultation in schools: Past, present, and future directions. Behavior Analyst Today, 4 (3), 318 33 0. Florida School Indicators Report (2003). Retrieved March 19, 2005 from http://info.doe.state.fl.us/fsir/ Flugum, K. R., & Reschly, D. J. (1994). Prereferral intervention: Quality indices and outcomes. Journal of School Psychology, 32 (1), 1 14. Fuchs, D., Fuchs, L. S., Bahr, M. W, Fernstrom, P., & Stecker, P. M. (1990). Prereferral intervention: A prescriptive approach. Exceptional Children, 56 (6), 493 513.
128 Fuchs, D., Fuchs, L. S., Bahr, M. W. (1990). Mainstream assistance teams: A scientific basis for the art of consultation. Exceptional Children, 57 128 139. Gilman, R., & Gabriel, S. (2004). Perceptions of school psychological services by educational professionals: Results from a multi state survey pilot study. School Psychology Review, 33(2), 271 286. Good, R.H., Simmons, D. C., & Smith, S. B. (1998). Effective academic interventions in the United States: Evaluating and enhancing the acquisition of early reading skills. School Psychology Review, 27 (1), 45 56. Graden, J. E. (1989). Redefining prereferral intervention as intervention assistance: Collaboration between general and special education. Exceptional Children, 56 227 231 Graden, J. E., Casey, A., & Bonstrom, O. (1983). Prereferral inte rventions: Effects on referral rates and teacher attitudes (Report No. 140). Minneapolis: Minnesota Institute for Research in Learning Disabilities. (ERIC Document Reproduction Service No. Ed 244 438) Graden, J. E., Casey, A., & Bonstrom, O. (1985). Imple menting a prereferral intervention system: Part II. The data. Exceptional Children, 51 487 496. Graden, J. E., Casey, A., & Christenson, S. L. (1985). Implementing a prereferral intervention system: Part I. The model. Exceptional Children, 51 377 384. Gresham, F. M. (1989). Assessment of treatment integrity in school consultation and prereferral intervention. School Psychology Review, 18 (1), 37 50. Grier, J. E. C. ( 2000 ). The relationship between teachers problem solving skills and their attitudes towards school inclusion. Unpublished educational specialist thesis, University of South Florida, Tampa. Grier, J. E. C. ( 2001 ). Factors relating to teachers attitudes towards school inclusion. Unpublished doctoral dissertation, University of South Flo rida, Tampa.
129 Hawkins, J., Martin, S., Blanchard, K. M., & Brady, M. P. (1991). Teacher perceptions, beliefs, and interventions regarding children with attention deficit disorders. Action in Teacher Education, 13 52 59. Harrington, R. G., & Gibson, E. (19 86). Preassessment procedures for learning disabled children: Are the effective? Journal of Learning Disabilities, 19 538 541. Individuals with Disabilities Education Act of 1997, Pub. L. No. 105 117 III Stat. 37 ( 1997 ). Individuals with Disabilities E ducation Improvement Act of 2004, Pub. L. No. 108 446 III Stat. 37 ( 2004 ). Inman, L., & Tollefson, N. (1988). Elementary teachers attitudes toward preassessment procedures. Psychology in the Schools, 25 331 337. Jacob Timm, S. & Hartshorne, T. S. (199 8). Ethics and law for school psychologists (3 rd edition). New York: John Wiley & Sons, Inc. Kling, B. (1997). Empowering teachers to use successful strategies. Teaching Exceptional Children, 30 (2), 20 24. Kovaleski, J. F., Gickling, E. E., Morrow, H., & Swank, P. R. (1999). High versus low implementation of instructional support teams: A case for maintaining program fidelity. Remedial and Special Education, 20 (3), 170 183. Kovaleski, J. F. (2002). Best practices in operating prereferral intervention t eams. In A. Thomas and J. Grimes (Eds.), Best practices in school psychology IV (pp. 645 655). Washington DC: National Association of School Psychologists. Kratochwill, T. R., Elliott, S. N., & Stoiber, K. C. (2002). Best practices in school based problem solving consultation. In A. Thomas and J. Grimes (Eds.), Best practices in school psychology IV (pp. 583 608). Washington DC: National Association of School Psychologists. Kratochwill, T. R., & Shernoff, E. S. (2004). Evidence b ased p ractice: Promoting e vidence b ased i nterventions in s chool p sychology. School Psychology Review, 33 34 48.
130 Kruger, L. J., Struzziero, J., Watts, R., & Vacca, D. (1995). The relationship between organizational support and satisfaction with teacher assistance teams. Remedial and Special Education, 16, 203 211. Lloyd, J. W., & Gambatese, C. (1990). Reforming the relationship between regular and special education: Background issues. In J. W. Lloyd, N. N. Singh, & A. C. Repp (Eds.). The regular education initiative: Alternative perspectives on concepts, issues, and models (pp.3 13). Sycamore, IL: Sycamore Publishing. Lloyd, J. W., Crowley, E. P., Kohler, F. W., Strain, P. S. (1988). Redefining the applied research agenda: Cooperative learning, prereferral, teacher consultation, and peer mediated interventions. Journal of Learning Disabilities, 21 (1), 43 51. Logan, K. R., & Stein, S. S. (2001). The research lead teacher model: Helping general education teachers deal with classroom behavior problems. Teaching Exceptional Children 33 (3), 10 15. Malouf, D. B., & Schiller, E. P. (1995). Practice and research in special education. Exceptional Children, 61 414 424. Mamlin, N. (1999). Despite best intentions: When inclusion fails. Journal of Special Education, 33 (1), 36 49. Mamlin, N., & Harris, K. R. (1998). Elementary teachers referral to special education in light of inclusion and prereferral: Every child is here to learn but some of these children are in real trouble. Journal of Educational Psychology, 90 (3), 385 396. McCarne y, S., Cummins Wunderlich, K., & Bauer, A. (1993). The pre referral intervention manual (2nd ed.). Columbia, MO: Hawthorne. McNamara, K., & Hollinger, C. (2003). Intervention based assessment: Evaluation rates and eligibility findings. Exceptional Childr en, 69 (2), 181 194.
131 Myers, C. L., & Holland, K. L. (2000). Classroom behavioral intervention: Do teachers consider the function of behavior? Psychology in the Schools, 37 271 280. National Association of School Psychologists (NASP) (1994). Assessment and eligibility in special education: An examination of policy and practice with proposals for change National Association of State Directors of Special Education. National Association of School Psychologists (NASP) (1995). Position statement: Rights witho ut labels. In A. Thomas and J. Grimes (Eds.), Best practices in school psychology III (pp. 1231 1232). Washington DC: National Association of School Psychologists. Nelson, J. R., Smith, D. J., Taylor, L., Dodd, J. M., Reavis, K. (1991). Prereferral inter vention: A review of the research. Education and Treatment of Children, 14 (3), 243 253. Newman, E. J. (1999). In the trenches: Increasing competency of teachers in training by having them conduct individualized interventions. Journal of Instructional Psy chology, 26 (1), 34 41. Noell, G. H., & Witt, J. C. (1999). When does consultation lead to intervention implementation? Critical issues for research and practice. Journal of Special Education, 33 (1), 29 35. Powell Smith, K. A., & Stewart, L. H. (1998). T he use of curriculum based measurement in the reintegration of students with mild disabilities. In M. R. Shinn (Ed.), Advanced applications of curriculum based measurement (pp. 254 307). New York: Guilford. Presidents Commission on Excellence in Special Education (2002). A new era: Revitalizing special education for children and their families Jessup, MD: ED Pubs. Pugach, M. C (1985). The limitations of federal special education policy: The role of classroom teachers in determining who is handicapped. The Journal of Special Education, 19 (1), 123 137.
132 Pugach, M. C., & Johnson, L. J. (1988). Peer collaboration: Enhancing teacher problem solving capabilities for students at risk. Paper presented at the annual meeting of the American Research Association New Orleans. (ERIC Document Reproduction Service No. SP 030 114) Raffaele, L. M., & Bradley Klug, K. L. (2000). Classroom interventions for children and adolescents with ADHD PowerPoint presentation, November 19, 2001. Rathvon, N. (1999). Effective sc hool interventions: Strategies for enhancing academic achievement and social competence. New York: Guilford. Reschly, D. J. (1988). Alternative delivery systems: Legal and ethical implications. In J. L. Graden, J. E. Zins, & M. J. Curtis (Eds.), Alternat ive educational delivery systems: Enhancing instructional options for all students (pp. 525 554). Washington, DC: National Association of School Psychologists. Safran, S. P., & Safran, J. S. (1996). Intervention assistance programs and prereferral teams: Directions for the twenty first century. Remedial and Special Education, 17 (6), 363 369. Sevcik, B., & Ysseldyke, J. (1986). An analysis of teachers prereferral interventions for students exhibiting behavioral problems. Behavioral Disorders, 11 109 117 Sindelar, P. T., Griffin, C. C., Smith, S. W., & Wanatabe, A. K. (1992). Prereferral intervention: Encouraging notes on preliminary findings. The Elementary School Journal, 92 (3), 245 259. Tilly, W. D., III (2002). Best practices in school psychology a s a problem solving enterprise. In A. Thomas and J. Grimes (Eds.), Best practices in school psychology IV (pp. 21 36). Washington DC: National Association of School Psychologists.
