Journal of Learning Disabilities http://ldx.sagepub.com/

Qualitative Differences in Learning Disabilities Across Postsecondary Institutions Robert Weis, Lauryn Sykes and Devanshi Unadkat J Learn Disabil 2012 45: 491 originally published online 18 March 2011 DOI: 10.1177/0022219411400747 The online version of this article can be found at: http://ldx.sagepub.com/content/45/6/491

Published by: Hammill Institute on Disabilities

and http://www.sagepublications.com

Additional services and information for Journal of Learning Disabilities can be found at: Email Alerts: http://ldx.sagepub.com/cgi/alerts Subscriptions: http://ldx.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://ldx.sagepub.com/content/45/6/491.refs.html

>> Version of Record - Oct 9, 2012 OnlineFirst Version of Record - Mar 18, 2011 What is This?

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

Journal of Learning Disabilities 45(6) 491­–502 © Hammill Institute on Disabilities 2012 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0022219411400747 http://journaloflearningdisabilities .sagepub.com

Qualitative Differences in Learning Disabilities Across Postsecondary Institutions Robert Weis, PhD,1 Lauryn Sykes1, and Devanshi Unadkat1

Abstract Many college students receiving accommodations for specific learning disability (SLD) do not meet objective criteria for the disorder. Furthermore, whether students meet criteria depends on the diagnostic decision model used by their clinician. The authors examined whether the relationship between diagnostic model and likelihood of meeting objective criteria is moderated by students’ postsecondary institution. They administered a comprehensive psychoeducational battery to 98 undergraduates receiving accommodations for SLD at 2-year public colleges, 4-year public universities, and 4-year private colleges. Most 4-year public university students failed to meet objective criteria for SLD. In contrast, most 4-year private college students met objective criteria based on significant ability–achievement discrepancies, and most 2-year public college students met objective criteria based on normative deficits in achievement and cognitive processing. Students who met objective criteria also differed significantly in degree of academic impairment. The authors’ findings indicate qualitative differences in SLD across postsecondary settings and have implications for the identification and mitigation of SLD in college students. Keywords specific learning disability, diagnosis, adults, postsecondary education

There is currently no consensus regarding an operational definition for specific learning disability (SLD) in college students (Fletcher, Lyon, Fuchs, & Barnes, 2007; Gregg, 2003). The Individuals with Disabilities Education Improvement Act (IDEIA, 2004) and corresponding regulations, which provide guidelines for the assessment of SLD in children, do not apply to individuals enrolled in college (Gregg, 2009a). The Diagnostic and Statistical Manual of Mental Disorders– Fourth Edition (DSM-IV; American Psychiatric Association [APA], 1994) diagnostic criteria for learning disorders do apply to adults; however, they include the presence of ability–achievement discrepancies, which appears to lack validity as a means of identifying SLD in children (Hoskyn & Swanson, 2000; Stuebing et al., 2002) and adults (Swanson, 2009). Most recently, comprehensive neuropsychological models of SLD have been developed, which include the presence of normative deficits in academic achievement, underlying cognitive processing problems, and a history of academic impairment in their criteria (Flanagan, Ortiz, Alfonso, & Dynda, 2006; Goldstein & Schwebach, 2009). However, these models have not gained widespread use among clinicians. This lack of consensus has reduced the reliability of SLD identification in college students. Two students, assigned the same SLD label and awarded the same academic accommodations, may have different patterns of academic and cognitive

ability, educational histories, and degrees of impairment. Two other students, one with and the other without SLD classification, may show similar levels of cognitive functioning and academic achievement. Indeed, recent research indicates that whether a college student meets objective criteria for SLD, and is given academic accommodations for this condition, depends largely on the assessment method and underlying operational definition of SLD used by his or her clinician (Proctor & Prevatt, 2003; Sparks & Lovett, 2009b). The presence or absence of the disorder, therefore, depends less on the characteristics of the individual student and more on the conceptualization of the clinician assigning the diagnostic label.

Reliability of SLD Identification Among University Students Two large studies have documented the lack of reliability in SLD identification among college students. Proctor and Prevatt (2003) tested 170 college students referred to a university-based 1

Denison University, Granville, OH, USA

Corresponding Author: Robert Weis, Denison University, 100 College Road, Granville, OH 43023 Email: [email protected]

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

492

Journal of Learning Disabilities 45(6)

clinic for academic problems. They found that the percentage of students meeting criteria for SLD varied as a function of the diagnostic decision model that was used. Specifically, 46.5% of students displayed a simple (1 SD) ability–achievement discrepancy, 24.7% showed a significant (1.3 SD) discrepancy, and 33.1% had normatively low standard scores (i.e., < 85) on at least one achievement domain. Similar results were obtained by Sparks and Lovett (2009b), who reviewed the psychological evaluation documentation of 378 university students diagnosed with SLD. In their study, 41.9% of students showed a simple (1 SD) discrepancy, 23.8% showed a significant (1.5 SD) discrepancy, and 6.9% met DSM-IV criteria (i.e., significant discrepancy and normatively low achievement). In both studies, agreement among diagnostic models was low; different interpretative models identified different students as having SLD. Furthermore, many college students receiving accommodations for SLD meet no objective criteria and display no normative deficits in academic achievement. In an early study, Ferri, Gregg, and Heggoy (1997) found that college students diagnosed with SLD showed no normative deficits in achievement; standard scores on measures of reading, spelling, and math fell within the normal range (M = 90–106). Similarly, Sparks and Lovett (2009b) found that the majority (54.8%) of students receiving accommodations for SLD met no objective criteria for the disorder and earned average to above average standard achievement scores (M = 90–119). In all, 60% had no history of SLD before beginning college. In their review of the empirical literature, Sparks and Lovett (2009a) found that college students diagnosed with SLD showed no normative deficits in reading (M = 97.5, n = 994), math (M = 97.7, n = 830), or written language (M = 90.7, n = 776) overall. These findings indicate that the diagnostic label SLD, when applied to college students, depends greatly on the diagnostic decision model used and may have little relationship to normative deficits in academic skills. Many college students classified with SLD appear to meet no criteria and show little objective evidence of impairment. However, these conclusions are based largely on samples of students enrolled in 4-year public universities. Is it possible that academic and cognitive functioning might be different for students enrolled at other types of postsecondary institutions?

