Journal of Applied Psychology

Brown, Clarence W.; Ghiselli, Edwin E.

doi: 10.1037/h0062754pmid: N/A

"Seven tests [arithmetic, reaction, dotting, tapping, two distance, mechanical] and an interest inventory were administered to 54 taxicab drivers and validated against their sales. With one possible exception [arithmetic], no single test gave adequate prediction. A simple weighted combination of the tests yielded a validity of .39. When the weighted battery was applied to another group of 29 drivers it was found to have a validity of .29 in the prediction of ratings of job proficiency."

Jenkins, James J.

doi: 10.1037/h0057721pmid: N/A

A sample of 92 weather forecasters completed the Minnesota Clerical Test, Minnesota Paper Form Board, Ohio State Psychological Exam, and the Strong Vocational Interest Blank. The forecasters were a highly select group, scoring in the 80th-90th percentile on the first three tests. In interests, they resembled a technical, skilled-trades interest group. A double cross-validation study indicated only one consistent predictor of forecasting accuracy, the Names section of the Minnesota Clerical Test, which correlated .31 with the criterion.

Hay, Edward N.

doi: 10.1037/h0059512pmid: N/A

The objections to creating a "large sample" by combining the results of several intact small samples are briefly mentioned.

Cattell, Raymond B.; Anderson, Jean C.

doi: 10.1037/h0056224pmid: N/A

"A previously completed factor analysis of 120 very diverse musical excerpts was used . . . [to construct] . . . a Music Preference Test set up t measure 11 factors by 100 items on two sides of a long-playing record." It was hypothesized that the factors represented dimensions of personality, since the item groups did not correspond to musical schools or periods. The hypothesis was confirmed by many significant correlations with the 16 Personality Factor Questionnaire Test on two samples of 102 and 71 Ss. Application of the Music Preference Test to 98 mental hospital patients yielded significant factor measure differences between psychotics and normals, and between various psychotic syndrome groups.

Canter, Ralph R.

doi: 10.1037/h0057500pmid: N/A

A rating-scoring technique for evaluating free response answers is described and illustrated. An experimental (trainee) group of 18 supervisors and a matched control group of 18 each gave free response answers to four human relations questions. The E group was trained, and then both groups answered the questions again. Four social scientist raters sorted the 72 responses to each question into a 7-category forced-normal distribution. The pre and post-test score for each respondent was the sum of the category numbers assigned by the four raters over the four questions. Validity of the device as a measure of training effectiveness was previously reported (see ^W25:^n 7152). Interrater reliability is reported here: the summated pretest score reliability was .85; posttest score reliability was .88.

Anikeef, Alexis M.

doi: 10.1037/h0062861pmid: N/A

Merit-rating rank scores assigned to 19 faculty members by their students (ranking distributions based on approximately 1500 cases) were correlated with grading leniency and absence extensiveness rankings for the instructors. Grading leniency correlated .73 (N = 13) with the merit rating on the freshman-sophomore level, and .43 (N = 17) on the junior-sevior level. Absence extensiveness correlated negatively on both academic levels. It is hypothesized that the decrease in correlation with grading leniency could be accounted for by the selection process operating during the freshman-sophomore years.

Anderson, Scarvia B.

doi: 10.1037/h0062970pmid: N/A

Horst's formula for estimating the reliability of means of unequal numbers of scores is compared with an intra-class correlation formula derived by Cureton. The two formulas gave the same numerical value for r for grade-point ratios based on three quarters' work, but gave different values for grade-point ratios based on one quarters' work. "Where reliability estimates of unequal numbers of ratings were to be made, we would generally tend to use Cureton's formula when we were interested in reliability for the prediction of population behavior from a sample of that population or prediction from one sample of a population to another sample of the population."

Coleman, William

doi: 10.1037/h0056294pmid: N/A

A trial battery of six tests was administered to entering freshman engineering students. The tests which produced the best multiple correlation coefficients with grades in individual courses in the engineering curriculum were the Cooperative Algebra, Cooperative English, and Bennett Mechanical Comprehension Tests. The A. C. E. Psychological Examination and the Minnesota Paper Form Board did not contribute significantly to the prediction of course grades.

Long, Louis; Perry, James D.

doi: 10.1037/h0054293pmid: N/A

For a sample of 433 engineering school graduates, the effectiveness of four measures (high school average, General Mathematical Test, Scientific Verbal Ability Test, and Comprehension of Scientific Materials Test) in predicting a weighted four-year grade-point average was studied. The correlations between the criterion and the measures varied from .30 to .50. The correlation between the criterion and a composite test score was .53. Correlations found between the criterion and the Strong and Kuder interest questionnaires were not high enough to warrant inclusion of the interest measures in the selection battery.

Stanley, Julian C.

doi: 10.1037/h0055943pmid: N/A

To make possible the construction of intra-individual profiles on the Allport-Vernon Study of Values, a centile table is presented, calculated on the published norms, showing the percentile equivalents for men and women separately for scores on each of the six scales of the test. Cautions in the use of the table are pointed out.