133 Upah, K. R. F., & Tilly, W. D., III (2002). Best practices in designing, i mplementing, and evaluating quality interventions. In A. Thomas and J. Grimes (Eds.), Best practices in school psychology IV, Volume I (pp. 483 501). Washington DC: National Association of School Psychologists. Viadero, D. (1994, October). The great div ide: The gap between research and practice is wider in education than in other fields such as medicine or business. Teacher Magazine, 22 24. Vogt, W. P (1999). Dictionary of statistics & methodology : A nontechnical guide for the social sciences (2 nd ed.) Thousand Oaks, CA : Sage Publications Wilson, C. P., Gutkin, T. B., Hagen, K. M., Oats, R. G. (1998). General education teachers knowledge and self reported use of classroom interventions for working with difficult to teach students: Implications for consultation, prereferral intervention and inclusive services. School Psychology Quarterly, 13 (1), 45 62. Wood, J. W., Lazzari, A., Davis, E. H., Sugai, G., & Carter, J. (1990). National status of the prereferral process: An issue for regular education. Action in Teacher Education, 12 50 56. Witt, J. C. (1996). Teachers resistance to the use of school based interventions. Journal of School Psychology, 24 37 44. Worthington, L. A., Wortham, J. F., Smith, C. R. B., & Patterson, D. (1997). Project facil itate: An inservice program for educators and parents. Teacher Education and Special Education 20 (2), 123 132. Ysseldyke, J. E., Pianta, B., Christenson, S., Wang, J., & Algozzine, B. (1983). An analysis of prereferral interventions. Psychology in the Schools 20 184 190.
135 Appendix A: Information Letter to Teachers You are invited to participate in a brief interview as part of a study at USF. Interviews will take place at XXX Elementary during the month of February 200 5, before or after school. An interview time will be individually scheduled with you if you decide to participate. Purpose of the Study : To learn more about teachers and what classroom based interventions they use with difficult to teach students. Abou t the Researcher : The Principal Investigator (PI) for this study is Jenine Sansosti, a graduate student in the school psychology program at the University of South Florida. This research project fills the requirement for the Ed.S. thesis as part of school psychology graduate training. Time Required : Approximately 20 25 minutes. Interview sessions will take place at your school, at a time that is convenient for you Jenine will contact you to schedule an interview. Format : Participation in this study consists of two parts: First, a brief questionnaire, to learn more about you as a teacher, will be put in your mailbox to complete on your own time. Next, during a meeting with Jenine, you will be given a brief description about a hypothetical student wi th academic and behavioral difficulties. You will be asked to think of various ways that you can help the student within your general education classroom. All information you share in the course of this study is completely confidential If you agree to participate in this study, please do not discuss the activities with other teachers at your school until data collection is complete, as it may influence the responses of others who are also participating in this study. Benefit to You and Your Students : For your participation in this research study, you will receive a $15 gift certificate to a local teacher supply store to use in your classroom. In addition, participating in this study may help you learn more about how to help children in your classroom who might be having difficulties.
136 Appendix A (continued) Interested? Heres wh at to do next : Complete the following contact information and return this form to Principal XXX. She will pass this on to Jenine, who will schedule an interview with you. Email address Best time to call Evening Phone May I call you at home? Best time to call Daytime Phone May I call you at school? Address ( optional so I can mail you the gift certificate) Grade You Teach S chool Name Your Name Please Check ALL Days/Times You Are Available for Interview This is just to help me get an idea your interview day/time will still be scheduled with you individually! Check if you are available on this day and circle which time would b e best for you: _____ Wednesday 02/02/05 Before school After school _____ Wednesday 02/09/05 Before school After school _____ Wednesday 02/16/05 Before school After school _____ Wednesday 02/23/05 Before school After school Other days/tim es you are available: ___________________________________________________________ Thank you in advance for making this project possible!! Jenine M. Sansosti, M.A., Doctoral Student in School Psychology 813/XXX.XXXX Fax: 813/XXX.XXXX
137 Appendix B: Summary of Pilot Results Participants Three individuals served as full protocol pilot participants. Two (1P3 and 2P2) were teachers at a local private school in Pasco County, and one (3PPreK) was a teacher at a local preschool. Each of these participants responded to the questionnaire and the structured interview, and also answered questions about the procedures and content of the study protocol. A fourth partic ipant (4PNG, NG indicating no grade), responded to the interview only (no questionnaire) to address an issue that arose from a change made to procedure. This is discussed in greater detail under Number of Interventions in the Changes to Interview Pro cedures section. Demographic Questionnaire Participants 1P3 and 2P2, teachers at a local private school, had significant difficulty responding to the questionnaire. They were unfamiliar with the term Child Study Team and did not have formal procedures for referring difficult to teach students to building level problem solving teams and did not have a school based or itinerant school psychologist with which they regularly consulted. Students with academic or behavioral challenges tended to be dealt wit h on an individual and somewhat informal basis. If school staff could not meet students needs, the students parents were recommended to get an evaluation by an outside psychologist or specialist. As such, participants 1P3 and 2P2 had difficulty respond ing to items 8 11 1 and 13c, 13d, and 13j. Further discussion with Linda and other school employees suggested that not all schools, whether public or private, have building level problem solving teams. As such, it was decided to add an item that specif ically asks if such a team exists. All references to such a team were modified to school based problem solving team, rather than Child Study Team or intervention team, to overcome the variability in terminology that may exist from school to school i n Hillsborough County. The order of items 11 (how many times have you consulted with an individual school psych or other ed. professional) and 12 (with which of the following individuals are you most likely to consult) was reversed. This was done because participants 1P3 and 2P2 both struggled to think of the number of occasions that they had consulted with someone about a student, but appeared to be better able to recall when they reviewed list of possible consultants on item 12. It seemed likely that p utting item 12 first could serve as a way to prime teachers for responding about their consulting behavior. Items 8 11, which require teachers to estimate instances of referral of consultation for the past three academic years, were altered to reference o nly the last two academic years. Teachers seemed to struggle to remember the third year. Jenine and Linda hypothesized that if they are unable to recall it with ease, their estimate is unlikely to be an accurate one. Item 13 was reworded, because part icipants 1P3 and 2P2 were both confused by the question. 1 Item numbers in this section reflect the numb ers on the original demographic questionnaire. As a result of modifications, the item numbers in the revised version no longer match the original.
138 Appendix B (Continued) Challenges in Interview Procedures Jenine coded pilot interviews for number of interventions, specificity, and type of intervention, and Linda conducted interrater reliability coding for participant 2P2. Items that were difficult to code were subsequently reviewed and discussed by Jenine and Linda. This led to suggested modifications for both the interview procedures and the intervention code structure. These problems and pr oposed solutions are discussed below. Number of Interventions In general, the coding form made counting interventions easy because they were written into a numbered space (e.g., #1___, #2 ___). However, on occasion it was difficult to determine at what point the teachers description of one intervention ended and another one began. This resulted in some disagreement on intervention numbers between Jenine and Linda and made reference to particular intervention numbers (e.g., 2.1 for Goal #2, Interventio n #1) complicated because they did not match on each coding form. To solve this problem and to potentially address some of the specificity problems described below, a modification to the interview procedure was proposed. Poker chips are now used to rep resent each individual intervention idea. Teachers are prompted to hold a poker chip in their hand while describing each thing they can do to help John and drop the chip into a plastic cup wh en finished describing an idea. The sound of the chip droppin g into the cup is clearly audible on the tape, so it is possible to hear when each intervention stops and another one begins. This strategy also creates a situation where teachers cannot go back to add to a description once theyve dropped it, so they need to describe it as thoroughly as possible before moving on to a new one. The poker chip strategy was implemented in the interview with 3PPreK, and she reported that it made sense to her and did not impede her description of ideas. However, she only provided three interventions for Goal 1 and two interventions for Goal 2. Because of the participants position as a preschool teacher, it was unclear whether her low number of interventions was due to an unfamiliarity with elementary age children (which she stated in follow up questions) or to the newly added poker chip procedure. To address this question, a fourth pilot participant (4PNG) was recruited. This participant is school psychologist and was expected to have many ideas in response to the vig nette. It was hypothesized that if the poker chip strategy was somehow interfering with ability to respond, this participant would also have relatively few intervention ideas. This participant, however, had eight ideas for Goal 1 and 5 ideas for Goal 2, and also stated that the poker chip strategy did not interfere with or distract during the interview in any way. Finally, two of the participants offered up the same intervention more than once for the same goal. As such, the procedure for counting the i nterventions was clarified to allow for the possibility of duplicate descriptions and non interventions, or statements that dont actually fit the definition of intervention.
