Comparative investigation of the psychometric properties of three tests of logical thinking

Abstract

Using a sample of 908 eleventh grade science stream male and female students from similar socioeconomic area schools, variance based psychometric properties of three paper-and-pencil tests of logical thinking (Longeot test, Lawson's test TOFR, and Tobin and Capie's test TOLT) are investigated. A sub-sample of 212 students took the three tests in randomly allocated different sequential orders of presentation, while 696 students took only two tests. AIfa coefficients for each test separately and for the three tests combined together, concurrent validity coefficients, measures of item difficulty, item discrimination, item-criterion correlation, and 30-day stability coefficients are calculated. Considering the relative homogeneity of the sample, the reliability coefficients of the tests are judged satisfactory, but concurrent validity coefficients are quite low which implies incongruency in decisions made on the basis of the three tests. Need for estimating various psychometric parameters of alternative tests of logical thinking over different grade populations is emphasized. ence teaching on a large scale for two reasons. One is that the time and preparation needed for the interviews are extensive and prohibitive, and the other is that the technique is not objective because of the difficulty in standardizing the

guided-inquiry based laboratory instruction: investigation of critical thinking skills, problem solving skills, and implementing student roles in chemistry

“…The TOLT is a 10-item written test that was designed to assess cognitive development with respect to specific Piagetian tasks. The TOLT contains two items for measuring each of the five formal reasoning modes namely (a) proportional reasoning, (b) controlling variables, (c) probabilistic reasoning, (d) correlational reasoning, and (e) combinatorial reasoning (Tobin, and Cape 1981;Triflone, 1987); Ahlawat, and Billeh, 1987).…”

“…Ten items on the TOLT measure formal thinking of students. Two items measure proportional reasoning, the next two measure students' ability to control variables, two questions are about probabilistic reasoning, the next two questions are on correlational reasoning, and the last two questions on the TOLT measure the combinatorial thinking of students (Trifone, 1987;Ahlawat, and Billeh, 1987). Each item on the TOLT (from 1 through 8) is in multiple-choice format, with a sub question that is also in multiple choice format.…”