Abstract

The purpose of this study was to examine the preliminary psychometrics to revisions to the decoding subtest of the Predictive Early Assessment of Reading and Language (PEARL) dynamic assessment and to further examine the predictive validity of the language subtest of the PEARL. Specific aims were to (a) determine if kindergarten students made gains on the PEARL decoding subtest from pretest to a delayed standard posttest and to a delayed novel posttest, indicating that floor effects were mitigated, (b) to examine if there was variance in the delayed standard posttest and novel posttest scores, (c) to determine if there was adequate fidelity of administration and adequate inter-rater reliability of a delayed standard posttest and a delayed novel decoding posttest, and (d) to examine the sensitivity of the language subtest of the PEARL dynamic assessment. The PEARL Dynamic Assessment was administered to 34 kindergarten students from one elementary school who were identified as at-risk from the DIBELS (Dynamic Indicators of Basic Early Literacy Skills) composite score (Good et al., 2004). Of those 34 students, the PEARL identified seven students at risk for decoding and 11 students at risk for language. The PEARL dynamic assessment contains two brief subtests, a decoding subtest and a language subtest. Both subtests use a pretest, teach, posttest format. Results indicated that the delayed standard posttest correct sounds were significantly different from the pretest correct sounds. The delayed novel posttest correct sounds were significantly different from the pretest correct sounds. For the delayed standard posttest correct words there was a significant difference when compared to the pretest correct words. The delayed posttest novel words were not significantly different from the pretest words. Fidelity of administration was calculated from ten separate administrators, with 100% fidelity of administration for nine of those administrations. The point-to-point interrater reliability of the delayed posttest sounds, and novel sounds was 94% and the point-to-point interrater reliability of the total words read correctly was 94%. Results indicated that the PEARL correctly identified 9/11 students as having language disorder, yielding a sensitivity index of 82%. The findings suggest that the modification to the decoding subtest of the PEARL has preliminary evidence of validity and reliability, and that the language subtest of the PEARL may be a valid tool to identify kindergarten students at risk for language disorder.

Degree

College and Department

David O. McKay School of Education; Communication Disorders

Rights

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