139 Appendix B (Continued) Data collected subsequent to these interventions (partic ipants 3PPreK and 4PNG) revealed that intervention counting was greatly improved. The sound of the poker chip dropping made it clear when one intervention stopped and another began. Additionally, it was observed that participant 4PNG came close to droppi ng a chip into the cup on several occasions but then appeared to think twice and continue to add to the description before dropping it, thus improving the specificity of response. Specificity. Coding of the pilot participants responses revealed several problems. First, discriminating between low medium and high specificity responses was difficult due to ambiguous definitions of these categories. Thus, the codes for specificity were revised and clarified to allow for easier differentiation between low medium and high specificity. Second, none of the interventions described by any of the pilot participants could be categorized as high specificity. Examination of the interview directions read to each participant revealed that the sample of a hig h specificity response actually matched with the medium specificity definition. Thus, the interview directions were modified to provide low medium and high specificity examples of responses taken directly from the code definitions. A phrase was also added requesting participants to describe your ideas in a way that someone else would know exactly what to do, based on your description. Two additional changes were made based on follow up comments from participants 3PPreK and 4PNG. They both noted th at it would have been helpful to have been told about the specificity descriptions before they had read the vignette. They both indicated that this might have impacted the way they read the vignette and the ideas they generated. Additionally, participant 3PPreK asked if she could make notes while reading the vignette, and this was permitted for participant 4PNG, who indicated this was indeed helpful. A statement encouraging participants to make notes if necessary was subsequently added. Intervention Typ e. Upon trying to code the pilot data, it became clear that many of teachers responses could not be classified by the existing intervention code (adapted from Ysseldyke et al., 1989). The following are items that could not be coded under current coding structure. Parenthetical notation before each item indicates participant code (e.g., 1P3) and the goal/number of intervention to which item refers (e.g., 1.3 refers to Goal #1, intervention #3). The bullet below the problematic statement is a suggestion for a new classification code or revision to an existing code. (1P3 1.3) Have student set a goal, see if the student recognizes it [own behavior] o Communication (student) (1P3 1.4) Alert whole class to raise hand o Communication (whole class) (1P3 1.6) Tell student I will call on you o Communication (student) (1P3 1.7) Tell student(s) about the importance of not calling out; discuss how it disrupts others thinking o Communication (whole class)
140 Appendix B (Continued) (1P3 2.1) Use nonverbals o Communication (student) (1P3 2.11) Talk to student to see what they think o Communication (student) (1P3 2.14) Set timer or put dots on clock (e.g., dot on 3 and 6); You need to stay in your seat between the dots. o Communication (student) o Behavioral cu e (2P2 1.1) Develop a signal between the student and the teacher, a reminder to raise hand that no one else knows. o Communication (student) o Behavioral cue (2P2 1.2) Have a discussion (privately) about the effect on the classroom. Ask him some strategie s, as a team what does he think might work? o Communication (student) (2P2 2.2) Gently remind him o Communication (student) o Behavioral cue (2P2 2.3) Make sure he knows that its a problem (he may not know). Make sure he knows when its OK to be out of sea t. o Communication (student) (2P2 2.6) Give him some appropriate outlet to be out of seat we dont know, is he, does he have ADHD. Some duty when finished with work, an opportunity for John to be out of seat. He might not be able to control urge to get out of seat, so give him opportunities to get out of seat that are acceptable. o Modify classroom structure to accommodate this, because this changes the structure for the child. o Note the presence of the hypothesis (does he have ADHD) within this stateme nt. (2P2 2.8) (reference to vignette, part that states other students laughing) Discussion with class re: acceptance and respect for other students in class, considering feelings, hurt feelings o Communication (class) (2P2 2.9) Dont know if this is du e to frustration or for attention; other students reactions may be encouraging his behavior. o This is not really an intervention, but a hypothesis. (2P2 2.10) Might be feeling self conscious, recognizing differences in own behavior and that of other stud ents. Would work on building him up, achieve small successes. o Emotional/social support o Note the presence of the hypothesis (might be feeling self conscious) within this statement. (2P2 2.12) Putting in writing what goals are, a contract, can be helpfu l. Students feel some sort of accountability when its down in writing and they sign their name to it. o Communication (student) o This could be a behavioral intervention if there was some reference to consequences related to the contract.
141 Appendix B (Contin ued) As a result, the following modifications to the intervention code were made: 1. Addition of two new categories: Communication (specify student, whole class, parent/family) and Emotional/Social Support. 2. Changes to Behavioral and Classroom Structure cat egories to clarify their definitions. a. Behavioral was changed because of a recurring question about Behavioral Cues (or antecedent interventions), but these were later coded as Communication Student. b. Classroom structure was changed because several pilot i nterventions indicated changes in classroom environment or assignment of duties/privileges that did not fit under the current definition. Addition of a procedure for coding hypotheses mentioned within intervention descriptions. Hypotheses are not formally coded because they are not mutually exclusive with any other intervention category, but rather underlined in the text for later analysis. After making these modifications, data from participants 1P3, 2P2, and 3PPreK were recoded to check the utility of t hese modifications. All participants responses were successfully categorized with the revised coding structure. Follow up Questions To aid in the analysis of the questionnaire, interview protocol, and coding procedures, follow up questions were asked of each pilot participant. Each question is listed below with a summary of participants responses. Did you understand everything required for the questionnaire and interview portions of this study? 1P3: Questionnaire didnt feel very confident answer ing them because we dont have things like that. Interview felt comfortable with it, understood everything. Felt like I was in the role of the student. I felt like I was being tested. 2P2: Understood, yes, though some [questions] were not directly relev ant to me as a private school teacher, but yes, I understood. 3PPreK: Yes. 4PNG: No when saying what actions would you use to help achieve Johns outcomes, I wasnt clear on what actions you wanted to know. I didnt know if you wanted information about what I would do as an educator, or what interventions I could use. Did you want process, trying to look at the root cause of the problem, more of a problem solving approach?
142 Appendix B (Continued) Was there any point at which you did not know how to re spond? What was it? 1P3: No. 2P2: No. 3PPreK: Probably a little more difficult to respond to the story because it was older children. 4PNG: (see above answer) Were there any parts of the questionnaire or interview that were unclear because of the way th ey were worded? 1P3: Nothing unclear, though there was a typo. 2P2: No. 3PPreK: No. 4PNG: (see above answer) Was there any terminology that you did not understand in either the questionnaire or interview? 1P3: Only terminology was related to teams we d ont have in private school. 2P2: No. 3PPreK: No. 4PNG: No. By asking you to be so specific in your descriptions of interventions, did you think that you were limited in the number of things you could say due to time constraints? 1P3: Didnt feel time c onstraint, but it was a list. 2P2: No (laughs) it was the end of the day and it wasnt time or anything like that, it was just trying to think on my feet. In reality, I would sit and really think about this student, and I would have days and days of anecd otal info. I have a journal and Ive written things and behaviors when things come up, and then I would have a much clearer picture of whats going on, so its just kind of general strategies that you would use at this point. You dont have any background information, does this child need to be tested? That kind of thing. 3PPreK: Yes, probably because I was thinking of a whole scenario, instead of like individual steps. 4PNG: Somewhat. For me, it would almost be better to write down all the steps necessa ry in a semi detailed fashion and then talk, but that might just be a personal preference. Did the poker chip strategy make sense to you? 3PPreK: Yes, for the purpose that you described, it did. A lot of times, I feel that the interventions all run toge ther and theres not specific ones but for what youre trying to look for, that makes sense. 4PNG: Yes.
143 Appendix B (Continued) Did the poker chip strategy distract you when you were trying to come up with ideas? 3PPreK: No. 4PNG: No. This whole sessio n lasted 30 40 minutes. Do you think this is a reasonable amount of time to ask teachers to participate? 1P3: Yes. 2P2: Sure. 3PPreK: Yes not too short, not too long. 4PNG: Yes. Most of the issues raised in these questions have been addressed in t he above mentioned modifications to procedure. The only remaining issue was from 4PNG, who indicated that it was not clear if responses to the vignette should be in the form of interventions or if ideas related to problem solving and information gatheri ng were acceptable. During this interview, I simply said, Do your best and reiterated that the protocol reads give a detailed description of all of the ways that you know of to work with the child in your classroom to help him achieve those goals. Th is participant interpreted that direction as including more problem solving oriented actions, and as such 8 of 13 of this participants ideas were categorized as information gathering, which was already a part of the existing coding structure. This may be an issue to attend to, both in terms of training other data collectors to respond in the same way, and in terms of subsequent analysis. At the present time, all participants ideas are referred to as interventions but in fact the information category does not currently fit with the Fuchs, Fuchs, and Bahr (1990) definition of intervention as provided in the Coding Procedures. It may be better to refer to teachers responses as actions which may include information gathering and/or intervention strat egies. It would certainly be important to note the degree to which teachers are inclined to seek additional information before providing intervention strategies in response to the vignette. Conclusion Pilot data collection revealed that, in general, the questionnaire, interview protocol, and coding procedures were generally effective in eliciting desired responses from the participants. Several problems were brought to light, including difficulties in counting and coding interventions, less specific resp onses than desired, and ambiguities in questionnaire wording. These problems were addressed with multiple modifications all study procedures and materials. Changes in coding structure made it possible to code all pilot participants responses. Implement ation of a new strategy for counting interventions (poker chip strategy) dramatically improved counting of discrete intervention ideas. According to participant feedback, allowing participants to take notes improved ability to respond to the vignette. Finally, all participants stated that the length of the interview was acceptable to them.
