A systematic review of the psychometric properties of standardized reading assessments

A systematic review of the psychometric properties of standardized reading assessments

First Author: Kelley Nelson-Strouts -- University of Kansas Medical Center
Additional authors/chairs: 
Mindy Sittner Bridges
Keywords: Assessment, Reading Assessment, Word reading, Reading comprehension
Abstract / Summary: 

Purpose: Research and clinical diagnostics widely use commercially-available, norm-referenced reading assessments. Being able to accurately and efficiently locate information regarding psychometric properties is critical to make informed choices of assessments to use in clinical practice and/or research. In this study, we describe a systematic process for evaluating standardized assessments widely used with children with or at risk for reading disability (see Plante & Vance, 1994).

Method: After an initial search, 12 assessments were reviewed. Criteria for inclusion included the following: standardized, norm-referenced assessments; for children ages 4;0-17;11; and published within the last twenty years in the United States. What constituted a reading assessment was based on the Simple View of Reading (Gough & Turner, 1986) and thus, tests of reading comprehension and tests of decoding were included. Each went through an extensive coding system on the identified psychometrics. Additionally, educators will be trained to complete the coding process and will be interviewed after as to the ease of locating this information.

Results: Preliminary results indicate administration procedures and evidence of item analysis were most often reported. Predictive validity, sensitivity, and specificity were the least reported criteria. Evidence of test-retest reliability was frequently reported, although the strength of this reliability varied. Normative samples were based on recent U.S. Census data but most failed to report key information such as participants’ language backgrounds.

Conclusions: There are notable deficits in the psychometrics of many tests of reading disorders, with evidence of validity and classification accuracy particularly lacking. Educators and researchers alike must be aware of these deficits in order to make informed decisions prior to a test’s use.