An Alternative Oral Proficiency and Expressive Vocabulary Assessment of Kindergarten English Language Learners

Abstract

The data used in this study were secondary, kindergarten data from a longitudinal, five-year, federal experimental research project: English and Literacy Acquisition (ELLA) (R305P030032). The overall goal of ELLA was to examine the impacts of two different programs (Bilingual and Structured English Immersion) on the performance of Spanish-speaking English language learners (ELLs) in grades K to 3. My first research question was to determine to what extent a curriculumbased measure could be developed and validated to measure oral proficiency and vocabulary knowledge among ELLs who are participating in a controlled oral language development intervention. In addressing validity the scores of the S4 were compared with the scores of the Woodcock Language Proficiency Battery - Revised (WLPB-R) and the IOWA Test of Basic Skills (ITBS) language and vocabulary subtests. The correlations were .283 to .445 and they were statistically significant (p<.01). The S4 underwent several iterations. With each iteration intrarater reliability improved (Kappa .817 to 1.00 and Cramer's V .330 to 1.00). Interrater reliability also improved (Kappa .431 to 1.00 and Cramer's V .616 to 1.00). The second research question was to determine to what extent teachers could use the Semantic and Syntactic Scoring System (S4) for the STELLA vocabulary fluency measure with minimal training to accurately assess students' vocabulary knowledge and oral proficiency. The teachers' Kappas ranged from .786 to 1.00 and Cramer's V from .822 to 1.00. On average they were able to score a given student measure in under 22 minutes. The third research question was to determine to what extent the Semantic and Syntactic Scoring System (S4) differentiates the level of knowledge regarding expressive vocabulary and oral proficiency of kindergarten students under two different program placements: enhanced Traditional Bilingual Education and the enhanced Structured English Immersion Program in comparison to the WLPB-R (language and vocabulary subtests). The S4 was able to distinguish between the control and experimental groups (unlike the other subtests); but was not able to distinguish program type (bilingual and structured English immersion).