Date Awarded


Document Type


Degree Name

Doctor of Education (Ed.D.)




Recent advances in microcomputers allowed for the development of a Computer Assisted Assessment (CAA) device which provided a higher degree of feedback and interaction for subjects than had previously been possible. This enhanced human/machine interactive capability was found to increase the comparability between a CAA device and a traditional test instrument. A counter-balanced, pre-post test comparison between the manual and CAA Raven Standard Progressive Matrices (RSPM) indicated no difference between test mediums, with the exception of the CAA pre-test requiring significantly longer for students to complete than the manual pre-test version. A reliability study of the CAA RSPM indicated this version of the RSPM produced a test-retest r = 88 with a two-week interval (N = 21). In addition to examining the comparability between test mediums, five CAA derived measures were developed to determine their efficacy as psychometric measures. Three measures of response time and two measures of 'motoric efficiency' or purposefulness were developed. Each measure provided a means of fractionating subjects' responses and recording these responses in milliseconds and measuring motoric output with a resolution of 1/80th of an inch. Eighty-five CAA administrations of the RSPM were obtained using 4th, 5th and 6th grade public school students. Discriminate analysis revealed 3 of the CAA derived psychometric measures produced significant results at the p < .10 level. A final post hoc comparison between previously administered WISC-R scores and the CAA RSPM was made with 20 students. Multiple regression and correlation analyses indicated significant relationships with the CAA RSPM and the WISC-R verbal and full scale IQ scales. In addition several CAA derived measures demonstrated significant relationships with the WISC-R verbal, performance and full scale IQ scores.



© The Author