RTI in Practice

A Practical Guide to Implementing Effective Evidence-Based Interventions in Your School

James L. McDougal, Suzanne B. Graney, James A. Wright, Scott P. Ardoin

• Publisher: Wiley
• Year: 2009
• Print ISBN: 9780470170731
• eBook ISBN: 9780470564493
• Edition: 1
• Topic: Psychology
• Subtopic: Research & Methodology in Psychology
• Index: Psychology

A must-have resource for educational professionals implementing Response to Intervention (RTI) programs in their schools

RTI in Practice: A Practical Guide to Implementing Effective Evidence-Based Interventions in Your School is an innovative and timely guide that presents concrete, balanced perspectives and directions for implementing an effective RTI model in your school.

Built upon the three tiers of effective general education and universal screening, targeted interventions, and intensive interventions, this authoritative resource addresses:

• Effective academic programs for all students
  
  

• How to use data to make decisions in general education
  
  

• Guidelines for setting goals, monitoring progress, and graphing intervention outcomes
  
  

• Multicultural considerations
  
  

Realistic case scenarios appear throughout to bring the implementation strategies to life, and the book is packaged with a CD-ROM containing numerous reproducible and customizable forms, surveys, and screening tools, as well as an annotated list of resources for charting and monitoring individual student and classroom progress.

RTI in Practice: A Practical Guide to Implementing Effective Evidence-Based Interventions in Your School is a complete resource providing educators and school professionals with the tips¿and tools needed for successful RTI program implementation.

PART I
INTRODUCTION

Chapter 1:
History of Learning Disabilities and Emergence of a New Model

LEARNING DISABILITIES: DEFINITION AND BACKGROUND

The concept of learning disabilities dates back to the early 1960s. In 1968 the label of “specific learning disability” was added as a federally designated category of handicapping conditions (Hallahan, Kauffman, & Lloyd, 1999). One of the first to address the definition of learning disabilities was Samuel Kirk. In 1962 Kirk wrote:

A learning disability refers to a retardation, disorder, or delayed development in one or more of the processes of speech, language, reading, writing arithmetic, or other school subject resulting from a psychological handicap caused by a possible cerebral dysfunction and/or emotional or behavioral disturbances. It is not the result of mental retardation, sensory deprivation, or cultural and instructional factors (Kirk, 1962, p. 263).

In Kirk’s description can be seen many components of the modern definition including a conceptualization that LD (1) is a deficit in processing (2) that results in reduced academic performance in one or more areas, (3) is possibly related to a cerebral (pertaining to the central nervous system) dysfunction, and (4) is not the result of other handicapping conditions. Later in 1965, Barbara Bateman proposed a modified definition of learning disabilities that removed emotional factors as causal in LD and more significantly suggested that it could be identified by an “educationally significant discrepancy” between estimates of intellectual potential and actual-performance level (for discussion, see Hallahan, Kauffman, & Lloyd, 1999; Smith, 1998). This discrepancy notion was further supported by the epidemiological work of Rutter and Yule in the early to mid-1970s. By studying the IQ predicted reading achievement of children ages 9 to 13 on the Isle of Wright they concluded that there was an abnormal distribution of reading performance scores suggesting that (1) reading underachievement occurred at a higher than expected rate and (2) that different patterns of sex distribution and of neurological deficit and development were observed in the “under achievement” group (Rutter & Yule, 1975). Thus support for the first severe discrepancy provisions for learning disabilities emerged.

THE HISTORY OF LD

Arguably the most important landmark legislation providing rights and educational privilege to students with disabilities was PL 94–142 enacted by Congress in 1975. Prior to 1975 approximately 200,000 individuals with significant disabilities were institutionalized in state-run settings and generally provided minimal standards of care (Ed.gov. 5/21/2007). Further, in 1970 only one in five children with disabilities was educated in public schools. Perhaps one of the most debated classification categories in the PL 94–142 regulations was with respect to learning disabilities.

