Dissertation/Thesis Abstract

Homogeneous grouping in the context of high-stakes testing: Does it improve reading achievement?
by Salcedo-Gonzalez, Trena, Ed.D., California State University, Fullerton, 2012, 89; 3528261
Abstract (Summary)

As accountability reform intensifies, urban school districts strive to meet No Child Left Behind mandates to avoid severe penalties. This study investigated the resurgence of homogeneous grouping methods as a means to increase reading achievement and meet English Language Arts Adequate Yearly Progress requirements. Specifically, this study investigated the relationship between achievement grouping methods and reading achievement as well as the differential effect that it may have on Latino students in particular.

A multi-level analysis was designed to examine the relationship between achievement grouping methods and reading achievement for a cohort of fourth-grade students. A hierarchical linear model was utilized to analyze the data and control for student and teacher characteristics. Students' ethnicity, prior reading achievement, gender, and socioeconomic status, as well as teachers' ethnicity, achievement grouping methods, and years of teaching experience were controlled. Descriptive statistics were also included to show the association between students' characteristics (gender, ethnicity, and socioeconomic status) and reading achievement growth.

The Hierarchical Linear Model found that there is no difference in reading achievement between students enrolled in homogeneous or heterogeneous classes. This study also found that achievement grouping methods did not have a differential effect on Latino students in comparison to non-Latino students. However, it did find that Latino students continued to experience less reading achievement growth in comparison to their peers. The chi square test of independence also showed ethnicity and socioeconomic status to be associated with lower reading achievement.

Based on these results, this study challenged the notion that leveling students for instruction leads to positive reading achievement outcomes. It does not enable lower performing students to excel nor has it been shown to equalize education. Even with the best intentions in mind, homogeneous grouping does little to improve student achievement. This study's findings also urge educational leaders to reexamine their school reform efforts to ensure that all students benefit academically, especially those who have been historically marginalized.

Indexing (document details)
Advisor: Ordonez-Jasis, Rosario
School: California State University, Fullerton
School Location: United States -- California
Source: DAI-A 73/12(E), Dissertation Abstracts International
Subjects: Instructional Design, Elementary education, Literacy, Reading instruction
Keywords: Achievement grouping, Equity, Hierarchical linear model, Latino, Latino students, No Child Left Behind, Reading achievement
Publication Number: 3528261
ISBN: 978-1-267-58656-8
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy