Date of Award

Spring 5-2008

Degree Type

Dissertation

Degree Name

Doctor of Education (EdD)

Department

Education & Educational Psychology

First Advisor

Karen A. Burke, CSJ, EdD

Second Advisor

Patricia Cyganovich, EdD

Third Advisor

Marcia A. B. Delcourt, PhD

Fourth Advisor

Bruce M. Shore, PhD

Abstract

This study examined the impact of the conceptually designed assessment, test debate and test analysis, on students’ critical thinking and ability to analyze literature. The test analysis and test debate process involved three steps: (a) teachers created and administered a multiple-choice exam that asked conceptual questions; (b) students participated in a Socratic test debate in which they were required to support their answers using specific textual references; and (c) students wrote a metacognitive reflection of the evolution of their thought process including an initial interpretation of the question, the points gleaned during the debate, and their final interpretation of the course concept or theme addressed in the question.

Using a sample of convenience (n = 157), this study assessed the use of test debate and test analysis in six separate classes among heterogeneously and homogeneously grouped students in grades 6 through 11. The quasi-experimental research design of this study used The California Measure of Mental Motivation (CM3), Advanced Placement English Language and Composition raw scores, and New York State English Language Arts assessments to consider how well the process enhances students’ critical thinking skills and students’ ability to read and analyze literature.

A two-group and three-group multivariate analysis of variance (MANCOVA) with the Literature Pretest covariate was conducted on the six dependent variables: Literary Analysis, Mental Focus, Learning Orientation, Creative Problem Solving, Cognitive Integrity, and Scholarly Rigor. The data set was analyzed using an independent variable with two levels and three levels.

The two-group MANCOVA data analysis revealed statistically significantly group difference on three of the six dependent variables (Creative Problem Solving, Scholarly Rigor, and Literary Analysis). The three-group MANCOVA produced similar results with regard to significance level, but examination of mean scores was not consistent with the findings of the two-group MANCOVA. A statistically significant effect of the independent variable three groups (trained and treatment, trained no treatment, and no treatment) existed for Mental Focus, Creative Problem Solving, Scholarly Rigor, and Literary Analysis. It can be concluded that the statistically significant multivariate effect was driven in part by the impact of grouping on these dependent variables.

Share

COinS