SLD in Students Attending Community and Private Colleges Most high school students with SLD who attend college enroll in 2-year public institutions rather than 4-year colleges or universities. Data from the National Longitudinal Transition Study–2 (Gerber, 2009), which reflect a nationally representative sample of adolescents with disabilities, indicate that 33% of youth with SLD attend college within 2 years after high school graduation. Most (66%) students with SLD enroll in 2-year colleges rather than 4-year institutions. At least three

factors might explain this apparent self-selection to 2-year colleges (Gregg, 2007, 2009b). First, the prevalence of SLD is significantly higher among elementary and secondary school students from low socioeconomic backgrounds than among middle-class students (Child Trends Data Bank, 2010). Adolescents from less affluent families who attend college may be more likely to enroll in community colleges, which are typically less expensive than 4-year universities. Second, most secondary students with SLD have histories of academic problems and current academic deficits. Furthermore, their scores on standardized entrance exams tend to be below average (Gerber, 2009). These students may be more likely to enroll in 2-year colleges, which typically have lower admissions standards than 4-year universities. Third, adolescents with SLD may be more attracted to the smaller class sizes or support services for students with disabilities at 2-year colleges compared to 4-year institutions (Gregg, 2007, 2009b). It is possible that students with SLD who attend 2-year public colleges might display a different pattern of cognitive ability and academic achievement scores than their counterparts at 4-year colleges and universities. Specifically, they may be more likely to meet objective criteria for SLD overall, to meet criteria based on normative deficits in academic achievement, and to have histories of academic problems beginning in childhood or adolescence. A second group of students, largely ignored in the research literature, is those enrolled in highly selective, private, 4-year colleges. Although high school students with SLD disproportionately enroll in 2-year public colleges, the percentage of students receiving accommodations for SLD is greatest at 4-year, private liberal arts colleges. Vogel and colleagues (1998) analyzed data from a nationally representative survey of disability directors at postsecondary institutions. The overall prevalence of identified SLD among students was 0.7%; however, prevalence varied significantly by type of institution. Specifically, the prevalence of students receiving accommodations for SLD ranged from 0.65% at public universities to 1.27% at community colleges to 3.01% at highly selective, private colleges. Indeed, the prevalence of SLD approached 10% of the student body at some of the private liberal arts institutions. Students attending highly selective private colleges may earn higher average cognitive ability scores than students attending other institutions. These higher cognitive ability scores might increase the likelihood that students would show significant discrepancies between their overall ability and academic achievement (Gregg, 2007; Gregg, Coleman, & Knight, 2003). Indeed, Lovett and Sparks (2010) found that college students with above average cognitive ability scores (47.1%) were almost 3 times as likely to show significant ability– achievement discrepancies as students with cognitive ability scores within the average range (16.2%). Students receiving accommodations for SLD, who attend private colleges, may

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

493

Weis et al. be described as both “gifted and learning disabled” (G/LD), a term usually reserved for students with above average cognitive ability and relative (but not normative) deficits in academic achievement (Mather & Gerner, 2008). These students often earn achievement scores within the normal range and are not diagnosed until after beginning college, when coursework demands overtax their ability to compensate for their learning problems (Gregg, Coleman, Davis, Lindstrom, & Hartwig, 2006).

The Current Study Emerging evidence, therefore, indicates a relationship between students’ likelihood of being diagnosed with SLD and the decision model used. Students attending predominantly 4-year public universities are more likely to meet criteria using the discrepancy model than approaches involving normatively low achievement. However, many university students do not meet any objective criteria whatsoever. The purpose of this study is to examine whether institutional setting might moderate the relationship between students’ likelihood of meeting objective criteria for SLD and the diagnostic decision model employed. Like Sparks and Lovett (2009b), we recruited students previously diagnosed with SLD and receiving academic accommodations for this condition. In this manner, we excluded from our study students experiencing academic difficulties for other reasons, such as psychosocial stress or other disabilities. Like Proctor and Prevatt (2003), we administered the same psychoeducational battery to all students. In this manner, we could compare the same test scores across students and reduce measurement error as the result of instrumentation differences (Kazdin, 2003). We also administered a comprehensive achievement battery to students to assess those domains of academic achievement identified in IDEIA (2004) as areas of potential disability. Finally, our study extends previous research by including students from three different postsecondary settings: 4-year private liberal arts colleges, 4-year public universities, and 2-year public colleges. As in previous studies, we expected a relationship between students’ likelihood of meeting objective criteria for SLD and the diagnostic decision model used. Furthermore, we expected institutional setting to moderate this relationship. Specifically, we expected students at 4-year private colleges to be most likely to meet criteria for SLD based on significant ability– achievement discrepancies. In contrast, we expected students at 2-year public colleges to be most likely to meet criteria for SLD based on assessment models whose components include normatively low achievement, cognitive processing problems, and a history of academic difficulties. Finally, we expected students at 4-year public universities to earn test scores falling somewhere between those of students at the other institutions, making them less likely than 4-year private college students to meet criteria based on ability–achievement discrepancies

and less likely than 2-year public college students to meet criteria based on low achievement, processing problems, and history of impairment. Consequently, most students enrolled in 4-year public universities would fail to meet any objective criteria for SLD. Such findings would indicate qualitative differences in SLD across institutions.

Method Participants A total of 98 full-time undergraduate students (54% female) participated in this study. Ages ranged from 18 to 27 years. The mean age was 22.02 years (SD = 3.9 years). Ethnicities included White (78.6%), African American (13.3%), Asian American (5.1%), Latino (2.0%), and Native American (1.0%). All students had been previously identified as having SLD and were receiving academic accommodations through their institution’s disability services office. Specifically, 34% had been diagnosed with reading disorder (RD) only, 31% with RD and disorder of written expression (DWE), 20% with mathematics disorder (MD) only, 13% with RD and MD, and 2% with DWE only. Of the students, 30% had also been diagnosed with attention-deficit/hyperactivity disorder. Students were enrolled in (a) one of two highly selective, private liberal arts residential colleges (n = 33), (b) one of two public 4-year universities (n = 33), or (c) one of two public 2-year nonresidential (i.e., community) colleges (n = 32). All postsecondary institutions were located in Ohio. Average undergraduate enrollment was 1,943 for the private liberal arts colleges, 29,757 for the public 4-year universities, and 16,737 for the public 2-year (community) colleges. Average annual tuition and fees at the three types of institutions was $37,770, $8,793, and $3,337, respectively. Average acceptance rates at the three types of institutions were 34.5%, 70.0%, and 100.0%, respectively. Average composite ACT and SAT scores at the 4-year private colleges were 29 and 1,283, respectively. Average composite ACT and SAT scores at the 4-year public universities were 25 and 1,158, respectively. ACT and SAT scores were not required for 2-year public college students (National Center for Education Statistics, 2010).

Measures Participants completed the Woodcock–Johnson III Tests of Cognitive Abilities and Achievement–Extended (WJ III; Woodcock, McGrew, & Mather, 2001) as a measure of cognitive and academic functioning. The WJ III is an individually administered, norm-referenced psychoeducational battery frequently used to identify SLD. The extended Tests of Cognitive Abilities yield an overall measure of General Intellectual Ability (GIA) and composites that reflect seven components of cognitive processing included in the Cattell–Horn–Carroll (CHC)

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

494

Journal of Learning Disabilities 45(6)

model: Comprehension-Knowledge (Gc), Long-Term Retrieval (Glr), Visual-Spatial Thinking (Gv), Auditory Processing (Ga), Fluid Reasoning (Gf), Processing Speed (Gs), and Short-Term Memory (Gsm). The extended Tests of Achievement yield a Total Achievement Index and composites that reflect the five possible areas of SLD recognized by IDEIA (2004): Basic Reading Skill, Reading Comprehension, Mathematics Calculation, Mathematics Reasoning, and Written Expression. The WJ III also permits discrepancy analysis between participants’ extended GIA and each dimension of achievement. All scores have means of 100 and standard deviations of 15. Considerable data support the reliability and validity of the WJ III (Flanagan, Ortiz, Alfonso, & Mascolo, 2006; McGrew & Woodcock, 2001). Internal consistency estimates for composites fall above .90. One-year temporal stability estimates exceed .80 or .90 for all composites. Interrater reliability on subtests requiring examiner judgment ranges in the upper .90s. Factorial validity is supported by exploratory and confirmatory factor analysis indicating a relationship between composite scores and the CHC model. Evidence of convergent validity comes from correlations with other standardized measures of cognitive ability and academic achievement.