144 Appendix C: Demographic Questionnaire Please tell us about yourself and your school 8. Does your school have a problem solving team that meets regularly to discuss teachers concerns about students academic or behavioral performance? (Circle yes or no and respond as directed if you answered no, proceed to #10) Yes We have a schoolwide problem solving team that meets on the following schedule (check appropriate item ): ___ weekly ___ monthly ___ as needed ___ other: ______________ No We do not have a schoolwide problem solving team. What happens at your school when you have concerns about a students performance? __________________________ __________________________ 9. Are you required to refer students with academic/behavior problems to a problem solving team before they can be referred to a school psychologist for special education eligibility testing? Yes We are required to attempt interventions for students with the problem solving team before they can be referred to a school psychologist. No We are not required to work with the problem solving team we can refer students to the school psychologist for testing at anytime. 1. Your age 2. Your gender a. Male b. Female 3. Your race/ethnicity a. White b. Black/African American/Caribbean Islander c. Hispanic d. Native American e. Asian/Pacific Islander f. Other (please describe ________________________) 4. How many total years have you been teaching, includi ng this year? (any grade level, not including internship) __________________________________________________ 5. Please list all Florida Certifications you hold. ______________________________________ ______________________________________ 6. What is the highest degree youve earned and in what area? a. High school diploma b. Bachelors in _________________________ c. Masters in ___________________________ d. Specialist in _________________________ e. Doctorate in _________________________ f. Other degree in ______________________ 7. What grade do you presently teach? a. 2 nd grade b. 3 rd grade
145 Appendix C (continued) Please tell us about your experiences in consulting with others regarding students with academic or behavior problems. 10. How many child ren have you referred to your schools problem solving team in each of the following years? *If your school does not have a problem solving team, please write N/A the lines to the right and move on to #11 Last Year (2003 2004) __________________________ T wo Years Ago (2002 2003) _____________________ 11. Of those above referred children, how many were eventually referred to the school psychologist or other personnel for evaluation for suspected disability in each of the following years? Last Year (2003 20 04) __________________________ Two Years Ago (2002 2003) _____________________ 12. Of those above referred children you referred for suspected disability, how many were eventually found to be eligible for ESE services in each of the following years? Last Year (2003 2004) __________________________ Two Years Ago (2002 2003) _____________________ 13. When you decide to consult an individual staff person regarding a difficult to teach student, with which of the following individuals are you most likely to consult? a. School psychologist b. School counselor c. Teacher (same grade level) d. Teacher (different grade level please specify __________) e. Special education teacher (please specify exceptionality taught ___________________________) f. Exceptional s tudent education (ESE) coordinator g. Specialist (e.g., reading, curriculum, etc.) h. Principal i. Other educational personnel (please specify ____________) 14. In each of the following years, how many times have you consulted an individual professional ( rather than in a team setting) about a difficult to teach student, such as a school psychologist or other education professional listed above? Last Year (2003 2004) __________________________ Two Years Ago (2002 2003) _____________________
146 Appendix C (c ontinued) Please tell us about your school problem solving teams practices with regard to students with academic or behavioral problems If you do not have a problem solving team at your school, please proceed to #16. Use these categories to guide your response to #15 DK Dont Know 1 Not at all 2 Rarely 3 Somewhat 4 Usually 5 Alwa ys a does the principal or assistant principal participate in team meetings? DK 1 2 3 4 5 b does the team look at schoolwide indicators (e.g., number of stude nts served by the team, number of students referred for special education, number of students retained) to determine the teams impact on the school as a whole ? DK 1 2 3 4 5 c does your school provide other opportunities to get information about inter ventions for students with academic/behavioral problems from inservice trainings, case studies, reading groups, etc.? DK 1 2 3 4 5 d are you (or someone else) required to collect data on the interventions you implement (e.g., curriculum based measurem ent data, baseline data, etc.)? DK 1 2 3 4 5 e does the team attempt to use intervention strategies with demonstrated research support ? DK 1 2 3 4 5 f does someone on the team assist you in getting interventions started in your classroom (e.g., a school psychologist demonstrates how to use a behavioral intervention)? DK 1 2 3 4 5 g does the team develop a plan to incorporate the intervention into your day to day instructional routine? DK 1 2 3 4 5 h does the team invite parents to partic ipate in selecting interventions for their children? 15. When you work with a problem solving team regarding concerns about a specific student DK 1 2 3 4 5
147 Appendix C (continued) Please tell us about your training experiences with regard to classroom based in terventions Use these categories to guide your response to #16 1 Not at all 2 Rarely 3 Somewhat 4 Often 5 Extensively a. Classes taken in college or graduate school 1 2 3 4 5 b. Inservice workshop(s) 1 2 3 4 5 c. Continuing Education Units (CEUs) obtained at non school workshops/professional conferences 1 2 3 4 5 d. Participation in intervention assistance teams or similar consultative groups 1 2 3 4 5 e. Supervised practice in developing and implementing interventions 1 2 3 4 5 f. Have taught/mentored others in developing and implementing interventions 16. To what extent have you participated in the following training experiences for learning about classroom based interventions for di fficult to teach students? 1 2 3 4 5 17. Do you feel you are adequately trained in the use interventions for classroom problems with difficult to teach students? 1 Not at all 2 A little bit 3 Somewhat 4 Mostly 5 Definitely Yes No 18. If not, would you like to receive additional training in developing and i mplementing classroom based intervention? I would like to learn more about interventions, especially about _______________________ _______________________ _______________________. I would not like to learn more about classroom based interventions at this time.
148 Appendix D: Interview Instructions 2 NOTE: ** Before you start the formal interview, you should explain to participants that you are required to read from a script to ensure consistency between all data collectors. This can help break t he ice and comfort them that you do know what you are doing this is just part of the procedure. J When we begin in a few moments, I am going to be asking you to read a vignette, or a made up description, about a child experiencing academic and behavior pr oblems in a third grade general education classroom. Along with this vignette, you will see two goals that have been developed to address this students difficulties. After youve had a chance to read this problem, I will be asking you to give a detailed description of all of the ways that you know of to work with the child in your classroom to help him achieve those goals. Please keep in mind that there are no right or wrong answers to any of the questions I will be asking in this interview. I am i nterested in hearing all the strategies you know of for solving our hypothetical problem! As you can see, the interview will be conducted privately between us. It should take about 20 minutes to complete. I will be audiotaping the interview, but your na me and any identifying information will be not be on the tape recording nor on the label. Following this session, I will use the audiotape to make a list of your ideas for helping the student. Your name will not appear on this list, nor will it be used i n any future publications or presentations resulting from this research. Another data collector may also listen to the tape later, to double check the accuracy of my list. Following the completion of the study, all interview tapes will be erased and lists will be discarded. I also want to remind you that your participation is strictly voluntary. If you would like to stop the interview, you may do so at any time. Lets begin with the vignette! 2 INTERVIEWER NOTES Bold type reflects phrases to be said aloud to interviewee. Parenthetical notes and non bold face type reflect interviewer instructions not to be read aloud. Do not use prompts to obtain additional information as teachers will already have been instructed to be as specific as possible.
149 Appendix D (continued) Heres how Id like you to tell me about your ideas for helping John. In front of you is a bag of poker chips. [Open bag so teacher can see inside] Each chip represents one complete idea for helping John improve in your classroom. Just as there are many, many things we can think of to help students, there are many chips in the bag. You do not have to use them all. When you are ready to describe an idea you have for helping John, [pick up a chip] take a chip from the bag and hold it in your hand. Describe your idea for helping John w ith lots of detail, and hold the chip the entire time you are describing it. When you are done with that idea and want to move on to another one, [drop chip into Goal 1 cup] drop the chip into the cup. Then pick up a new chip [pick up a chip] and repeat this process. Unfortunately, that means you cant go back to add to an idea once you have dropped it into the cup [point to cup] so try to describe your idea as completely as possible before you drop it [drop chip in cup] I also want to remind you to b e as specific as possible in your descriptions. Give as much detail as you can Try to describe what you would do in a way that is so clear that I, as another educational professional, would know exactly how to implement your idea just from hearing your description. Let me give you some example responses that provide low, medium and high levels of detail. While I give these examples, I will show you how to use the poker chips like I just described. [Get ready to demonstrate, but dont pick up yet!] If I asked you to describe the types of things you might do to help John succeed in the classroom, and you said, [pick up a chip] I could change the workload, [drop chip into Goal 1 cup and pause for 2 seconds] that would be a low detail response. That is too general and doesnt tell me exactly how you are planning to help John. If you said, [pick up a chip] I could shorten his daily math assignments, [drop chip into Goal 1 cup and pause for 2 seconds] that would give a medium amount of detail I ha ve a better idea of what you want to do, but Im still not completely sure how you would do it. Finally, if you said, [pick up a chip] I would take Johns math worksheets and cut them into strips of five problems each. When he finishes one strip, he wil l come up to my desk, and I will tell him hes doing a good job and give him another strip. This will break down his work into smaller chunks and allow him to get a brief rest and some praise in between sets of problems. [drop chip into Goal 1 cup and pa use for 2 seconds] this would be a highly detailed response. I would know exactly how to implement this idea based on your description. This is the kind of response were looking for.
150 Appendix D (continued) Do you have any questions about how to use th e chips, or how to describe your ideas in detail? [After answering any questions, empty Goal 1 cup and place bag and cup in front of teacher.] OK, then, lets move on to the vignette. Please take as much time as you need to read this material carefully and feel free to jot down any notes as necessary. When you are ready, I will be asking you to describe as many ways as you know of to reach the goals that are presented. Let me know when you have finished reading and are ready to discuss your ideas on ho w the goals for this child might be achieved. I will need to start the tape before you begin sharing your ideas. [Give vignette, scratch paper, and a pen to teacher and allow them to read. Continue when s/he indicates to do so] OK, then, why dont we ge t started. First, though, I need to start the tape and state your participant code, so later on, I know who were listening to. Ready? Start audiotape now! Pause 3 5 seconds, then state participant code (e.g., 2P2). Lets start with the first goa l, stop talking out in class? Describe all possible ways you know to help John achieve this goal. Remember, always hold a chip when youre describing an idea, drop it into the cup when youre done, and be as detailed as possible. Go ahead. When the t eacher has stopped suggesting interventions for the first goal, ask: Is there anything else you can think of to help John achieve the first goal? Continue asking the above question each time the teacher stops, until the teacher indicates that he/she cann ot think of any additional strategies. Then move Goal 1 cup off to the side (do NOT empty it!!) and place Goal 2 cup in front of the teacher. Continue with: Lets move on to the second goal, stay in his seat? Describe all possible ways you know to he lp John achieve this goal. Remember, always hold a chip when youre describing an idea, drop it into the cup when youre done, and be as detailed as possible. Go ahead. When the teacher has stopped suggesting interventions for the second goal, ask: Is there anything else you can think of to help John achieve the second goal?
