Lester Mann

The Cognitive Levels Test: Its Relationship with Reading and Mathematics Achievement.

Any instrument designed to measure broad cognitive ability is expected to correlate substantially with valid measures of academic achievement. This report describes an investigation designed to evaluate the relationship between the Cognitive Levels Test, a newly available instrument, and two tests of academic achievement that have been widely used for nearly two decades: the Woodcock Reading Mastery Tests and the KeyMath Diagnostic Arithmetic Test. The study included children in grades K-2 who were enrolled in a remedial summer school program. The results showed the Cognitive Levels Test (Cognitive Index) to be rather highly correlated with the KeyMath Diagnostic Arithmetic Test Total score (r=.72) and moderately correlated with the Woodcock Reading Mastery Tests Total Reading score (r=.55). A repeated measures analysis of variance comparing standard scores for the Cognitive Levels Test with those of the Woodcock Reading Mastery Tests indicated few mean differences between the two sets of scores. Implications concerning the validity of the Cognitive Levels Test were discussed.

Cognition and Academic Achievement: The Relationship of the Cognitive Levels Test, the KeyMath Revised, and the Woodcock Reading Mastery Tests-Revised.

This study evaluated the ability of the Cognitive Levels Test to estimate current achievement in mathematics (as measured by the KeyMath Revised) and reading (as measured by the Woodcock Reading Mastery Tests-Revised) among a group of normal children attending a private school. The validity coefficients resulting from the analysis indicated that the Cognitive Levels Test scores were moderately to highly correlated with the WRMT-R and KMR, respectively. Repeated-measures analyses of variance yielded no significant main effect for the CLT/KMR scores, but did identify a main effect for the CLT/WRMT-R scores. Follow-up multiple comparisons revealed one significant difference between the various mean scores: The CLT Verbal Reasoning mean was significantly greater than the WRMT-R Passage Comprehension mean. Given the substantial correlations and generally nonsignificant differences between mean scores, the results were judged as strong evidence of the validity of the CLT for the purpose of estimating math and reading achievement.

Cognition and Academic Achievement: The Relationship of the Cognitive Levels Test to the Keymath and Woodcock Reading Mastery Tests

This study evaluated the ability of the Cognitive Levels Test to estimate current achievement in mathematics (as measured by the KeyMath Diagnostic Arithmetic Test) and reading (via the Woodcock Reading Mastery Tests) among a group of 43 handicapped and at-risk children. The validity coefficients resulting from the analysis indicated that the Cognitive Levels Test provides valid estimates of mathematics and reading achievement. A repeated measures analysis of variance yielded just one score that was significantly different from other scores in the set. That is, the Letter Identification score of the Woodcock Reading Mastery Test was found to be significantly greater than all other reading and cognitive scores. Although the Cognitive Levels Test scores met criteria for convergent-discriminant validity with regard to the KeyMath Diagnostic Arithmetic Test scores, the results were less promising regarding the Woodcock Reading Mastery Test.

Models for Measuring Induced Affective Change: Some Neglected Methodological Considerations

The measurement of affect and other personality characteristics has usually been carried out in a predispositional, correlational framework. However, the measurement of varying levels of affect induced by different treatments during an experiment has received less attention because of the methodological problems involved. Typical weaknesses of research concern test-retest effects, test sensitization, test-wiseness, lack of control groups, inductive construct validity of measures, emotional boundedness, purging effects, and short treatment duration. A manageable design paradigm for gaining sound results on experimentally induced affect is presented.

Test Anxiety and Defensiveness Experimentally Induced by Four Conditions of Testing Arousal.

To study the relationships among frequency of testing, arithmetic learning and retention, predisposition al test anxiety, defensiveness against admission of test anxiety, and induced test anxiety, eighty sixth-grade Ss were randomly assigned to four arousal conditions: tests every day, tests every other day, tests once a week, and daily practice. Teachers were randomly rotated daily. The study lasted 5 weeks. An achievement post test was given at the end of the study and again 2 weeks later. Induced test anxiety was measured at the end of each week. On both achievement posttests, the only significant difference was in favor of the daily test group over the weekly test group. Induced test anxiety was found to operate similar to predispositional test anxiety.

The Relative Predictive and Construct Validities of the Otis-Lennon Mental Ability Test, The Lorge-Thorndike Intelligence Test, and the Metropolitan Readiness Test in Grades Two and Four: A Series of Multivariate Analyses.

1 This study was supported cooperatively by RISE, an Elementary and Secondary Education Act of 1965 Title III project (OEG-1-67-3010-2696), the School District of Cheltenham Township, and a predoctoral USOE fellowship in educational research granted to J. R. McGowan at Lehigh University, Bethlehem, Pa. However, the opinions expressed herein do not necessarily reflect the policies or positions of the cooperating agencies. The senior author