Procedure Participants were recruited by disability directors at each institution. Inclusionary criteria were (a) previous diagnosis of RD, MD, and/or DWE, (b) documentation on file with disability office, (c) receiving academic accommodations for disability, and (d) full-time student status. Documentation of SLD varied by postsecondary institution; however, all institutions required a recent (i.e., within the previous 3–5 years) evaluation conducted by a qualified professional (e.g., psychologist, educational specialist) consisting of a comprehensive psychoeducational test battery. Furthermore, the report of the evaluation had to specify the diagnosis of SLD and include norm-referenced test scores. These requirements are consistent with principles outlined by the National Joint Committee on Learning Disabilities (2007). Exclusionary criteria included any conditions that might interfere with standardized test administration (e.g., limited proficiency in English, visual impairment). The percentage of undergraduate students receiving accommodations for SLDs was 5.3% at the private 4-year colleges, 1.1% at the public 4-year universities, and 2.4% at the public 2-year colleges. All students meeting criteria for the study were solicited by email, in writing, or in person by disability directors. Participants were administered the WJ III by one of five research assistants trained by the first author, a licensed psychologist. Training consisted of formal instruction in standardized testing and scoring, practice administrations, and

test protocol review (Wendling & Mather, 2001). Test administration occurred at each participant’s college or university, in the disability office, testing center, or private study room in the library. Students were paid for their participation. Each participant’s test scores were examined to determine whether he or she met criteria for SLD using three diagnostic decision models: (a) the discrepancy model, (b) the DSM-IV model, and (c) the comprehensive cognitive model of SLD. Discrepancy model. Participants met criteria for SLD according to the discrepancy model if their score on any of the five achievement composites was at least 1.5 SD lower than their GIA extended standard score. This procedure reflects the traditional notion of ability–achievement discrepancy required by most school districts prior to IDEIA (Mather & Schrank, 2003) and is the most commonly used method for identifying SLD in college students (Madaus & Shaw, 2006). GIA extended scores reflect a weighting of cognitive subtest scores based on principal component analysis. GIA scores, therefore, are believed to provide a relatively unbiased estimate of general intelligence (g). GIA extended scores correlate from .71 to .76 with other cognitive ability composites that do not rely on such weightings, such as composites obtained from the Wechsler and Stanford–Binet intelligence tests (Schrank, Miller, Wendling, & Woodcock, 2010). DSM-IV model. Participants met criteria for SLD according to the DSM-IV model if they (a) showed a significant ability– achievement discrepancy as defined by the discrepancy model and (b) earned a standard score less than 85 (16th percentile) on any of the five achievement composites. This procedure reflects the DSM-IV (APA, 2000) diagnostic criteria for learning disorders, namely, the presence of both significant ability–achievement discrepancy and significant impairment in at least one achievement domain. Although DSM-IV does not explicitly require a normative deficit in achievement, this criterion has been used by previous authors to operationally define the DSM-IV “impairment” criterion (Lovett & Sparks, 2010; Proctor & Prevatt, 2003). Comprehensive cognitive model. Recently, Flanagan and colleagues (Flanagan, 2003; Flanagan, Ortiz, Alfonso, & Dynda, 2006; Flanagan, Ortiz, Alfonso, & Mascolo, 2006) have developed an operational definition of SLD that incorporates many components of contemporary cognitive assessment and is based on CHC theory. Unlike other assessment approaches, their model does not include ability–achievement discrepancy as a criterion. To be classified with SLD, participants must display a significant academic skill deficit in at least one achievement domain. According to the model, an absolute, not relative, deficit in achievement is a necessary component of SLD; without normative deficits in achievement, there is no objective evidence of disability and no need for academic accommodation. In this study, a standard score less than 85 on at

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

495

Weis et al. least one of the five achievement composites meets this criterion. Second, participants must show a significant cognitive deficit or processing problem that is theoretically or empirically related to their achievement deficit. Including the presence of an underlying cognitive processing problem in the criteria for SLD has a long history; it is based on the conceptualization of SLD as a psychological processing disorder and is a component of many contemporary approaches to SLD identification. According to Flanagan, Ortiz, Alfonso, and Mascolo (2006), problems with language development and lexical knowledge (Gc), phonological awareness and processing (Ga), and rapid automatic naming (Glr) are most often associated with reading and writing deficits. In contrast, problems with fluid reasoning (Gf), working memory (Gsm), and perceptual speed (Gs) are most often associated with mathematics deficits. However, the authors acknowledge that other cognitive processing problems can underlie these disorders. Indeed, subsequent research has identified other associations between cognitive processing and achievement deficits in adults (Dehn, 2008; Goldstein & Schwebach, 2009; Gregg, 2009b; Semrud-Clikeman & Fine, 2008; Swanson, 2009; Wolf, Schreiber, & Wasserstein, 2008). Consequently, in this study, a standard score less than 85 on any cognitive ability composite would indicate a cognitive processing problem and meet this criterion. To meet the third criterion, participants must show a pattern of academic and cognitive functioning that is consistent with SLD. Specifically, normative deficits in achievement and cognitive processing must be circumscribed, that is, they must exist within an otherwise normal cognitive profile. This criterion differentiates individuals with circumscribed cognitive deficits (i.e., SLD) from individuals with pervasive cognitive deficits (e.g., mental retardation). In this study, participants meet this criterion if the majority of their cognitive composite scores are within normal limits (≥ 85). Finally, participants must display “significant or substantial academic failure or other restrictions/limitations in daily life” (Flanagan, Ortiz, Alfonso, & Dynda, 2006, p. 823). In clinical settings, college students with SLD typically show real-world impairment in academic achievement, as evidenced by poor school performance or a history of academic difficulties (Gerber, 2009). In this study, students meet this criterion if they reported an onset of any academic difficulties prior to beginning college. Evidence of academic difficulties might include (a) repeating a grade, (b) failing a course because of reading-, writing-, or math-related difficulties, (c) being referred for psychological testing to rule out suspected learning disabilities, or (d) receiving special education or other academic instruction outside the regular classroom environment because of academic deficits. In summary, individuals meet criteria for the comprehensive cognitive model only if they display all four components of the

model: low achievement, underlying cognitive processing problems, circumscribed deficits, and a self-reported history of academic difficulties. Exclusionary criteria include sociocultural factors or physical and psychiatric conditions that might better account for the individual’s learning problems.

