Inter- and Intra-Test Correlates of the TOEFL

Abstract

The present study provides validity information for the TOEFL by examining some of its interand intra-test correlates. Inter-test correlates included: 1) grades in an intensive English program, 2) accumulated scores from objective quizzes administered after each of 15 lessons in a course designed to teach listening comprehension and note taking skills, and 3) scores on the Comprehensive English Language Test (CELT) with subtests representing structure, listening comprehension and vocabulary abilities. In addition, intercorrelations were done within ability groups (high, medium and low, as measured by students' levels in their English program) to determine if correlation patterns vary according to the academic level of the student. Intra-test correlates consisted of investigations of correlations among subtests within the TOEFL. Factor analyses were used to aid in interpretation of the various correlation structures. The purpose of this study is to initiate a validation study of the content of the TOEFL. One factor was identified through factor analysis of TOEFL subtest scores, with the reading comprehension subtest having the highest factor loading. The interrelated nature of the TOEFL subtests is supported by positive correlations (greater than .50) within TOEFL subtests and between TOEFL and CELT subtests. A high correlation between the listening comprehension subtest of the TOEFL and another listening comprehension measure, the listening tracts, as well as a considerable correlation between listening tract scores and TOEFL totals, suggest that listening comprehension may be a separate skill that is significantly interrelated with total score success. Grades in an intensive English program are not predictors of TOEFL success, although relative academic level is.