An associative evaluation of selected interdisciplinary courses
Three studies were conducted to evaluate learning in selected interdisciplinary courses using the Word Association Test (WAT) as the measurement instrument. The Ss in Study I were students enrolled in the course "Technology and Society." Ss were given one of two forms of a 45-item WAT during the first and last weeks of the course. Completed forms were sorted into three groups on the basis of their academic major: Engineering, Liberal Arts, and Other (Ss with majors not belonging to either of the first two Colleges).
Distributions of responses for each of these different groups for each administration were compared using the overlap coefficient. Both the Engineering--Liberal Arts and the Liberal Arts--Other group comparisons showed significant increases in response similarity. The Engineering--Other group comparison revealed no significant differences in response similarity. In addition, the average overlap coefficient was found to be significantly higher for the postcourse administration. These results indicate that for three of four comparisons students produced more homogeneous response distributions after course exposure. The range of reliability estimates (using Cronbach's coefficient alpha-α) were from .903 to .963.
The Ss in Study II were ten matched students enrolled in the course, "Aspects of the Urban Environment," who completed both administrations of a 30-item WAT. The results showed a significant increase in mean association scores indicating that Ss were able to give a higher number of acceptable responses after course exposure. The reliability estimates were found to be α = .781 for the precourse and α = .914 for the postcourse. Finally, responses to one stimulus term for each administration were presented to illustrate qualitative changes in responding as a function of course exposure.
The Ss in Study III were students (Precourse N = 49; Postcourse N = 31) enrolled in the course, "Human Needs and Services," who completed both administrations of a 30-item WAT. The results indicated a significant decrease in mean association score indicating fewer acceptable responses were given after course exposure. The reliability estimates were found to be α = .9l8 for the precourse and α = .932 for the postcourse. Finally, responses to one stimulus term were presented to illustrate qualitative changes in responding as a function of course exposure.
The Word Association Test is a reliable measurement tool for assessing changes in response distributions resulting from exposure to interdisciplinary courses. It is sensitive to both academic major differences among Ss and to changes due to course impact in terms of overlap of response distributions as well as to mean association scores.
Thesis75b.S243.pdf
6.02 MB
Adobe PDF
132cd05560588008ece66e50b708fd3d