Results Demographic Differences Across Institutions We conducted a series of analyses, using SPSS, examining differences in demographic variables as a function of institution type. First, students attending 2-year public colleges were significantly older (M = 24.53, SD = 5.58) than students attending 4-year private colleges (M = 20.48, SD = 1.10), t(63) = 4.08, p < .001, d = 1.01, and 4-year public universities (M = 21.13, SD = 2.30), t(63) = 3.22, p = .002, d = 0.80. Second, there was a borderline-significant association between ethnicity and type of postsecondary school, χ2(2, N = 98) = 5.09, p = .078, Cramer’s V = .23. Follow-up contingency tests indicated that students attending 2-year public colleges (34.4%) were more likely to identify themselves as non-White than students attending 4-year private colleges (12.1%), χ2(1, N = 65) = 4.53, p = .033, Cramer’s V = .26. Third, 63% of participants reported an onset of learning problems before beginning college. However, there was a significant association between reported age of onset and institution type, χ2(2, N = 98) = 21.35, p < .001, Cramer’s V = .46. Students attending 2-year public colleges (93%) were more likely to report a first experience of learning problems in childhood or adolescence than students attending 4-year private colleges (39%), χ2(1, N = 65) = 21.45, p < .001, Cramer’s V = .57, and 4-year public universities (58%), χ2(1, N = 65) = 11.46, p = .001, Cramer’s V = .42.

Main Effects We conducted a Cochran test, examining differences in the number of students who met criteria for SLD across the three diagnostic models. Results showed significant overall differences, QC(2, N = 98) = 16.31, p < .001, Kendall’s W = .08. Follow-up McNemar’s tests, evaluated at p < .017 to control for familywise error, indicated that the number of students meeting criteria using the discrepancy model (36.7%) was significantly greater than the number using the DSM-IV model (14.3%). There was no difference in frequency between the comprehensive cognitive model (23.5%) and the other models. We conducted an overall contingency analysis examining the relationship between the number of students who met any criterion for SLD and institutional setting. Results indicated a significant overall association, χ2(2, N = 98) = 7.78,

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

496

Journal of Learning Disabilities 45(6)

Table 1. Number of Students Meeting and Not Meeting Criteria for Specific Learning Disability as a Function of Diagnostic Model and Institution Type Diagnostic Model Ability–Achievement Discrepancy

Institution Type 4-year private college (n = 33) 4-year public university (n = 33) 2-year public college (n = 32)

Meeting Criteria

Not Meeting Criteria

21  9  6

12 24 26

Comprehensive Cognitive Model

DSM-IV

%

Meeting Criteria

Not Meeting Criteria

63.6a 27.3b 18.8b

4 4 6

29 29 26

%

Meeting Criteria

Not Meeting Criteria

%

12.1b 12.1b 18.8b

 1  4 18

32 29 14

3.0b 12.1b 56.3a

Note: DSM-IV = Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (American Psychiatric Association, 2000). Percentages reflect the number of students meeting criteria. Different subscripts in the same row reflect different frequencies according to McNemar tests evaluated at p < .017. Different subscripts in the same column reflect different frequencies according to chi-square tests evaluated at p < .017.

p = .02, Cramer’s V = .28. Follow-up chi-square tests, evaluated at p < .017, indicated that students attending 4-year private colleges (63.6%), χ2(1, N = 66) = 6.07, p = .014, Cramer’s V = .30, and 2-year public colleges (62.5%), χ2(1, N = 65) = 5.94, p = .016, Cramer’s V = .29, were significantly more likely to meet any criterion for SLD than students attending 4-year public universities (33.3%).

Tests of Moderation We conducted an extended Mantel–Haenszel test using the Compare 2 module for Windows Programs for Epidemiologists (Abramson, 2004) to determine whether there was an overall relationship between diagnostic model and students’ likelihood of being classified as having SLD, controlling for institutional setting. Results (see Table 1) indicated a significant omnibus relationship between diagnostic approach and likelihood of classification, χMH2(1, N = 98) = 4.657, p = .031. More importantly, the heterogeneity chi-square test was significant, χ2(2, N = 98) = 39.32, p < .001, indicating that institutional setting moderated the relationship between diagnostic decision model and likelihood of diagnosis (see Note 1). To explore the role of institutional setting as an effect moderator, we conducted a series of Cochran tests to examine possible relationships between the number of students meeting criteria for SLD and diagnostic model within each of the three types of institutions. Each test was evaluated at p < .017. Results indicated a significant overall relationship between the model of assessment and frequency of diagnosis for students attending 4-year private colleges, QC(2, N = 33) = 34.90, p < .001, Kendall’s W = .53, and 2-year public colleges, QC(2, N = 32) = 18.00, p < .001, Kendall’s W = .28, but not 4-year public universities, QC(2, N = 33) = 5.56, p = .062, Kendall’s W = .08. Table 1 (rows) presents the results of follow-up McNemar tests, each evaluated at p < .017. Students attending 4-year private colleges

were more likely to meet criteria for SLD using the discrepancy model than any other interpretative model. In contrast, students attending 2-year public colleges were more likely to meet criteria for SLD using the comprehensive cognitive model than any other assessment model. Students attending 4-year public universities showed equal likelihood of meeting criteria using all three diagnostic models. Next, we conducted a series of chi-square analyses to examine possible relationships between the number of students meeting criteria for SLD and type of institution within each of the diagnostic models. Each test was evaluated at p < .017. Results indicated an overall association between type of institution and frequency of diagnosis using the discrepancy model, χ2(2, N = 98) = 16.00, p < .001, Cramer’s V = .40, and the comprehensive cognitive model, χ2(2, N = 98) = 29.19, p < .001, Cramer’s V = .55, but not the DSM-IV model, χ2(2, N = 98) = 0.773, p = .679. Table 1 (columns) presents the results of follow-up chi-square tests, evaluated at p < .017, within each diagnostic model. Students attending 4-year private colleges were more likely to meet criteria using the discrepancy model than students attending other institutions. In contrast, students attending 2-year public colleges were more likely to meet criteria using the comprehensive cognitive model than students attending other institutions.

Students Meeting Objective Criteria Across Institutions For our final analyses, we identified only those students who met objective criteria for SLD according to any of the three diagnostic models. Then, we conducted a multivariate analysis of variance (MANOVA) using SPSS, examining differences in cognitive ability scores as a function of institution type. Results (Table 2) indicated a significant multivariate effect, Wilks’s Λ = .196, F(16, 84) = 6.60, p < .001, η2 = .55. Follow-up tests showed significant differences for

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

497

Weis et al. Table 2. Standardized Test Scores for Students Meeting Criteria for Specific Learning Disability as a Function of Institution Type Institution Type 4-Year Private

4-Year Public

2-Year Public

Follow-Up ANOVA

Composite

Score

SD

Score

SD

Score

SD

F(2, 49)

p

η2

General Intellectual Ability Comprehension-Knowledge Long-Term Retrieval Visual-Spatial Thinking Auditory Processing Fluid Reasoning Processing Speed Short-Term Memory Total Achievement Basic Reading Skill Reading Comprehension Mathematics Calculation Mathematics Reasoning Written Expression n

111.76 114.81 106.10 112.10 114.09 109.24 100.86 112.19 100.76 99.86 105.10 96.14 98.48 108.43

6.07a 9.48a 4.93a 7.54a 9.22a 6.01a 4.51a 10.84a 7.40a 10.10a 9.13a 10.92a 10.55a 8.14a

104.73 102.91 106.73 107.00 105.91 105.36 98.91 109.82 95.27 91.91 101.09 100.36 100.55 96.91

9.82a 12.20b 7.28a 8.84a 12.01a,b 9.24a 9.33a 12.06a 10.86a 10.50a,b 10.54a 12.24a 12.63a 13.64b

84.90 86.30 87.90 92.00 98.25 90.35 89.55 87.45 84.20 84.85 91.35 82.80 86.40 87.55

