The major purpose of this study was to develope a biographical questionnaire suitable for Chinese college students and to examine the validation of biodata and personality as useful predictors of broadly defined college student performance.Empirical criterion keying was the scaling method of biodata. First interviewed and investigated some college students of grade two using critical incident technical to realize how kinds of past life history and behaviors could influence college performance. Then based on the biographical information blank developecd by Zhang Yi Chen, developed the initial biodata items. Finally the final questionnaire was formed by item analysis.Revised Behaviorally anchored rating scale for multiple dimensions of college performance developed by Frederick L Oswald, contained eleven dimensions to assess college performance. The revised version had satisfactory psychometrics characters, including good reliability and validity. The scale could distinguish effectively the different level of college performance of students'.344 students from five universities in Nanjing were selected as subjects. The main results are as follows: biodata had good reliability and validity, the internal consistency reliability was 0.776, criterion related validation was 0.559, the cross validation was 0.385. Through regression analysis, twelve biodata items and four personality dimensions had great prediction to college performance. And biodata accounted for significant incremental validation over personality. Through variance analysis, there were significant difference on fifteen biodata items and ten personality dimensions among the high performance group, middle performance group and low performance group. The ten personality dimensions were: warmth, emotional stability, dominance, liveliness, rule consciousness, social boldness, vigilance, apprehension, perfectionism, and tension. |