Content area
Abstract
The purpose of the study was to predict critical thinking abilities by way of teacher assessment of intelligent behavior. Forty-nine, grade 5, Catholic, elementary school students, from a midwestern urban area, served as subjects. The Cornell Test of Critical Thinking--Level X, Third Edition (Ennis, Millman, & Tomko, 1985) was administered during the sixth week of the 1992-1993 school year to obtain a measure of critical thinking ability. Teacher assessment of each student, using a 14-item Intelligent Behavior Rating Scale based upon Costa's (1991) 14 Characteristics of Intelligent Behavior, was obtained following administration of the Cornell test.
To determine the relationship between the assessment of intelligent behavior and critical thinking ability in this pre-experimental research design, multiple linear regression statistical analysis was used.
Based upon the findings of this study the null hypothesis, the Intelligent Behavior Rating Scale does not predict student performance on the Cornell Test of Critical Thinking--Level X, Third Edition, was rejected at the.05 level of significance. Neither teacher, gender, nor method of scoring the Cornell was significant in the predictive value of the Intelligent Behavior Rating Scale upon student performance on the Cornell test.
Analysis of the data appeared to support the following conclusions: (1) The Intelligent Behavior Rating Scale is a potential reliable alternative to multiple-choice testing when attempting to authentically assess critical thinking skills or intelligent behavior of intermediate level, elementary school students. (2) Teacher bias is not a significant factor when rating students on the Intelligent Behavior Rating Scale. (3) Gender of students does not affect how teachers rate individual students using the Intelligent Behavior Rating Scale or performance on the Cornell test.
