Date of Award:

5-2010

Document Type:

Dissertation

Degree Name:

Doctor of Philosophy (PhD)

Department:

Special Education and Rehabilitation

Advisor/Chair:

Dr. Ronald B. Gillam

Abstract

This study investigated the validity of measures that were hypothesized to account for significant variance in English reading ability. During kindergarten, 63 bilingual Hispanic children completed letter identification, English and Spanish phonological awareness, rapid automatized naming, and sentence repetition static assessment tasks. They also completed a dynamic assessment nonsense-word decoding task that yielded pretest to posttest gain score, response to decoding strategy, and temporally related working memory information. One week prior to kindergarten, information was gathered regarding socioeconomic status, preschool attendance, English and Spanish language dominance, and language ability. At the end of first grade, the same children completed word identification, decoding, and reading fluency tasks designed to represent the narrow view of reading. Reliability, content relevancy, construct validity, and predictive evidence of validity were examined. The letter identification task, the English-only and Spanish-only tasks, and a composite of the participants' best English and Spanish scores accounted for significant variance in first-grade word-level reading. However, the Spanish and BLS static measures did not account for significant, unique variance over and above English-only static measures, and the English-only static measures did not account for significant, unique variance over and above the letter identification static measure. The dynamic assessment measure pertaining to the response to reading strategy instruction accounted for equivalent variance in first-grade word-level reading when compared to a combination of letter identification and BLS static measures. The dynamic assessment measure yielded the highest classification accuracy, with sensitivity and specificity at or above 80% for all three formative criterion reading measures, including 100% sensitivity for two out of the three first-grade measures. The dynamic assessment of reading strategy surfaced as a parsimonious, valid means of predicting first-grade word-level reading ability for Hispanic, bilingual children. When compared to multiple English, Spanish, and BLS static measures, the dynamic measure accounted for equivalent variance in the majority of first-grade reading measures and had superior classification accuracy.

Share

COinS