151 Appendix D (continued) Continue asking the above question each time the teacher stops, until the teacher indicates that he/she cannot think of any additional strategies. Stop audiotape now! Thank teacher for his/her participation, and remind him/her not to discuss the activities of the study with other teachers at the school until all participating teachers have been interviewed.
152 Appendix E: Standardized Vignette PROBLEM STATEMENT John is a third grade student who is working slightly below grade level in the areas of Reading and Math. During the first few weeks of the school year, John appeared quiet and well behaved in your class of 22 children, but lately youve noticed several behaviors that concern you. For example, John has begun disrupting class by talking out on a regular basis. He frequently calls out the correct answers; however, many of his comments are loud and have nothing to do with the lesson. During perio ds designated for independent seatwork, John often plays with any objects left on his desk. He seems to be constantly out of his seat, either under his desk retrieving dropped articles, looking at the other students papers, or grabbing materials from the ir desks. On many occasions, John complains that he cant do this stuff, and it seems like you must redirect him back to task repeatedly until it is completed. While his daily worksheets and papers are messy, most of his answers are correct. His actio ns often draw some laughter from the other children in class. More often, however, his actions appear to annoy his peers. John has few friends. On many occasions, Johns behavior has disrupted all productive classroom activity and demanded a great deal of your attention. TEACHER GOALS You have decided that you want to help John learn to: 1. Stop talking out in class. 2. Stay in his seat.
153 Appendix F: Coding Form GOAL 1 : Stop talking out in class # ____ G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # ____ G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F Grade 2 3 DC: JS _____
154 Appendix F (continued) GOAL 2 : Stay in seat # ____ G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotio nal/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # ____ G Com munication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F Grade 2 3 DC: JS _____
155 Appendix F (continued) SUMMARY OF RESPONSES Total number of interventions offered: Goal 1: Goal 2: 1 2 3 4 5 6 7 8 9 10 Sum of Spec. Ratings (Sum across row) Total Freq. for this Type Mean Spec. Rating (A) Instruct. / = (B) Behavioral / = (C) Classroom structure / = (D) Interdiscip. Support / = (E) Information Gathering / = (F) Materials / = (G) Comm. (Student, Class, Family) / = (H) Emotional/ Social Support / = (I) Compound / = Total (Sum down) / = Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F Grade 2 3 DC: JS _____ Were any hypo theses offered? Yes No If so, how many?
156 Appendix G: Code Definiti ons NUMBER OF INTERVENTI ONS : 1. The final number of interventions is determined using the interview coding form. Each separate intervention should be written in a new intervention box. You should be able to hear on the tape the chip drop sound that separa tes one intervention from another. DO NOT number interventions as you go, but rather, enter interventions into each box and follow the procedures below before numbering. 2. After completing the transcription from tape to interview coding form, review non in terventions and duplicates. o Non interventions : Occasionally, teachers will describe ideas about student behavior that are not actually interventions (e.g., dont know if this is due to frustration or attention other students reactions may be encourag ing his behavior). Use the following definition to determine if the statement is actually an intervention: Interventions are defined as a teachers modification of instruction or classroom management to better accommodate a difficult to teach pupil wi thout disabilities (Fuchs, Fuchs, & Bahr, 1990). Thus, any actions a teacher describes that are offered with the intention of improving student behavior or performance may be considered interventions. Cross out any descriptions that do not fit this inter vention definition. o Duplicates : If interventions are mentioned more than once for the same goal, cross out the least specific version(s) of that intervention. It is OK for the same intervention to be suggested once for Goal 1 and then again for Goal 2. 3. Af ter coding interventions, the Total, Goal 1 and Goal 2 will go in the Summary of Responses Box at the end of the Coding Form.
157 Appendix G (continued) SPECIFICITY OF INTERVENTIONS : In the designated space below the intervention, highlight in black the spe cificity of the teachers intervention description. Determine the specificity of intervention descriptions using the following code (adapted from Gresham, 1989): Specificity Rating Definition Examples Low specificity descriptions consist of nonspecific or vague recommendations intervention could not be implemented based on current description alone I could use one of those B Mod things Use nonverbals I could give him more help in reading Moderate specificity description contains some, but not c omplete, detail intervention could be implemented if some additional details were to be provided Develop a signal between the student and the teacher, a reminder to raise hand that no one else knows. He could earn chips if he stays in his seat for the whole lesson High specificity descriptions demonstrate a detailed plan for assisting the hypothetical student intervention could be implemented on the basis of this description alone should not have questions about the who, what, when, where, why, how of the intervention During the recess period every other day, John and a paraprofessional would sit in the Reading Corner of the classroom and John would read aloud for 20 minutes. The para could keep track of errors and words read correctly per minute, an d she and John could chart his progress on a special graph. o If participant refers to a predetermined consequence system in the classroom, each step needs to be explained fully to receive a rating of 3. HYPOTHESES Within intervention descriptions, teache rs may hypothesize about potential causes of behavior. These are neither counted nor coded, but should be underlined for later analysis. (E.g., Might be feeling self conscious, recognizing differences between his Bx and other students Would work on bui lding him up, achieve small successes.)
158 Appendix G (continued) TYPE OF INTERVENTIONS : In the designated space below each intervention description, highlight in black the code for the type of intervention described. If an intervention clearly includes mo re than one type of assistance delivered to the student simultaneously, record it as a Compound intervention (I). Determine the type of intervention described using the following code (adapted from Ysseldyke et al., 1989): Intervention Type Definition Examples (A) Instructional A change in the teachers approach to instructing the child Individualized help with classroom work Restating directions (for academic work) Curriculum modifications (B) Behavioral Consequence oriented approach to change identi fied behavior, using positive or negative reinforcement, removal from reinforcement, or application of punishment. Differential reinforcement of alternative behaviors (target student or other students) Time out (removal from reinforcement) or other removal from classroom. Positive reinforcement in the form of praise, stickers/tokens/points, etc. (C) Classroom structure Changes in the amount of the structure provided for student within the classroom context. Not limited to instructional tasks may include changes to students responsibilities/duties that impact level of structure, or changes to the classroom environment as a whole. Move students seat Peer tutor/buddy Assign student duties to allow appropriate out of seat opportunities Allow student to sta nd or move while working, but in an appropriate, predetermined way. (Student) work with aide o Use of an aide is considered change in classroom structure rather than interdisciplinary support because it does not involve assistance from professionals of othe r disciplines. (D) Interdisc iplinary support Additional specialized assistance student receives directly from other school personnel. Pre taught vocabulary with the resource teacher, Counseling with the school counselor, Social skills training from scho ol psychologist (E) Information gathering Teacher requested or teacher gathered additional information regarding the student. Review the students cumulative file Call parents to ask questions about behavior at home o A call home in this context is consider ed information gathering rather than communication parents because its purpose is to get more information, not to make changes in student behavior. Refer to child study for additional evaluation Gather baseline data on problem behavior Continued on nex t page
159 Appendix G (continued) (F) Materials Specifically identified materials used to supplement instruction or remediation, such that the materials themselves are the primary intervention tool. Audio visual tapes Manipulatives (G) Communi cation Studen t Whole class Parent/ family Conversations, comments, or nonverbal cues directed at the student, class, or parent that are intended to change student(s) behavior Tell student about the importance of not calling out, discuss how it disrupts others thinking ; Ask student why they are out of seat or asking them what is going on for them at home; Remind student of classroom rules; Allow student opportunities to write down comments to share with teacher (journal, etc.) o Student Alert the whole class to raise thei r hand o Whole class Conference with parents to come up with a plan to change behavior at school and at home o Parent/family (H) Emotional/ Social Support Efforts on teachers part to provide emotional support to the student, increase students self esteem o r provide/enhance student friendships. Work on building him up, achieve small successes Talking one on one with the student for purposes of supporting concerns (not for information gathering or changing the behavior) o A student talking with a guidance couns elor would be considered interdisciplinary support because the teacher is not implementing this intervention. Pair him up with someone who can serve as a mentor/buddy o A peer buddy in this context is considered emotional/social support rather than change in classroom structure because its purpose is increase student friendships, not provide academic/behavioral support in the classroom. (I) Compound Intervention An intervention which consists of more than one code above. The intervention must be described i n a way that it is clear that the multiple components are intended to be delivered simultaneously. Developing a behavioral contract (B Behavioral), which is monitored by the guidance counselor (D Interdisciplinary Support) and which is sent home to p arents as a means of communication about his behavior (G Communication Family). o It is not necessary to note the other categories the intervention can be coded on, as done above. Simply code as Compound. o If two or more interventions are described but both came be coded the same (e.g., ignoring and differential reinforcement both behavioral) DO NOT code as compound. Code as behavioral.
