Predicted differences and differences between predictions

Mollenkopf, William G.

doi:10.1007/bf02288869

Cited by 14 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Philosophical complexity arises from the attempt to devise a system of tests which is sufficiently stable to meet the needs of the Air Force personnel programs while being sufficiently flexible for use." Early research on enlisted classification was influenced by theories and statistical procedures developed by Brogden (1946Brogden ( , 1951Brogden ( , 1954, Horst (1954), Mollenkopf (1950), and Thorndike (1949).…”

Section: Air Force Classification Testsmentioning

confidence: 99%

Enlisted Selection and Classification Tests: Precursors of the ASVAB

Thompson¹

2007

View full text Add to dashboard Cite

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

show abstract

Section: Air Force Classification Testsmentioning

confidence: 99%

Enlisted Selection and Classification Tests: Precursors of the ASVAB

Thompson¹

2007

View full text Add to dashboard Cite

show abstract

“…Earlier arguments for a distinct approach when placement or classification were the goals were based on ensuring low intercorrelations between job-specific predictor composites, and hence obtaining factor pure tests; i.e. tests loading high on unique factors for different jobs (Mollenkopf, 1950;Thorndike, 1950). Validity generalization proponents, however, using large datasets of test and criterion scores from the US Employment Service (Hunter and Hunter, 1984) and US military (Hunter, 1985), showed that gains in predictive validity from job-specific predictor composites with low intercorrelations were a product of artefactual error, such as sampling error, rather than true differential prediction of performance across jobs.…”

Section: Designing Test Batteries Composites and Predictors Of Performance: Predictive Validity Or Differential Assignment?mentioning

confidence: 99%

Classification techniques for person‐job matching

Zeidner

Scholarios

Johnson³

2001

Kybernetes

View full text Add to dashboard Cite

This paper presents the case for personnel systems based on maximizing the differential information gathered about individual abilities and their match to jobs. In the context of assignment to multiple jobs, such systems are shown to be more effective than those based on the currently dominant paradigm of maximizing predictive validity. The latter paradigm favours the measurement of general cognitive ability over multiple specific aptitudes. Recent differential approaches use computer simulation modelling of alternative hypothetical systems to evaluate potential efficiency. The paper reviews the theoretical background on the structure of human abilities which has led to these contrasting approaches to personnel system design, and presents evidence, based on the US Army selection and classification system, in support of the alternative approach. Individual test/aptitude profiles improve the efficiency of personnel selection and classification as well as academic, vocational and career counselling. They also provide a broader, potentially fairer definition of talent than a unidimensional indicator of cognitive ability, and a foundation for the design of learning and decision environments around learner and user profiles.

show abstract

“…Dif'ferentialPrediction. Since we know from Mollenkopf (1950) that predicted differences are equal to the difference between predictions, we are concerned here simply with the level of significance found in the differences between the predictions for two courses read from the expectancy charts. Starting with the ordinary formula, where a student has little information to go on and will welcome the statement that he has four out of five chances of doing better in one field than in another.…”

Section: Appendix Demonstration Of Expectancy Charts For Comparative mentioning

confidence: 99%

Comparative Prediction of Success and Satisfaction in College Major Fields

French¹

1961

ETS Research Bulletin Series

View full text Add to dashboard Cite

This bulletin completes the report of the Comparative Prediction Study ‐ College Level. It is to be regarded as a second part of RB‐59‐10, which described the background and method of the study, the aptitude tests, interest scales and personality scales, the administration to college freshman, the intercorrelations that were computed, and the need for averaging validity coefficients. The results of the testing are presented here as if the study had been done at the “average college.” Even this requires rather voluminous tables. The analysis shows the multiple validities that can be obtained from the battery and evaluates the battery's capability for differential and comparative prediction by computing the intercorrelations among predicted criteria and the validity of the predicted differences between criteria. An explanation and a demonstration is made of statistical corrections for the length of the tests, restriction of range in the major‐field groups, and “shrinkage” expected when the multiple validities are tried out on new groups. The battery is evaluated according to specifications laid down previously. A summary of the evaluation is as follows: The general level of prediction of major‐field grades is good (over .62) for some fields, but lower than this for too many others. Both the aptitude tests and interest scales contribute to the general validity; the personality scales were found to be much less useful. Validity for freshman grades is somewhat poorer; validity for satisfaction is low and limited entirely to the interest scales. Differential prediction of major‐field grades is good, mainly because of the presence of the interest scales.

show abstract

Predicted differences and differences between predictions

Cited by 14 publications

References 4 publications

Enlisted Selection and Classification Tests: Precursors of the ASVAB

Enlisted Selection and Classification Tests: Precursors of the ASVAB

Classification techniques for person‐job matching

Comparative Prediction of Success and Satisfaction in College Major Fields

Contact Info

Product

Resources

About