5.64b 7.46c 10.59b 6.14b 12.69b 8.81b 8.98b 10.14b 5.93b 9.86b 10.25b 10.54b 8.60b 8.24c

58.31 34.64 32.48 40.24 10.16 31.33 12.38 30.14 23.87 11.34 10.26 11.54 9.58 24.53

< .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 .001 < .001

.70 .58 .57 .62 .29 .56 .33 .55 .49 .32 .29 .32 .28 .50

21

11

20

Note: Different subscripts in the same row are significantly different at p < .017.

all dependent variables. Students attending 4-year private and public schools earned significantly higher GIA scores than students attending 2-year public colleges. Ability composites reflected these differences in GIA. In general, scores for 4-year private college students were toward the high end of the average range (M = 100–114), scores for 4-year public university students were squarely within the average range (M = 98–109), and scores for 2-year public college students were toward the lower end of the average range (M = 86–98). We conducted another MANOVA examining differences in students’ academic achievement scores as a function of institution type. Results showed a significant multivariate effect, Wilks’s Λ = .334, F(12, 88) = 5.35, p < .001, η2 = .42. Follow-up tests indicated that Total Achievement scores were significantly higher for students enrolled in 4-year private and public schools than for students enrolled in 2-year public colleges. Composites reflected these differences in Total Achievement. Composite scores for students enrolled in 4-year public and private schools fell within the average range (M = 91–108) whereas achievement scores for students enrolled in 2-year colleges fell within the low average to average ranges (M = 82–91).

Discussion The purpose of our study was to examine the possibility that institutional setting might moderate the relationship between college students’ likelihood of meeting objective criteria for SLD and the decision model used to identify the condition.

Such evidence would suggest qualitative differences in students classified with SLD across postsecondary institutions. As in previous studies, we found a significant overall relationship between diagnostic model and the number of students meeting objective criteria for SLD. The greatest percentage of students was classified using the discrepancy model (36.7%), followed by the comprehensive cognitive (23.5%) and DSM-IV models (14.3%). The discrepancy model may have been most likely to identify college students as having SLD because, by definition, it requires students to show only a relative (not normative) deficit in academic skills and favors students with average or above average cognitive ability (Gregg, 2009a). Operational definitions that include normative deficits in academic ability identify fewer college students with SLD because students who evidence such normative deficits may be less likely to attend college (Gerber, 2009). Our results are also consistent with previous studies (Sparks & Lovett, 2009a) suggesting that many college students classified with SLD do not meet any objective criteria for the disorder. Specifically, 46.9% of students in our study showed neither a significant ability–achievement discrepancy nor normative deficits in achievement. These findings indicate that many clinicians may base their diagnosis of SLD on factors unrelated to these objective criteria, such as inspection of intraindividual discrepancy scores, use of profile analysis, or reliance on clinical judgment (Gregg, 2003; Hagin, 2003; McCoach, Kehle, Bray, & Siegle, 2004). Use of intraindividual discrepancy scores and profile analysis has been criticized for their lack of reliability and inability to differentiate individuals with and without learning problems

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

498

Journal of Learning Disabilities 45(6)

(Fletcher et al., 2007). Although clinical judgment certainly plays a role in diagnosis, its overreliance can compromise the reliability of SLD identification. To the extent clinicians rely on unreliable diagnostic approaches or minimize objective criteria in reaching diagnostic decisions, the SLD label loses its ability to aid in professional communication, to guide interventions, and to facilitate evidence-based research. Our study extends our knowledge of SLD in college students by demonstrating differences in the number of students meeting objective criteria for SLD across postsecondary settings. Students enrolled in 4-year private (63.6%) and 2-year public colleges (62.5%) were more likely to meet criteria for SLD than students enrolled at 4-year public universities (33.3%). More importantly, our tests of moderation indicate that the effect of assessment method on the number of students classified with SLD depends on institutional setting. Students receiving accommodations at 2-year public colleges were most likely to meet objective criteria for SLD using the comprehensive cognitive model. Approximately one half (53.6%) showed normative deficits in academic achievement and underlying cognitive processing problems, but comparatively few displayed significant ability–achievement discrepancies (18.8%). The magnitude of these academic deficits was generally one standard deviation below the mean, at approximately the 16th percentile. Achievement scores on the WJ III within this range correspond to academic skills similar to those evidenced by students enrolled in the 6th through 8th grades (McGrew & Woodcock, 2001). Consequently, these students likely experience significant difficulty performing college-level academic tasks without accommodations. Furthermore, most of the 2-year college students displayed cognitive processing problems in addition to normative deficits in achievement. Finally, almost all (93%) of the 2-year public college students receiving accommodations for SLD reported experiencing academic problems in childhood or adolescence. Although self-report data are susceptible to cognitive biases and distortions in recall, the available data suggest that most of these students had a history of academic difficulties before college. In many ways, the 2-year public college students who met objective criteria for SLD in our sample reflect the traditional scientific and legal conceptualizations of SLD. First, SLD has always been understood as a disorder characterized by learning impairment (Flanagan, Ortiz, Alfonso, & Dynda, 2006; Gregg, 2009b). Although experts disagree whether the learning deficit should be relative or normative, this impairment should interfere with daily life activities and, in the case of students, academic achievement. Consequently, evidence of learning deficits that interfere with functioning is a component of many contemporary models of SLD (Brueggemann, Kamphaus, & Dombrowski, 2008; Flanagan, Ortiz, Alfonso, & Dynda, 2006). The community college students in our study who met criteria for SLD clearly experienced difficulty

performing academic tasks. Furthermore, their impairment was relative to the general population of young adults, not to other high-achieving college students. Second, both professional (Learning Disabilities Roundtable, 2002) and legal (Taymans, 2009) standards attribute the academic deficits shown by individuals with SLD to underlying cognitive processing problems (Dehn, 2008; Mather, 2009). Adults with SLD use different and more inefficient cognitive processing strategies than adults without SLD (Gregg et al., 2008). Furthermore, the existence of cognitive processing deficits appears to differentiate adults with and without SLD (Swanson, 2009). The existence of one or more cognitive processing deficits, therefore, is necessary to identify SLD as a disorder of “basic psychological processes” (IDEIA, 2004) and differentiate it from other causes of low academic achievement such as impoverished educational experiences or socioeconomic disadvantage (Hale, Kaufman, Naglieri, & Kavale, 2006; Kavale & Spaulding, 2008; Semrud-Clikeman & Fine, 2008). Third, the academic and cognitive deficits shown by the community college students in our study were circumscribed. From their first conceptualization, learning disabilities have been viewed as “specific” to a limited number of cognitive domains; they do not reflect pervasive deficits in cognitive functioning. For this reason, SLD is often viewed as a disorder characterized by “unexpected underachievement,” the existence of circumscribed cognitive deficits in the presence of otherwise intact processing integrities (Kavale, Kaufman, Naglieri, & Hale, 2005). Although the average cognitive ability scores of community college students meeting objective criteria were rather low, this average reflects the wide range of circumscribed deficits shown by individual students who met criteria for SLD according to the comprehensive cognitive model. Finally, SLD is usually conceptualized as a developmental disorder, which typically emerges during childhood or early adolescence and can compromise academic achievement over time (Dombrowski, Kamphaus, & Reynolds, 2004; SemrudClikeman & Fine, 2008). Although it is possible that SLD might not be identified until late adolescence or early adulthood, especially among individuals with well-developed compensatory strategies, most individuals with SLD report an onset of academic problems earlier in life. For example, adults with SLD typically report the onset of their reading or mathematics problems sometime during elementary school (Krueger & Grafman, 2008; Sherman, 2008). The community college students in our study reported this early onset. In contrast, the students in our study who were enrolled at private liberal arts colleges were most likely to meet objective criteria for SLD based on a significant ability–achievement discrepancy. Approximately 64% of these students showed a significant discrepancy, but comparatively few displayed normative deficits in academic skills necessary to meet criteria outlined in the DSM-IV (12.1%) or comprehensive cognitive (3.0%) models. Indeed, the average achievement scores of