160 Appendix H: Completed Coding Form for Participant 4A3 GOAL 1 : Stop talking out in class # 1 Well, John is going to need to know first of all his boundaries, making sure that those are clear and that he understands what the expectations are in the class. Once he understand s that were all in this together and when you disrupt people you keep them from learning, and going through the whole process, more of a one on one, spending that extra minute with him (Continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 2 Giving him that pat on the back of praise w hen hes doing good, and just really going through with him when hes making poor choices, giving him immediate feedback as to his warning, you know whatever signal it is that you signal. In my class, I use a sticky, and when you get stickied they know to stop and think, what is happening, what am I doing that is a poor choice? So once hes warned on his sticky, you know, when I verbally tell him thats hes stickied he should try to reflect. (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____
161 Appendix H (continued) # 3 And then at that p oint, if he doesnt, anothe r intervention that I might try would be setting something up with him as far as consequence. Which would be if he chose to keep doing what hes doing, then he would have to go through the steps of what the consequences are for everybody in the class. (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 4 It sounds like he is going to be an everyday disruption, so every child is different and what works for one child doesnt work for all of them. But usually when the connecti on is made, they respond to you, because they know you care about them and the other kids in the class, and they are going to want to do good for you. So I really personally think that relationship is the most important thing and the expectations for what you expect of him. So as long as he knows your expectations and he knows you care about him, then you are probably going to get a positive response where he wants to do good for you. (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____
162 Appendix H (continued) # 5 If theres not a response like that, it could be where you look out for something else happening, as far as home, things that are going on at home, which is another connection for you to work with his parents and see if they are seeing the same things going on at home. B ut really, take care of it right off the bat. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity DC: Is there anything else you can think of to help John achieve the first goal? # 6 We could go through things that he enjoys doing in the classroom, ways that you could reward him, whether it be 15 minutes or on the computer, whatever, something extra that makes him want to work quietly so he can earn something that is important to him and a little bit of something to celebrate his success (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity DC: (after a pause) If you have other ideas, just keep going. At the end, when youre done, Ill keep prompting you, but for now, if you have other things, just keep going. 4A 3 : Yeah, well, not really. You know, for the most part, th at works really well. In all the years, for me, it works well you know you might have one child that might be a severe disruption, but for the most part, as long as you are consistent, and they know exactly, immediately what their next consequence is goi ng to be, then they can control themselves. And if they are having a hard time controlling themselves, then you know thats something you sit down and go through with the parents as far as what youre faced with and kind of brainstorm together, what can w e do to fix this, because things arent working out for him and hes going to end up academically slipping. And that really throws the red flag up to the parents, but usually once the parents get involved things start to get better. DC: Again, so I just have to clarify, is there anything else that you can think of to help John achieve the first goal? 4A3: No Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____
163 Appendix H (continued) GOAL 2 : Stay in seat # 1 I think that it goes right along with the other, stop talking out in class. When he knows that the class is set up in a situation where hes going to have to be in his seat at times, and the whole class knows that they cant get up and come to you because the expectation is that if they need you then they can raise their hands and they can come to you. And again, when theyre not following procedure, whatever signal it is, whether its five fingers up, one finger up, whatever it is when you know that your student needs you and you need to come to them. I dont let them come to me. (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 2 If hes having a hard time out of his seat, you know, disturbing other people, then thats going to be an immediate consequence of being stickied and when that continues, his consequence would be our classroom procedure for when they a re breaking the rules (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 3 and just really having a little bit of lee way for him. Depending on, if hes standing up at his chair, I can deal with that, unless were having a test or something but you gotta have patience and understand that every child is different. Some kids have a hard time. (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____
164 Appendix H (continued) # 4 But if hes just flat out being defiant then hes going to have the consequence and go through what we talked ab out for the first goal of calling out, which is working for positive things, praise, and when he does good then at the end of the week he gets 30 minutes of free time. (continues on) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 5 Again, interaction with the parents to see wh at you can do to help him understand that home and classroom is very clear and connected. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity DC: Anything else you can think of for helping John stay in the seat? 4A3: No pretty much like I previously stated, a parent conference. Definitely a parent conference so they know whats going on and hopefully with that connection that he can get himself focused and back on track. DC: Ok, well, thanks 4A3: OK, no wait, let me tell you this, I want to say this. Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____
165 Appendix H (continued) # 6 For severe disruptions, they would eventually go through the Child Study Team for ideas and youll go through that process, because when its sever e disruptions, youve gotta look at alternative possibilities for the classroom. He might need a behavio r packet to see if hes got some emotional things going on Thats always a possibility too. G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity DC: OK, anything else? 4A3: No. Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____
166 Appendix H (continued) SUMMARY OF RESPONSES Total number of interventions offered: 12 Goal 1: 6 Goal 2: 6 1 2 3 4 5 6 7 8 9 10 Sum of Spec. Ratings (Sum across row) Total Freq. for this Type Mean Spec Rating (A) Instruct / = (B) Behavioral 2 1 1 1 2 7 / 5 = 1.4 (C) Classroom structure 1 1 / 1 = 1 (D) Interdiscip S upport / = (E) Information G athering 1 1 / 1 = 1 (F) Materials / = (G) Comm (Student, C lass, Family) 1 1 1 3 / 3 = 1 (H) Emotional/ Social Support 1 1 / 1 = 1 (I) Compound 2 2 / 1 = 2 Total (Sum down) 15 / 12 = 1.3 Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 03.02.05 Grade 2 3 DC: JS _____ W ere any hypotheses offered? Yes No If so, how many? 1
167 Appendix I: Completed Coding Form for Participant 3B3 GOAL 1 : Stop talking out in class # 1 I would give John a limit of questions. For example, I would give him 20 sticks, and each time he would like to ask a question, and any time he would like to ask a question about a subject, being math, reading, w riting, he would have to hand me a stick. It is up to him to decide he could use them all then, and if he doesnt have them at the time to ask another question, then he doesnt get to ask another question. The sticks are his limit to ask a question, and if he doesnt have one, then hes not allowed to do that. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 2 Basically, after reading this, he seems very bored and hes just doing this to get attention so I would try to pair him with another student that may be n ot popular, but Im just thinking of my own kids, I have my kids in different groups. Im trying to think of a personality thats not as strong as his, he seems to have a very strong personality so I would pair him with another personality thats maybe n ot as strong and help work together. I actually did that with one of my own students, and thats actually working because they work together in pairs. And he seems to be getting the attention he needs that I cant give at the time, you know, the peer wil l give him the attention he needs and I will withdraw and let the peer give him the attention he needs to work on the specific task. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ JH ___
168 Appendix I (continued) GOAL 2 : Stay in seat # 1 Again, going along with this, I still feel hes bored. So not necessarily to give him more work, but along the same lines of giving him his work in chunks, such as his math, reading, and writing. Give it to him in a portion, so that he raises his hand, tells me hes done, so he can look forward to me giving him more work so that he will say, Im done with this point, and I will go over, check his work, give praise, and then I say, OK, heres the next portion. Do this and when youre done, raise your hand, wait for me to come, dont come to me just constantly reinforcing that I will come to him. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotio nal/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ JH ___
169 Appendix I (continued) SUMMARY OF RESPONSES Total number of interventions offered: 3 Goal 1: 2 Goal 2: 1 1 2 3 4 5 6 7 8 9 10 Sum of Spec. Ratings (Sum across row) Total Freq. for this Type Mean Spec. Rating (A) Instruct. 3 3 / 1 = 3 (B) Behavioral / = (C) Classroom structure 3 3 6 / 2 = 3 (D) Interdiscip. Support / = (E) Information Gathering / = (F) Materials / = (G) Comm. (Student, Class, Family) / = (H) Emotional/ Social Support / = (I) Compound / = Total (Sum down) 9 / 3 = 3 Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ JH ___ Were any hypo theses offered? Yes No If so, how many? 2
170 Appendix J : Completed Coding Form for Participant 3D2 GOAL 1 : Stop talking out in class # 1 It seems like John has a lot of attention seeking behavior and so it seems to me it seems to me that he needs a lot of positive reinforcement and positive attention. I would probably start with some kind of chart on his desk just for him that when he was doing a good job and not talking out in class and he raised his hand, I could come give him a sticker on his chart some kind of positive reinforcement, so that he is getting the attention he needed in a positive attention in a positive way instead of a negative way. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 2 If the positive reinforcement doesnt work then we might have to move to some negative consequences, like taking away privileges, such as having a ticket or a c ard pulled every time he talked out in class, and if he gets a ticket pulled so many times, then he loses some kind of privilege like P.E. or fun centers, or something like that. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ DD __
171 A ppendix J (continued) # 3 The next step, what I would probably do is bring him some kind of outside help, like call a parent or ask the guidance counselor to come talk to him, or the principal or something, so he would be able to continue with his progress for not talking out in class. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ DD __