While crafting a definition of LD in 1976 for the PL 94–142 regulations, the United States Department of Education (USDOE) considered the addition of a severe discrepancy formula (e.g., achievement falling 50 percent or more below the child’s expected achievement level) within the LD definition. While these efforts were offset by a number of objections from national experts of the time offering an array of conceptual and statistical difficulties with this procedure, the notion of seemingly objective discrepancy criteria was not entirely abandoned. The final definition for learning disabilities in PL 94–142 was as follows:

The term “specific learning disability” means a disorder in one or more of the basic psychological processes involved in understanding or using language, spoken or written, which may manifest itself in an imperfect ability to listen, speak, read, write, spell, or to do mathematical calculations. The term includes such conditions as perceptual handicaps, brain injury, minimal brain dysfunction, dysfunction, dyslexia, and developmental aphasia. The term does not include children who have learning disabilities which are primarily the result of visual, hearing, or motor handicaps, or mental retardation, or emotional disturbance, or of environmental, cultural, or economic disadvantage. (U.S. Office of Education, 1977, p. 65083)

While the actual definition in the pivotal regulations did not include a severe discrepancy formula, the section of the law that identified criteria for identifying students with learning disabilities stipulated that:

1. A team may determine that a child has a specific learning disability if:
   1. The child does not achieve commensurate with his or her age and ability levels in one or more of the areas listed in paragraph (a) (2) of this section, when provided with learning experiences appropriate with the child’s age and ability levels; and
   2. The team finds that a child has a severe discrepancy between achievement and intellectual ability in one or more of the following areas:
      1. Oral expression;
      2. Listening comprehension;
      3. Written expression;
      4. Basic reading skill;
      5. Reading comprehension;
      6. Mathematics calculation; or
      7. Mathematics reasoning
2. The team may not identify a child as having a specific learning disability if the severe discrepancy between ability and achievement is primarily the result of:
   1. A visual, hearing, or motor handicap;
   2. Mental retardation;
   3. Emotional disturbance; or
   4. Environmental, cultural, or economic disadvantage. (Federal Register, Dec. 29, 1977, p. 65083)

Therefore, while the severe discrepancy language did not make it into the formal LD definition, the inclusion of the preceding language essentially added these procedures to the classification. Following the publication of PL 94–142 most states adopted severe discrepancy provisions in their identification procedures for learning disabilities (e.g., Frankenberger & Franzalglio, 1991). However states varied in terms of the tests used to ascertain a discrepancy, the formulas used to compute the discrepancy, and the magnitude required for identification purposes (for discussion, see Fuchs, Mock, Morgan, & Young, 2003).

CRITICISMS OF DISCREPANCY-BASED MODELS

Criticisms of discrepancy-based models for understanding and identifying learning disabilities are numerous and have a long history. Essentially these criticisms can be conceptualized along two domains: problems with the reliability of a discrepancy-based approach for identifying students with disabilities and problems with the discrepancy-based model for conceptualizing and treating students with learning disabilities. Therefore, basic criticisms of the discrepancy-based model are that this method for understanding and identifying learning disabilities lacks adequate reliability and validity. In terms of reliability, the 300 percent increase noted in the population of students identified with learning disabilities over the last 30 years suggests a lack of stringent criteria for making the diagnosis (President’s Commission on Excellence in Special Education, 2002).

PROBLEMS WITH RELIABILITY

One specific difficulty hampering reliable diagnosis is that there are four major methods for determining the presence of a severe discrepancy and each uses different criteria. The methods include assessing the discrepancy in terms of (1) deviation from grade level, (2) Standard deviation from the mean, (3) Standard Score comparison, and (4) Standard Regression analysis. The first, deviation from grade level, suggests that if Kate is in the fourth grade yet reads at a second-grade level then she may be seen as having a severe discrepancy in her reading achievement. In this method Kate’s academic performance is compared to her peers. The second method, standard deviation from the mean, might assess Kate on an individually administered achievement test. Given that her score overall or in a specific academic area was at least a standard deviation below the norm she may be perceived as evidencing a severe discrepancy commensurate with an LD diagnosis. This method would compare Kate’s achievement with that of a standardized sample of same-age students from across the country. In the third method, Standard Score comparison, Kate’s performance on an individually administered intelligence test would be compared to her performance on an individually administered achievement test. If she achieved an IQ score of 100 (average score) and an achievement score one or more standard deviations below the mean, she may be seen as evidencing a severe discrepancy commensurate with an LD diagnosis. With this method Kate’s academic performance is compared to her performance on an intellectual assessment. Given that the comparison groups for Kate’s academic performance differ across these three methods (e.g., compared to peers, a national sample, and to her own IQ score), it is not hard to imagine why the result would be different for students diagnosed as learning disabled depending on the discrepancy method utilized. In essence, different methods of calculating a discrepancy will result in different students being classified. The fourth method, Standard Regression analysis, utilizes the Standard Score comparison technique and additionally employs a regression formula as an attempt to statistically account for the measurement error associated with the tests, the reliability of them, and the correlations between them. While this is perhaps the most psychometrically sound method for assessing IQ/ achievement discrepancies, it is not without additional inherent difficulties.

