Premium
COMPARISON OF PREDICTIVE VALIDITY AND DIAGNOSTIC ACCURACY OF SCREENING MEASURES OF READING SKILLS
Author(s) -
Klingbeil David A.,
McComas Jennifer J.,
Burns Matthew K.,
Helman Lori
Publication year - 2015
Publication title -
psychology in the schools
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.738
H-Index - 75
eISSN - 1520-6807
pISSN - 0033-3085
DOI - 10.1002/pits.21839
Subject(s) - predictive validity , psychology , fluency , reading (process) , curriculum based measurement , receiver operating characteristic , test (biology) , test validity , variance (accounting) , criterion validity , incremental validity , concurrent validity , psychometrics , developmental psychology , clinical psychology , statistics , construct validity , mathematics education , curriculum , pedagogy , mathematics , law , internal consistency , business , biology , paleontology , accounting , political science , curriculum mapping , curriculum development
Assessment data must be valid for the purpose for which educators use them. Establishing evidence of validity is an ongoing process that must be shared by test developers and test users. This study examined the predictive validity and the diagnostic accuracy of universal screening measures in reading. Scores on three different universal screening tools were compared for nearly 500 second‐ and third‐grade students attending four public schools in a large urban district. Hierarchical regression and receiver operating characteristic curves were used to examine the criterion‐related validity and diagnostic accuracy of students’ oral reading fluency (ORF), Fountas and Pinnell Benchmark Assessment System (BAS) scores, and fall scores from the Measures of Academic Progress for reading (MAP). Results indicated that a combination of all three measures accounted for 65% of the variance in spring MAP scores, whereas a reduced model of ORF and MAP scores predicted 60%. ORF and BAS scores did not meet standards for diagnostic accuracy. Combining the measures improved diagnostic accuracy, depending on how criterion scores were calculated. Implications for practice and future research are discussed.