172 Appendix J (continued) GOAL 2 : Stay in seat # 1 The same as what I would do for the first one, I would have some kind of behavior modification chart on his desk, just for him, and if he could stay in his seat for a given amount of time, we would start off small like 10 or 15 minutes, and if he could stay in his seat in that 15 minutes then I could come over and give him a sticker. And if he earns so many stickers then he gets to go to the treasure box or some other kind of positive reinforcement. So I would start small with small time increments, and then when he got used to that, I would up that to half an hour of staying in his seat, or 45 minutes and so on, until hopefully he could stay in his seat all the time. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotio nal/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 2 Just like I s aid with the first goal, if that didnt work, we would probably have to move to something negative like taking away a privilege of some sort: teacher P.E., fun centers, something like that. (CHIP) G Com munication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ DD __
173 Appendix J (continued) # 3 And just like with the fi rst one, if both of those didnt work, I would probably send a letter home to the parent or talk to the guidance counselor. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specific ity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ DD __
174 Appendix J (continued) SUMMARY OF RESPONSES Total number of interventions offered: 6 Goal 1: 3 Goal 2: 3 1 2 3 4 5 6 7 8 9 10 Sum of Spec. Ratings (Sum across row) Total Freq. for this Type Mean Spec. Rating (A) Instruct. / = (B) Behavioral 2 2 3 1 8 / 4 = 2 (C) Classroom structure / = (D) Interdiscip. Support / = (E) Information Gathering / = (F) Materials / = (G) Comm. (Student, Class, Family) / = (H) Emotional/ Social Support / = (I) Compound 1 1 2 / 2 = 1 Total (Sum down) 10 / 6 = 1.67 Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.16.05 Grade 2 3 DC: JS __ DD __ Were any hypo theses offered? Yes No If so, how many? 1
175 Appendix K: Completed Coding Form for Participant 2E2 GOAL 1 : Stop talking out in class # 1 I would definitely explain to the class what the class rule is that we cannot talk out for disruptive reasons and I would provide nonexamples and examples of how we should respond with raising our hand. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 L ow Specificity 2 Moderate Specificity 3 High Specificity # 2 If the problem still continues, I would praise him every time he does not talk out in class. I would also give him a physical token that he can see: a sticker, points on a chart, something like that. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 3 I would also put him on an hourly contract, with the goal being just for talking out, to not talk out. And the hourly contract, out of 6 hours a day would first start out with him only needing to achieve 4 out of 6 to have success and the n as he improves, 5 out of 6, and then 6 out of 6. So an hourly goal chart. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
176 Appendix K (continued) # 4 I would also sit him next to a student who is a very good example for him to follow, a positive role model. And I would encourage the s tudent to help him not to call out. (CHIP) G Communication A Instruct B Behaviora l C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 5 I could also put him in a social skills group that meets with our school psychologist weekly. They work on using appropriate responses when students need something or would like to answer or make a comment or ask a question, and they work on a weekly basis on specific goals that the teacher would like to be worked on. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 6 I would also have a nightly contract that would go home every night to be signed by the parent and this way the parent knows what we are working on in class and at the end of the week, if he brings it back signed, he can received another reward or some sort of a token. (CHIP) G Communication A Instruct B Behaviora l C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
177 Appendix K (continued) # 7 I always have co operative learning groups in my classroom, where students work in teams and if he shows that he does not talk out in class, his team could receive bubbles on a surprise chart and he could work toward earning that surprise. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
178 Appendix K (continued) GOAL 2 : Stay in seat # 1 Sometimes students physically cannot stay still all the time; I would not expect him to do that at all times. If he is doing independent work, I would allow him to stand up as he does his work, as long as he is on task. So that would be one accommodation I would make for him as long as his is at his spot, then he will be allowed to stand to complete his work. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotio nal/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 2 Another thing I would do is add another goal to the contract of talking out, you can have two goals, but I would never have more than to goals. And stay in his seat would be the other goal that I monitor on an hourly basis. Starting with a goal of 4 out of 6, then working up to a goal of 6 out of 6. G Com munication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
179 Appendix K (continued) # 3 Praise I could have a buddy teacher that would regularly check on his progress, and when he can show that he has spent from morning until lunch in his seat or working hard to stay in his seat before lunch time, at midday he could go see that b uddy teacher and show his contract. Some sort of time where he is physically leaving the room, that gives him a small little break to get out of his seat because sometimes students who have this problem need that movement. So by doing that with a buddy t eacher, its not always me thats giving that consequence, hes got someone else he can show that he is making that progress. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specific ity 3 High Specificity # 4 Teacher proximity circulating around the class, as much as possible. If I see that he may be wanting to get up or get down to the floor, or whatever hes doing, patting his shoulder just to let him know Im here if you need something, you dont need to get up out of your seat. So teacher proximity of always moving around to show him t hat he doesnt have to get up to come to me, I can go to him. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificit y 2 Moderate Specificity 3 High Specificity # 5 Also, positive role models, those buddies who sit next to him and show the good examples of staying in your seat. A lot of it has to do with who they are sitting next to, someone that can get along with them and encourage them to do well. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
180 Appendix K (continued) # 6 Again, communication with the parents, letting them know what were working toward so that when he gets that communication goi ng home daily, they can reward him at home too or they talk to him, you know What is going on here, why are you constantly out of seat? (CHIP) G Communication A Instruct B Behaviora l C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 7 Looking at why hes getting out of his seat is it that he doesnt underst and the work ? I could definitely cut his workload down to see Tell him, I want you to do these 5 problems, give me a thumbs up when youre ready, Ill circulate back to your desk and check on you. Giving him shorter assignments and a silent signal to l et me know to come to him. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 8 I also do silent signals for water, bathroom, sharpening pencil. This way they know the code, and they dont have to get up or call out or be walking around the room. So silent signals f or routine procedures in the classroom. (CHIP) G Communication A Instruct B Behaviora l C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
181 Appendix K (continued) # 9 Another thing I could do is provide the child with a tally chart taped to his desk in the corner, and every time that I circulate to his seat and he is in his seat, I put a ta lly mark and every time they reach 5 tally marks to reward the behavior, they could get a sticker, they could get a piece of candy. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 10 We do I Spys where we can reward a student for staying in their seat, or a ce rtain goal theyre working towards, or they could go on the morning show to show an accomplishment or good thing theyve done, and that could be added as part of their reward. (CHIP) G Communication A Instruct B Behaviora l C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 11 I also would sit him in a low traffic area in the classroom. Really think about where that child should be sitting, where theres not so many distractions. Definit ely a low traffic area. (CHIP) G Communication A Instruct B Behavioral C Classroom Struct ure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.23.05 Grade 2 3 DC: JS __ IR _
182 Appendix K (continued) SUMMARY OF RESPONSES Total number of interventions offered: 18 Goal 1: 7 Goal 2: 11 1 2 3 4 5 6 7 8 9 10 Sum of Spec. Ratings (Sum across row) Total Freq. for this Type Mean Spec. Rating (A) Instruct. 3 3 / 1 = 3 (B) Behavioral 2 2 2 3 1 10 / 5 = 2 (C) Classroom structure 2 3 2 3 1 2 13 / 6 = 2.2 (D) Interdiscip. Support 3 3 / 1 = 3 (E) Information Gathering / = (F) Materials / = (G) Comm. (Student, Class, Family) 2 1 3 / 2 = 1.5 (H) Emotional/ Social Support / = (I) Compound 2 2 2 6 / 3 = 2 Total (Sum down) 38 / 18 = 2.1 Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F Grade 2 3 DC: JS _____ Were any hypo theses offered? Yes No If so, how many? 1
183 Appendix L: Completed Coding Form for Participant 4F3 GOAL 1 : Stop talking out in class # 1 The first thing that I would definitely do is get in touch with the parents and let them know that its become an increasingly more difficult problem, and tell them that I think that, especially since he already i s slightly below grade level, Im afraid that it would continue to inhibit his learning. So to just make them aware that I am going to try some interventions and also ask them do they have any suggestions, because maybe its something theyve seen at home or maybe they could tell me more about why this has happened, why all the sudden the change, and let them know what I plan on doing, and more specifically tell them these are the things Im going to do and ask them to ask him regularly so hes accounta ble to me and to them for how hes progressing. Ill either write a note in his agenda that goes home everyday to tell the parents, today was a good day, this is what happened or this is what were still working on. (CHIP) G Communication A Instruct B B ehavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 2 Then I woul d really focus on reminding him, by positive reinforcement of the other students, the appropriate way to respond in class. If calling out is now a problem for him and it wasnt as much before, he clearly just needs to be reminded of what the rules are. S o if he does call out, I would not respond to his answers, I would remind him, This is the proper procedure, we need to wait and raise our hand and wait to be called on and then I would praise the other students who did just by saying Thank you, I like the way you raised your hand and waited for me to call on you, so that hes reminded of that and also sees that you get good positive attention when you do that. Hopefully that will be a nice reminder and a gentle way to reinforce that behavior for him a nd for everyone else too. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
