Publication Date

Spring 2020

Advisor(s) - Committee Chair

Sarah Ochs (Director), Ryan Farmer, and Carl Myers

Degree Program

Department of Psychology

Degree Type

Specialist in Education


Many guidebooks and publication manuals for intelligence tests recommend that practitioners calculate subtest and index level discrepancy scores along with many other types of scores. The reliability of the discrepancy scores, however, are not provided in the test manuals. The purpose of the study was to determine the reliability of the discrepancy scores in the Woodcock Johnson Test of Cognitive Abilities (WJ IV COG, 2014). Using data from the WJ IV technical manual, the reliability of the discrepancy scores were examined at both the subtest and index level. The results were compared to the reliability criteria recommended by the assessment community for both hypothesis generation and clinical decision making. At the subtest-level, the reliability coefficients ranged from .61 to .93. The composite-level reliability coefficients ranged from .59 to .93. Limitations and future directions are discussed.


Other Education | Other Social and Behavioral Sciences | School Psychology