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

499

Weis et al. these students were within the normal range (Total Achievement = 101, 53rd percentile), indicating no evidence of normative impairment. In contrast, average cognitive ability scores were generally within the high end of the normal range (GIA = 112, 79th percentile), indicating well-developed cognitive skills. The combination of their relatively high cognitive ability scores and relatively low academic achievement scores would lead many clinicians to consider these students as G/LD (Lovett & Sparks, 2010). Although G/LD is not an official diagnostic category in any nosological system, several researchers have described individuals with characteristics of both giftedness and learning disability (Brody & Mills, 1997; McCoach, Kehle, Bray, & Siegle, 2001; Nielsen, 2002). Although it is relatively rare, there are instances of adults who show both above average cognitive ability (FSIQ > 120) and normative deficits in at least one achievement domain. However, the G/LD label is most frequently assigned to individuals who earn above average cognitive ability scores and show relative deficits in academic skills. Some authors (Gregg, Coleman, Lindstrom, & Lee, 2007; Mather, 2009; Mather & Gregg, 2006) argue that these individuals need not display normative deficits in achievement; by relying on their strong cognitive abilities, hard work, and perseverance, these individuals can compensate for their learning problems and develop academic skills within the normal range. Like most of the liberal arts students in our study, many individuals classified as G/LD are not identified until after beginning college, when “the academic demands begin to exceed (their) ability to compensate” (Brody & Mills, 2004, p. 77). Critics argue that SLD classification should not be given to adults who demonstrate academic skills in the normal range (Gordon, Lewandowski, & Keiser, 1999; Lovett, Gordon, & Lewandowski, 2009). First, critics argue, federal legislation requires the presence of normative, not relative, deficits in academic functioning. According to the Americans with Disabilities Act Amendments Act (2008), disability refers to “a physical or mental impairment that substantially limits one or more of the major life activities of an individual.” Guidelines provided by the Equal Employment Opportunity Commission (2002) further state that the limitation must “amount to a significant restriction when compared to the abilities of the average person in the general population.” Relative deficits, which do not impair the major life activities of most people, do not meet this standard. Critics also argue that relative deficits in achievement are not sufficient to justify the provision of academic accommodations (Gordon et al., 1999; Lovett & Lewandowski, 2006). Given the scarcity and cost of academic support services, their provision to students who show relative deficits might mean the restriction of services to students with normative deficits, that is, students who need these services to successfully complete college-level work.

It is noteworthy that the DSM-IV model yielded the most consistent prevalence rates across postsecondary institutions. Although the discrepancy model favored students with higher cognitive functioning, and the comprehensive cognitive model favored students with lower cognitive functioning, the DSM-IV interpretative model was less influenced by students’ cognitive ability. However, many professionals would likely consider the DSM-IV model too conservative; approximately 80% of students in our sample receiving accommodations for SLD would not meet criteria according to this model. Furthermore, the DSM-IV model is based on the presence of significant ability–achievement discrepancies, a criterion that may have limited validity as a means of identifying SLD in adults (Swanson, 2009). Although professionals do not agree on an operational definition of SLD in college students at this time, adoption of such a definition would likely result in considerable changes to the prevalence of SLD at some institutions. An operational definition based on significant ability–achievement discrepancy would likely decrease the number of students at 2-year public colleges with SLD. Furthermore, exclusive reliance on the discrepancy criteria to identify SLD would deprive most individuals who show normative deficits in academic achievement and cognitive processing the accommodations they need to complete their degrees. In contrast, an operational definition based on normative deficits in academic achievement and underlying cognitive processing problems would virtually eliminate SLD classification among students at selective colleges and universities. At the very least, professionals may wish to specify whether the deficits shown by students classified with SLD are normative or relative. Specification might help clinicians and disability directors at postsecondary institutions identify accommodations and plan interventions to improve students’ academic functioning (Dehn, 2008, Gregg, 2009a). For example, interventions designed to improve the cognitive processing of college students with SLD might be particularly useful for individuals who meet criteria according to the comprehensive cognitive model but may be ineffective for students who show significant ability–achievement discrepancies but no cognitive processing problems. In contrast, academic skill development workshops may be especially useful for students with SLD based on ability–achievement discrepancies, to help these individuals acquire the skills necessary to achieve their academic potentials. Specification would also allow researchers to identify more homogeneous groups of college students with SLD for their studies. For example, studies involving children (Swanson & Hoskyn, 1999) and adolescents (Swanson, 2001) indicate that the magnitude of the discrepancy between cognitive ability and reading achievement moderates the efficacy of treatment; youth with larger discrepancies show lower effect sizes in treatment whereas youth with circumscribed deficits in both

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

500

Journal of Learning Disabilities 45(6)

ability and reading achievement respond most favorably to interventions. Similar research, involving homogeneous groups of adults with SLD, is necessary to identify possible Ability × Treatment interactions in this population. Of course, professionals can continue to rely on the myriad operational definitions of the disorder and corresponding assessment strategies, which would likely ensure that the number of students who receive accommodations for SLD, yet meet no objective criteria for the disorder, remains high. This practice would also perpetuate the problem of low reliability in SLD classification among college students and open clinicians and postsecondary schools up to criticism that the SLD label offers unfair advantages to students with the sophistication and financial resources to seek out and acquire SLD classification. Our study is limited by its reliance on relatively few students from each of the three institutions who participated in testing. Although the percentage of participants who met objective criteria for SLD was similar to those obtained in previous studies involving larger samples (Proctor & Prevatt, 2003; Sparks & Lovett, 2009b), it is possible that our findings would be different if we were able to administer comprehensive psychoeducational batteries to more students. Our study is also limited by the fact that we examined the components of students’ cognitive abilities assessed by only the WJ III. Although the domains we assessed reflect many components of the CHC model, we did not assess some dimensions of cognitive processing (e.g., rapid naming) believed to be relevant to all learning problems (e.g., basic reading skills; Daniels, 2008). It is possible more students would have shown cognitive processing problems if we could have assessed cognitive skills not included in the WJ III. Ideally, our methods would have resembled the assessment approach used by clinicians who select tests based on each student’s presenting academic difficulties. By relying on a common psychoeducational battery, we sacrificed flexibility in assessment for standardization. Finally, we did not assess disabilities related to oral language. Although none of our participants reported a history of oral language problems, it is possible that some might have met objective criteria for SLD in the areas of listening comprehension or oral expression. A third limitation concerns external validity. Although our study extends the research literature on college students with SLD by examining students at three different postsecondary institutions, we do not know whether the institutions that we included in our study represent other 4-year private, 4-year public, and 2-year public colleges, respectively. Furthermore, we cannot be certain that the students who participated in our study from these institutions are representative of other students with documented or undocumented SLD at their respective schools. A final limitation lies in the lack of a nondisabled comparison group at each institution. In future studies, students with SLD not only could be compared across institutions but also could be compared to their classmates without SLD to provide additional control.

