ERIC Number: ED023123
Record Type: RIE
Publication Date: 1968-Feb
Reference Count: 0
New Procedures for Scoring Psychological Measurements (Development of Moderated Scoring Keys for Psychological Inventories). Final Report.
Prediger, Dale J.
The three major project objectives were as follows: (1) development of procedures for determining the optimum number of subgroups (and hence, moderated scoring keys) required for maximizing the predictive effectiveness of an inventory; (2) development of a single scale for reporting the scores obtained from a set of moderated keys; and, (3) determination of the accuracy of moderated scoring key predictions of college attendance as compared with predictions obtained from conventional keying techniques. Basic data consisted of biographical inventory responses and academic aptitude test scores for approximately 20,500 high school boys. Scoring keys were formed for each of 12 ability level subgroups and various combinations of these subgroups. The keying procedures developed as the primary objective of the project work well when applied to actual data. Although a statistically significant difference in favor of the moderated keys was obtained, academic ability was not found to be an effective moderator variable. However, a hit rate of 77% was achieved by biographical data as a predictor of college attendance versus nonattendance. This rate and the equivalent point biserial correlation coefficient of .60 were substantially higher than the corresponding figures for academic aptitude used alone. (AUTHOR)
Publication Type: N/A
Education Level: N/A
Sponsor: Office of Education (DHEW), Washington, DC. Bureau of Research.
Authoring Institution: Toledo Univ., OH.