In a replication of an earlier study Mercer, Jordan, Allsopp, and Mercer (1996) surveyed all state education departments in the United States and found that 98 percent of them included a discrepancy in their definition of and identification criteria for learning disabilities. As indicated in the 1997 NYS Part 200 Regulations of the Commission of Education, “a student who exhibits a discrepancy of 50% or more between expected achievement and actual achievement determined on an individual basis shall be deemed to have a learning disability.” This determination in contemporary assessment was often completed using an intelligence test as the measure of expected achievement and a norm-referenced, standardized, academic test as a measure of actual achievement. The difference between the two scores is used to assess the discrepancy.

This brings us to the second major difficulty significantly hampering the reliability of LD diagnoses made with discrepancy based methods: The norm-referenced, standardized measures commonly employed in this assessment process are inadequate for measuring both expected achievement and actual achievement. In terms of expected achievement, while IQ tests are good general predictors of educational attainment they are inadequate for assigning an expected achievement outcome for individual students for several reasons. First, IQ test components most linked with reading performance are often verbally mediated and are somewhat dependent on reading. Therefore poor readers may have lower verbal IQ test scores and therefore be denied special education services due to a lack of assessed discrepancy (see Siegel, 1989; Stanovich, 1989). Secondly, this approach assumes that IQ can accurately predict academic performance. To explore this further we can look at the correlations between IQ and achievement reported on the most recent version of a popular standardized achievement measure, the Wechsler Individual Achievement Test-Second Edition (WIAT-II, 2002). The examiner’s manual of the WIAT-II reports that the correlations between full-scale ability (assessed by the WISC-III) and achievement (assessed by the WIAT-II) range from .3 to .78. To understand how well the WISC-III predicts achievement we can square these correlations to determine the amount of shared variance between these scores. The result suggests that the WISC-III accounts for 9 to 61 percent of the variance in a given student’s achievement test score. This also suggests that from 39 to 91 percent of the student’s achievement score is not accounted for by the IQ test. This lends considerable doubt to the notion that an IQ test can accurately assign an expected level of achievement, at least at the level of the individual student. Second, with respect to actual achievement, the concept that a student’s actual academic performance can best be assessed with a norm-referenced test administered at a single point in time has received considerable criticism as well. Among these criticisms are that nationally normed standardized achievement assessments often do not reflect the skills in a given local curriculum, they suffer from regression to the mean effect, and the fact that all psychometric tests include measurement errors that vary across students and across characteristics of the student (see Francis, Fletcher, & Morris, 2003). In single point assessments measurement error creates fluctuations in test scores that vary by test, age, ability level, and ethnicity. Applying cut-off scores to these types of score distributions is problematic since there is generally little or no actual difference between children at or around that cut-off regardless of their assigned status. Score fluctuations (above or below assigned cut-off scores) have been assessed in both real and simulated data sets suggesting that up to 35 percent of cases change status based on measurement error when single tests were used. Similarly, with respect to discrepancy scores, actual data from the Connecticut Longitudinal Study, analyzed by Francis et al. 2005, found that approximately 20 percent to 30 percent of students studied change disability status from third to fifth grade based on discrepancy scores.

Given the cited limitations with the discrepancy model it is easy to see how it lacks reliability in diagnosis. The fact that different criterion are used across different states significantly impairs consistency in identification. In addition, the limited ability of IQ tests to predict the achievement of an individual measurement error, and the difficulties associated with assigning cut-offs in either single test or discrepancies between tests significantly limit the reliability of this approach. In sum the use of discrepancy-based psychometrically oriented models for diagnosis are unreliable and insufficient to accurately designate individuals with learning disabilities (Francis, et al., 2005; Fletcher et al., 2005).

PROBLEMS WITH VALIDITY

In addition to reliability concerns, discrepancy-based models also h...

Table of contents

• Cover
• Contents
• Title Page
• Copyright
• Preface
• Part I: INTRODUCTION
• Part II: TIER I: EFFECTIVE GENERAL EDUCATION AND UNIVERSAL SCREENING/PROGRESS MONITORING
• Part III: TIER 2: TARGETED INTERVENTIONS AND PROBLEM SOLVING FOR STUDENTS AT RISK FOR FAILURE
• Part IV: TIER 3: INTENSIVE INTERVENTIONS/ INDIVIDUAL EDUCATION PLAN CONSIDERATION AND DEVELOPMENT
• Part V: ORGANIZATIONAL CONSIDERATIONS AND CONCLUSIONS
• References
• Author Index
• Subject Index
• About the CD-ROM
• Wiley End User License Agreement