184 Appendix L (continued) # 3 I would have to make sure that he has the same rules and consequences as the other studen ts do. Even though I might implement some other interventions for him, he is still going to have to realize that within our behavior system, if you are warned, if its so severe, if its too the point that its disruptive and the other kids arent even ge tting a chance to answer because hes calling out so much, then I would have to say, OK, this is your warning, after that youre going have to follow the normal discipline procedure, which in my class would be first a hole punch on your behavior card, an d then if you get up to 5, he loses free time, things like that. So I would have to do that, even though I might be doing some additional things to help slow down the process of getting to that, that would be mandatory for him to realize that we are going to work on this together but I cant treat you differently than any of the other students, because if they broke the rule, this is what I would do so you have to have the same consequences. (CHIP) G Communication A Instruct B Behavioral C Classroom Structu re D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity # 4 I would also encourage him that maybe if he was wanting to answer so badly, especially for questions that are not right and wrong, if he just has an idea that he wants to share and get out there, Im not always going to be able to get to him. Hes one out of 22, and Im not always going to be ab le to hear his ideas, and if he feels that they are really extremely important, I will give him a piece of paper and make sure that he always has paper and he write those down for whatever question it is. Even if its just to go OK, that was my idea and he can know that he can share with me later so he still feels like I can tell her, shell know that I was on the right track, and I can reinforce that. And then Im giving him a little more one on one time later, even if its on the way to lunch we can look over that sheet and he can say this is what I was trying to say, this was my point, and I can let him know OK, thats right, where can we go and we can just have our own mini conference, even on a daily basis if we needed to just to help him feel like he can still be heard, I can still know what he was thinking, but in the proper setting and with the proper classroom discipline still in place. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gather ing F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
185 Appendix L (continued) # 5 I would also want to keep some kind of tracking system for him. If its such a big problem that its interrupting instruction, then they have all kinds of charts or grids that I could put on his desk and we would just st ick to the goal of not calling out in class. I would write it on the top of the chart, Monday through Friday, and I would definitely do it hourly, you know, check his progress and give him a smiley face if he did well that hour or a sad face if he didnt, and then develop some kind of reward system with him. Either if you do well for a day, you will get something that he enjoys, whether its computer time, or library time, or helping another teacher or, just a good note home to his parents. Or if you get three this week, you know, start with something small that he feels like he can accomplish, because if he doesnt feel like its possible its not going to work and hes not going to do it. So start with small goals, maybe the first week say, one hap py face a day, or a week, or wherever we need to start with him to let him see and then I would keep those, to show to his parents and for my own records so we could look at his progress. It would serve well as anecdotal notes for me on how hes been doi ng, and has it been working, has it been worth it, and have him tell me, what would you like to have as a reward if you get this? Would you like a popsicle or a homework pass or something like that that would help him feel like OK, I want to do this. And I think if he sees that chart its going to be an excellent visual reminder, and I wouldnt even have to take time away from other students to say anything, I could just point to that chart and he would know dont forget especially if its right ther e on the desk. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
186 Appendix L (continued) # 6 I pull smaller groups for reading and things like that, and I think that it would be very helpful for him to be in a smaller group like that where he is going to have more of a chance to talk, and so tapping into and reinforcing that behavior in that small group would be an excellent way for him to feel like you did it, Im proud of y ou, you didnt call out in the reading group just for like 20 minutes, you took your turn when you were supposed to, because usually when we do reading groups they can talk whenever they want, but it would be good for him if we could say alright, just fo r that particular reading group, if you have a comment, just knock on the table and if thats what works for him, he could do that in class. If thats what works for him, Id be fine with him carrying that over, so giving him a setting where he does feel like he can accomplish that. Because maybe it is more difficult in the larger group setting, he might feel like he is getting lost. So I would definitely make sure that when we are working in smaller groups that he is achieving that goal then too, and t hen taking that to determine where to go from there. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
187 Appendix L (continued) GOAL 2 : Stay in his seat # 1 Similar to wh at I would do for the other situation, I would definitely contact the parents and let them know that this is an increasingly difficult problem that were facing. And with something like staying in his seat, some students need to not always be like everyon e else. I would ask the parents if maybe do they have a hard time at home staying in their seat at the dinner table, are they able to just sit and watch TV or carry on a conversation without being up. And really, based on what the parents say is going to have a lot to do with what I decide to do. If the mom says, I just cant even get him to stay in his seat while we are having dinner, then I would think OK, maybe he needs something different. Or if they say, Yes, hes fine at home, I dont underst and why its a problem at school, then I would know which way to go from there. Because I would not have a problem if he just needs to stand, he can stand, thats fine with me. Or if maybe he just needs a different place, to sit to feel comfortable, tha ts fine, Im willing to work on that. But we need to know if its even possible first, so that would be the first thing I would do with the parents. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gather ing F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
188 Appendix L (continued) # 2 Then, also similar to the other one, I would have an incentive chart for him. Something like staying in your seat, he is clearly going to know whether or not hes doing it. If we decide that he can stay in his seat and that is what he needs to do, then I would once again put some chart on his desk. And this would probably be something that I might even have him take responsibility for Any time that you are out of your seat, I am just going to put my finger on that chart and you need to mark or tally the number of times that you are out of your seat. And then I would talk with him and say, OK, whats the reason? Is it just because youre bored, or do you want to want to get up and talk to ot her people? Why are you doing this, no one else is doing this and it cant be any different for you, unless you can give me a good reason why it should be. And I think that probably the first day of doing that would make him realize, oh, I am out of my seat 20 times, thats a little excessive! And once again, let him know that I am going to send that home, and just daily track this to see is it getting better, what can we do, and offer those little incentives along the way If you can just stay in you r seat while we are doing this reading group, then you can stand up for 10 minutes of math or whatever. And quite possibly a good incentive that would work for that would be maybe helping him going around, finding something that he can do, that he can ac hieve only if he can meet his goal of only if he can meet his goal of not getting out of his chair or staying in his seat or only getting out of his seat only 3 times a day. Start with the small goals and let him work up to that, and give him those reward s for that. Because maybe he just needs movement, so just say OK, if you can just stay in your seat during the reading lesson, then you can pass out the papers for the math lesson. And continue to give him that, and or you can go and stand next to a po ssible peer tutor that he might need, to give him something that he can accomplish, and a way to do it, and make it increasingly more difficult so that he can finally get that behavior down and under control. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specificity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
189 Appendix L (continued) # 3 Definitely having him look at the class around him and although its difficult to positively reinforce the other kids for that (like Thank you for s taying in your seat! because that really is going to single him out), but maybe if he was working in a smaller group setting, he would see No one else around me is up out of their seat, and just start to point that out to him about other children, just as a way to remind him of the rules. Not yelling from across the room, John, sit down! but just going over to him and saying, Just look at your classmates around you; everyone else is seated and it should be the same for you, thats our deal, thats ho w it works. So just reminding him Thats what the rules are, those are the expectations, and I dont have a reason to make them any different for you, so Im not going to. Youre going to have to stay in your seat or were going to have to continue wit h the discipline procedures. (CHIP) G Communication A Instruct B Behavioral C Classroom Structure D Interdisc. Support E Info Gathering F Matl s Student Class Family H Emotional/ Social Support I Compound 1 Low Specific ity 2 Moderate Specificity 3 High Specificity Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____
190 Appendix L (continued) SUMMARY OF RESPONSES Total number of interventions offered: 9 Goal 1: 6 Goal 2: 3 1 2 3 4 5 6 7 8 9 10 Sum of Spec. Ratings (Sum across row) Total Freq. for this Type Mean Spec Rating (A) Instruct / = (B) Behavioral 3 3 6 / 2 = 3 (C) Classroom structure 3 3 / 1 = 3 (D) Interdiscip S upport / = (E) Information G athering / = (F) Materials / = (G) Comm (Student, C lass, Family) 3 3 3 9 / 3 = 3 (H) Emotional/ Social Support / = (I) Compound 3 3 3 9 / 3 = 3 Total (Sum down) 27 / 9 = 3 Participant 1 2 3 4 5 6 7 8 Date of Iview: School P A B C D E F 02.17.05 Grade 2 3 DC: JS _____ W ere any hypotheses offered? Yes No If so, how many? n/a
191 Appendix M : Correlation s Among Specificity Rating s Behavioral Intervention Specificity C'rm Structure Intervention Specificity Comm unication Intervention Specificity Compound Intervention Specificity B ehavioral Intervention Specificity 1.00 C'rm Structure Intervention Specificity .314 1.00 Communication Intervention Specificity .477 .574 1.00 Compound Intervention Specificity .356 .430 .703 1.00
Appendix N: Point Biserial Correlations (r p b ) Between Selected Teacher Characteristics and Intervention Types Suggested Intervention Types Suggested Teacher Characteristics Instructional Behavioral Classroom Structure Interdisc. Support Info. Gathering Materials Communication Emotional/ Social S upport Compound Years of Teaching Experience ( N =29) .190 .169 .181 .305 .012 .247 .147 .052 .488 Frequency of P articipating in IA Teams ( N =22) .040 .143 .029 .434 .167 .396 .064 .056 .164 Referral to E ligibility R ate ( N =24) .163 .295 .057 .006 .01 9 .028 .517 .059 .105 Composite T raining E xperience s Score ( N =29) .005 .074 .346 .009 .176 .263 .162 .162 .217 Composite IA P ractices of S chool Score ( N =25) .191 .020 .064 .119 .117 .194 .152 .170 .050 N values for each correlation are reported due to the irregularities in responses to questions about IA teams. 192
Appendix O: Phi Coefficients ( r ) Between Intervention Types Instructional Behavioral Classroom Structure Interdisc. Support Info. Gathering Materials Communication Emotional/ Social Sup port Compound Instructional 1.000 Behavioral .136 1.000 Classroom Structure .125 .139 1.000 Interdisciplinary Support .139 .074 .139 1.000 Information Gathering .086 .112 .025 .183 1.000 Materials .125 .139 .261 139 .159 1.000 Communication .036 .136 .125 .136 .247 .125 1.000 Emotional/Social Support .256 .139 .051 .197 .025 .159 .125 1.000 Compound .313 .068 .224 .213 .214 .048 .154 .048 1.000 193