As it stands, professionals have erected a veritable “Tower of Babel” in the field of SLD research involving college students (Flanagan, Ortiz, Alfonso, & Dynda, 2006). The term SLD does not convey the same meaning to all (or perhaps even most) researchers and clinicians. Students assigned the same diagnostic label across institutions appear to have different cognitive and academic abilities, histories, and degrees of impairment. If the science of learning disability research and the practice of SLD identification and mitigation are to advance, a common understanding of this disorder must be developed. Our research provides a first glimpse of the repercussions such a common definition might have on the prevalence of SLD at various postsecondary institutions. Declaration of Conflicting Interests The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Financial Disclosure/Funding This research was supported by grants from the Laurie Bukovac Hodgson and David Hodgson Endowed Fund for Science Research and the Denison University Research Foundation.

Note 1. The heterogeneity chi-square is the sum of the Mantel– Haenszel chi-squares in each stratum (institution) minus the overall Mantel–Haenszel chi-square (Abramson, 2010).

References Abramson, J. H. (2004). WINPEPI computer programs for epidemiologists. Epidemiologic Perspectives and Innovations, 6, 1–10. Abramson, J. H. (2010). WINPEPI Compare 2 manual. Retrieved from http://www.brixtonhealth.com/ American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author. Americans with Disabilities Act Amendments Act, Pub. L. No. 110-325 (2008). Brody, L. E., & Mills, C. J. (1997). Gifted children with learning disabilities: A review of the issues. Journal of Learning Disabilities, 30, 282–296. Brody, L. E., & Mills, C. J. (2004). Linking assessment and diagnosis to interventions for gifted students with learning disabilities. In T. M. Newman & R. J. Sternberg (Eds.), Students with both gifts and learning disabilities (pp. 73–93). New York, NY: Kluwer. Brueggemann, A. E., Kamphaus, R. W., & Dombrowski, S. C. (2008). An impairment model of learning disability diagnosis. Professional Psychology: Research and Practice, 39, 424–430. Child Trends Data Bank. (2010). Child Trends of the National Health Interview Surveys. Retrieved from http://www.child trendsdatabank.org/ Daniels, A. (2008). Reading fluency in adults. In L. E. Wolf, H. E. Schreiber, & J. Wasserstein (Eds.), Adult learning disorders (pp. 121–125). New York, NY: Psychology Press.

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

501

Weis et al. Dehn, M. J. (2008). Cognitive processing deficits. In R. J. Morris & N. Mather (Eds.), Evidence-based interventions for students with learning and behavioral challenges (pp. 258–287). New York, NY: Routledge. Dombrowski, S. C., Kamphaus, R. W., & Reynolds, C. R. (2004). After the demise of the discrepancy: Proposed learning disabilities diagnostic criteria. Professional Psychology: Research and Practice, 35, 364–372. Equal Employment Opportunity Commission. (2002). Enforcement guidance on reasonable accommodation and undue hardship under the Americans with Disabilities Act. Washington, DC: Author. Ferri, B. A., Gregg, N., & Heggoy, S. J. (1997). Profiles of college students demonstrating learning disabilities with and without giftedness. Journal of Learning Disabilities, 30, 552–559. Flanagan, D. P. (2003). Use of the Woodcock–Johnson III within the context of a modern operational definition of learning disability. In F. A. Schrank & D. P. Flanagan (Eds.), WJ III clinical use and interpretation (pp. 200–227). San Diego, CA: Academic Press. Flanagan, D. P., Ortiz, S. O., Alfonso, V. C., & Dynda, A. M. (2006). Integration of response to intervention and norm-referenced tests in learning disability identification: Learning from the Tower of Babel. Psychology in the Schools, 43, 807–825. Flanagan, D. P., Ortiz, S. O., Alfonso, V. C., & Mascolo, J. T. (2006). The achievement test desk reference (2nd ed.). New York, NY: John Wiley. Fletcher, J. M., Lyon, G. R., Fuchs, L. S., & Barnes, M. A. (2007). Learning disabilities: From identification to intervention. New York, NY: Guilford. Gerber, P. J. (2009). Transition and adults with learning disabilities. In J. M. Taymans (Ed.), Learning to achieve: A review of the research literature on serving adults with learning disabilities (pp. 211–252). Washington, DC: National Institute for Literacy. Goldstein, S., & Schwebach, A. (2009). Neuropsychological basis of learning disabilities. In C. R. Reynolds & E. Fletcher-Janzen (Eds.), Handbook of clinical child neuropsychology (pp. 187–201). New York, NY: Springer. Gordon, M., Lewandowski, L., & Keiser, S. (1999). The LD label for relatively well-functioning students: A critical analysis. Journal of Learning Disabilities, 32, 485–490. Gregg, N. (2003). Diagnostic issues surrounding students with learning disabilities. In S. A. Vogel, G. Vogel, V. Sharoni, & O. Dahan (Eds.), Learning disabilities in higher education and beyond (pp. 93–105). Baltimore, MD: York. Gregg, N. (2007). Underserved and unprepared: Postsecondary learning disabilities. Learning Disabilities Research and Practice, 22, 219–228. Gregg, N. (2009a). Accommodations: Evidence-based accommodation research specific to the adolescent and adult population with learning disabilities. In J. M. Taymans (Ed.), Learning to achieve: A review of the research literature on serving adults with learning disabilities (pp. 119–180). Washington, DC: National Institute for Literacy.

