Analysis of the Reliability of the Cognitive Ability Test as a Predictor of Intellectual Giftedness

Abstract

The problem: This study examined the reliability of the Cognitive Ability Test by Thorndike and Wagen as a tool for identification of gifted students, Procedure. Statistical analysis using a chi-square and a t-test were used to examine the reliability/ repeatability of this test. The first grade scores were compared with the fourth grade scores of the same students using a correlated t-test. Within group comparisons were based on a chi-square analysis. Findinqs. This study showed that there was a high relationship between Test 1 and Test 2, therefore, Test 1 is highly predictable of Test 2. However, this was not true in the upper range scores (those in the 130 and above range). Conclusions. This study found that the upper range scores in the Cognitive Ability Tests do not repeat in a subsequent test of the same students. Recommendations: Use of the Cognitive Ability Test as a predictor of gifted and talented students should be reevaluated since this study found that the very range where the scores are used for classification in the gifted and talented programs is the range where the test lacks repeatability, and thus reliability.