Abstract

Specific learning disabilities (SLD) are among the most diagnosed developmental disorders in children, with specific learning disability in reading (SLDR) as the most common. SLDR typically occurs due to deficits in phonological processing that can negatively impact decoding and/or reading comprehension skills. Reading deficits are often diagnosed using formal assessments such as the Comprehensive Test of Phonological Processing (CTOPP-2), the Gray Oral Reading Test (GORT-5), the Wechsler Intelligence Scale for Children (WISC-V), and the Woodcock Johnson Achievement Battery (WJ-IV-A). I did not find any research in which the theoretical factor models proposed in the technical manuals of these assessments were confirmed in a clinical sample. Benefits of using a community sample compared to a standardized sample include improved ecological and consequential validity (e.g., fewer diagnostic exclusion criteria, regional sampling differences). Additionally, the theoretical factor structure for the WJ-IV-A and WISC-V have been criticized in the literature through secondary analyses on the normative sample. Improving consequential validity can increase the confidence of clinicians in making diagnostic decisions and treatment recommendations using data from these measures. I also used generalizability theory to evaluate whether the WJ-IV-A, CTOPP-2, and GORT-5 should be used interchangeably as is often done in clinical practice. Data from psychological assessments from the BYU Comprehensive Clinic, a community clinic, during 2014-2021 were extracted into a database containing 587 individuals and used in analyses. I performed CFA using the proposed theoretical factor structure for each of the CTOPP-2 and GORT-5 and adapted theoretical factor structures for the WJ-IV-A and WISC using the community sample. Factor loadings for each of the models were generally moderate to strong, and multiple goodness of fit indices indicated adequate to good fit, which suggests that these four measures are typically measuring their proposed constructs in a diagnostically diverse, regional sample. The relative coefficients of generalizability analyses using the CTOPP-2, GORT-5, and WJ-IV-A were high when using within-subjects data. These findings indicate that the reading subtests across these three measures are reliably measuring the same reading skills. Implications of this dissertation include additional psychometric evidence for the theoretical factor structures of the CTOPP-2, GORT-5, WJ-IV-A, and WISC-V. Furthermore, results provided preliminary evidence for using the CTOPP-2, GORT-5, and WJ-IV-A interchangeably and, thus, shortening SLDR evaluation batteries.

Degree

PhD

College and Department

Family, Home, and Social Sciences; Psychology

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2023-08-11

Document Type

Dissertation

Handle

http://hdl.lib.byu.edu/1877/etd13377

Keywords

Specific Learning Disorder in Reading, confirmatory factor analysis, generalizability

Language

english

Download

Share

COinS