Gregg, N. (2009b). Adolescents and adults with learning disabilities and ADHD: Assessment and accommodation. New York, NY: Guilford. Gregg, N., Bandalos, D. L., Coleman, C., Davis, J. M., Robinson, K., & Blake, J. (2008). The validity of a battery of phonemic and orthographic awareness tasks for adults with and without dyslexia and attention deficit/hyperactivity disorder. Remedial and Special Education, 29, 175–190. Gregg, N., Coleman, C., Davis, M., Lindstrom, W., & Hartwig, J. (2006). Critical issues for the diagnosis of learning disabilities in the adult population. Psychology in the Schools, 43, 889–899. Gregg, N., Coleman, C., & Knight, D. (2003). Use of the Woodcock– Johnson III in the diagnosis of learning disabilities. In F. A. Schrank & D. P. Flanagan (Eds.), WJ III clinical use and interpretation (pp. 125–174). San Diego, CA: Academic Press. Gregg, N., Coleman, C., Lindstrom, J., & Lee, C. (2007). Who are most, average, or high-functioning adults? Learning Disabilities Research and Practice, 22, 264–274. Hagin, R. A. (2003). Diagnosing adults with learning disabilities: Essential components. In S. A. Vogel, G. Vogel, V. Sharoni, & O. Dahan (Eds.), Learning disabilities in higher education and beyond (pp. 83–91). Baltimore, MD: York. Hale, J. B., Kaufman, A., Naglieri, J. A., & Kavale, K. A. (2006). Implementation of IDEA: Integration response to intervention and cognitive assessment methods. Psychology in the Schools, 43, 753–770. Hoskyn, M., & Swanson, H. L. (2000). Cognitive processing of low achievers and children with reading disabilities: A selective meta-analytic review of the published literature. School Psychology Review, 29, 102–119. Individuals with Disabilities Education Improvement Act of 2004, 20 U.S.C. § 1400 et seq. (2004). Kavale, K. A., Kaufman, A., Naglieri, J., & Hale, J. (2005). Changing procedures for identifying learning disabilities: The danger of poorly supported ideas. School Psychologist, 59, 15–25. Kavale, K. A., & Spaulding, L. S. (2008). Is response to intervention good policy for specific learning disability? Learning Disabilities Research and Practice, 23, 169–179. Kazdin, A. E. (2003). Research design in clinical psychology. Boston, MA: Allyn & Bacon. Krueger, F., & Grafman, J. (2008). Disorders of mathematics: Implications for adult functioning. In L. E. Wolf, H. E. Schreiber, & J. Wasserstein (Eds.), Adult learning disorders (pp. 191–218). New York, NY: Psychology Press. Learning Disabilities Roundtable. (2002). Specific learning disabilities: Finding common ground. Washington, DC: U.S. Department of Education. Lovett, B. J., Gordon, M., & Lewandowski, L. J. (2009). Measuring impairment in a legal context: Practical considerations in the evaluation of psychiatric and learning disabilities. In S. Goldstein & J. Naglieri (Eds.), Assessing impairment: From theory to practice (pp. 93–103). New York, NY: Springer. Lovett, B. J., & Lewandowski, L. J. (2006). Gifted students with learning disabilities: Who are they? Journal of Learning Disabilities, 39, 515–527.

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014

502

Journal of Learning Disabilities 45(6)

Lovett, B. J., & Sparks, R. S. (2010). Exploring the diagnosis of “gifted/LD:” Characterizing postsecondary students with learning disability diagnoses at different IQ levels. Journal of Psychoeducational Assessment, 25, 17–28. Madaus, J. W., & Shaw, S. F. (2006). The impact of the IDEA 2004 on transition to college for students with learning disabilities. Learning Disabilities Research and Practice, 21, 273–281. Mather, N. (2009). The intelligent testing of children with specific learning disabilities. In J. C. Kaufman (Ed.), Intelligent testing: Integrating psychological theory and clinical practice (pp. 30–52). New York, NY: Cambridge University Press. Mather, N., & Gerner, M. E. (2008). Postsecondary students with high abilities and reading disabilities: Case analysis and commentary. Learning Disabilities, 15, 121–129. Mather, N., & Gregg, N. (2006). Specific learning disabilities: Clarifying, not eliminating, a construct. Professional Psychology: Research and Practice, 37, 99–106. Mather, N., & Schrank, F. A. (2003). Using the Woodcock–Johnson III discrepancy procedures for diagnosing learning disabilities. In F. A. Schrank & D. P. Flanagan (Eds.), WJ III clinical use and interpretation (pp. 175–198). San Diego, CA: Academic Press. McCoach, D. B., Kehle, T., Bray, M. A., & Siegle, D. (2001). Best practices in the identification of gifted students with learning disabilities. Psychology in the Schools, 38, 403–411. McCoach, D. B., Kehle, T. J., Bray, M. A., & Siegle, D. (2004). The identification of gifted students with learning disabilities: Challenges, controversies, and promising practices. In T. M. Newman & R. J. Sternberg (Eds.), Students with both gifts and learning disabilities (pp. 31–46). New York, NY: Kluwer. McGrew, K. S., & Woodcock, R. W. (2001). Technical manual, Woodcock–Johnson III. Itasca, IL: Riverside. National Center for Education Statistics. (2010). Integrated Postsecondary Education Data System. Retrieved from http://nces .ed.gov/ipeds/ National Joint Committee on Learning Disabilities. (2007). The documentation disconnect for students with learning disabilities: Improving access to postsecondary disability services. Retrieved from http://www.ahead.org/uploads/docs/resources/ njld_paper.pdf Nielsen, M. E. (2002). Gifted students with learning disabilities: Recommendations for identification and programming. Exceptionality, 10, 93–111. Proctor, B., & Prevatt, F. (2003). Agreement among four models used for diagnosing learning disabilities. Journal of Learning Disabilities, 36, 459–466. Schrank, F. A., Miller, D. C., Wendling, B. J., & Woodcock, R. W. (2010). Essentials of WJ III cognitive abilities assessment. New York, NY: John Wiley.

Semrud-Clikeman, M., & Fine, J. G. (2008). Brief comprehensive neuropsychological assessment of children and adults with learning difficulties and disabilities. In R. C. D’Amato & L. C. Hartlage (Eds.), Essentials of neuropsychological assessment (pp. 245–291). New York, NY: Springer. Sherman, J. C. (2008). Normal and learning disabled (LD) development of academic skills. In L. E. Wolf, H. E. Schreiber, & J. Wasserstein (Eds.), Adult learning disorders (pp. 3–24). New York, NY: Psychology Press. Sparks, R. L., & Lovett, B. J. (2009a). College students with learning disability diagnoses: Who are they and how do they perform? Journal of Learning Disabilities, 42, 494–510. Sparks, R. L., & Lovett, B. J. (2009b). Objective criteria for classification of postsecondary students as learning disabled: Effects on prevalence rates and group characteristics. Journal of Learning Disabilities, 42, 230–239. Stuebing, K. K., Fletcher, J. M., LeDoux, J. M., Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2002). Validity of IQdiscrepancy classifications of reading disabilities: A metaanalysis. American Educational Research Journal, 39, 469–518. Swanson, H. J. (2001). Research on interventions for adolescents with learning disabilities: A meta-analysis of outcomes related to higher-order processing. Elementary School Journal, 101, 331–348. Swanson, H. J. (2009). Assessment of adults with learning disabilities. In J. M. Taymans (Ed.), Learning to achieve: A review of the research literature on serving adults with learning disabilities (pp. 15–70). Washington, DC: National Institute for Literacy. Swanson, H. J., & Hoskyn, M. (1999). Definition × treatment interactions for students with learning disabilities. School Psychology Review, 28, 644–658. Taymans, J. M. (2009). Introduction. In J. M. Taymans (Ed.), Learning to achieve: A review of the research literature on serving adults with learning disabilities (pp. 1–13). Washington, DC: National Institute for Literacy. Vogel, S. A., Leonard, F., Scales, W., Hayeslip, P., Hermansen, J., & Donnelis, L. (1998). The National Learning Disabilities Postsecondary Data Bank: An overview. Journal of Learning Disabilities, 31, 234–247. Wendling, B. J., & Mather, N. (2001). Woodcock–Johnson III Tests of Cognitive Abilities and Tests of Achievement: Examiner training booklet. Itasca, IL: Riverside. Wolf, L. E., Schreiber, H. E., & Wasserstein, J. (2008). Adult learning disorders. New York, NY: Psychology Press. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock– Johnson III. Itasca, IL: Riverside.

Downloaded from ldx.sagepub.com at OhioLink on October 4, 2014