Date of Award

Fall 2011

Degree Type

Dissertation

Degree Name

Doctor of Education (EdD)

Department

Teacher Leadership for Learning

First Advisor

Dr. Angela Blaver

Second Advisor

Dr. T. C. Chan

Third Advisor

Dr. Benjamin Lester

Abstract

To comply with federal accountability requirements, English learners (ELs) in Georgia must demonstrate progress in academic English language proficiency (ELP) and academic content-area achievement as measured annually by two state-mandated Title I and Title III assessments, the Accessing Communication and Comprehension in English State-to-State for English Language Learners® (ACCESS for ELLs®) and the Criterion-Referenced Competency Test (CRCT). Using a critical ELP assessment validation framework, the intent of this study was to investigate the relationship of students’ performances on Title assessments, analyze student level variability in the predictive power of the ELP assessment for academic achievement, and examine claims made when using assessment data for measurement, classification, and prediction. Partial correlations, multiple regressions, t-tests, and ANOVA analyses were performed on district level data for 1059 elementary 1st – 5th grade native Spanish speaking students who took both assessments in 2010. The results confirm that when student characteristics are held constant, a significant positive, though moderate, relationship exists between Els’ performances on the ACCESS for ELLs® and CRCT. The findings show that the time spent in English language development programs along with disability status and grade levels explain more variance in CRCT scores than students’ ELP scores and gender. Discussion is provided regarding the implications of these findings to support educators’ claims that EL assessment data is valid for high-stakes accountability decisions.