NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED079401
Record Type: RIE
Publication Date: 1973-May
Pages: 12
Abstractor: N/A
Reference Count: 0
ISBN: N/A
ISSN: N/A
Differential Validity: A Problem with Tests or Criteria?
Hollmann, Thomas D.
The evidence used in condemning a test as racially biased is usually a validity coefficient for one racial group that is significantly different from that of another racial group. However, both variables in the calculation of a validity coefficient should be examined to determine where the bias lies. A study was conducted to investigate the construct validity of a set of predictors and a set of criterion scales separately for blacks and whites. Data were collected during a project to validate a test battery and a followup study of the effectiveness of the resultant selection procedures. Ss totaled 70 blacks and 104 whites. Tests used were the Adaptability Test, the Spelling scale, word meaning scale, checking scale and copying scale of the Purdue Clerical Adaptability Test and the ten scales of the Guilford-Zimmerman Temperament Survey. The criterion scales consisted of 5 specially constructed, behaviorally anchored rating scales measuring accuracy, information, attitude, initiative, and knowledge of procedures. Eleven cases of differential validity occurred. The pattern of test-criterion relationships was obviously different for the two groups. The factor patterns for the 15 predictors were for the most part similar for blacks and whites. The analysis of the factor structure of the predictors suggested that they measure the same or similar constructs within the two groups. Results indicate that the criterion scales, more than the predictors, contributed to the differential validity and single-group validity. (KM)
Publication Type: N/A
Education Level: N/A
Audience: N/A
Language: N/A
Sponsor: N/A
Authoring Institution: N/A
Note: Paper presented at a meeting of the Midwestern Psychological Association, May, 1973