Erich P. Prien

The use of structured multi-domain job analysis for the construction of assessment center methods and procedures

This article addresses the defensibility of the assessment center based on a content oriented validation strategy. One reason for the confusion in this area is the fact that there are no agreed upon methods for constructing the assessment procedures. Therefore, the purpose of this article is to describe how structured multi-domain job analysis information can be used to develop an assessment center which is well grounded with regard to content validity.

An analysis of organizational behavior and organizational performance

Abstract Organizational unit measures of eight characteristic variables and 10 performance variables were obtained in an industrial plant employing some 25,000 persons in 64 units. Intercorrelations were generally low, within and between the variables in the two domains. A factor analysis yielded seven dimensions with significant loadings of variables from each domain. The results are discussed in terms of past research and some plausible causal relations.

An analysis of organization characteristics

Abstract A report of a factor analysis of 38 measures of organization characteristics was obtained for 107 small metal working firms. Nine factors were derived and interpreted with reference to the results of other empirical studies. The results differ from those of other studies and there appears to be limited concensus across all studies. Factors which appear in three or more studies are: Organization size, formalization, centralization of authority, extent of technology, and standardization with reference to skills and product complexity. The failure of researchers to interpret results more thoroughly seriously limits attempts to compare study results.

Interrater Reliability in Job Analysis: Differences in Strategy and Perspective

In spite of the increasing use of the content validity model for employee selection, there has been little focused research on the psychometric properties of the job analysis ratings used to determine job content. Sources of systematic variance include differences in job performance strategies and differences in perspectives on the job. In the current study, task importance ratings for a single job (social worker) are examined to determine whether rater experience and race will have significant effects on job performance strategy and job perspectives, as measured by job analysis ratings.

The Ghiselli Self-Description Inventory: A Psychometric Appraisal

A multitrait-multimethod analysis and a centroid factor analysis were used to examine and evaluate the measurement properties of the Ghiselli Self-description Inventory. The results support the use of the inventory and suggest the instrument can provide a relatively quick and simple method of measuring occupationally related personality traits and abilities.

PSYCHOMETRIC EVALUATION OF AN INTEGRATED ASSESSMENT PROCEDURE

The identification of supervisory and management potential has been an issue of priority in personnel selection for at least the past 40 years. Consequently, a wide variety of procedures have been developed and researched. A recently proposed method is the Brush and Schoenfeldt (1980) integrated assessment approach. Such an assessment system was developed and implemented at a southeast steel plant. A multitrait-multimethod-multirater analysis was used to assess the convergent and discriminant validity of the integrated assessment judgments. The results supported the use of the procedure for supervisors but not self-assessments. In each case, some evidence for convergent and discriminant validity was obtained. These results suggest that integrated assessment judgments of job skills possess the psychometric properties necessary for operational use.

Synthesizing Minimum Qualifications Using an Occupational Area Job Analysis Questionnaire

Determining minimum qualifications is largely an impressionistic process loosely related to training and education ratings of individuals. In determining and publishing minimum qualifications, the hiring official essentially delegates the authority for individual assessment of competency to external agencies. Minimum qualifications are, at best, a surrogate estimate of an individuals potential to perform the target job duties based on inferred knowledge, skills and abilities and competencies required for effective performance. Also, this requires the additional inferential leap backward in time to the individuals experience. Validation research on minimum qualifications has been limited, and, in application, the procedure is supported by a very impressionistic and global judgment process. The project described employs a systematic procedure linking the knowledge, skills and abilities and the competency content for an occupational area to the typical education, training and experience activities judged as required to achieve functional competence. Minimum qualifications for any single target position can then be synthetically derived from the database.

A Content-Oriented Approach to Setting Minimum Qualifications

Minimum qualifications (MQs) are typically estimated using a whole job, impressionistic approach. This article reports a series of quantitative job analysis studies using a decomposed approach with replication, and then extended to a comparison using a holistic approach for a sample of jobs. The results indicate that Subject Matter Experts (SMEs) agree on the years of educational attainment scaling of GED scale interval definitions (replicated), that job analysts can reliably assign GED ratings to tasks (replicated) and that job analysts can reliably assign GED ratings using a holistic approach. The decomposed and the holistic approach differ slightly in their results, with the decomposed approach producing a more conservative estimate of MQs and less differentiation among jobs. The conclusion drawn from these studies is that educational MQs can be established for clerical jobs using either of these techniques.

PEER RATINGS AND LEADERLESS GROUP DISCUSSIONS FOR EVALUATION OF CLASSROOM PERFORMANCE

An attempt was made to evaluate the internal consistency, temporal reliability, and validity of peer ratings as used in a classroom situation. Twenty-six Ss were divided into four small groups as part of a 6-wk. course in industrial psychology. Ss designed the rating device along four performance-related dimensions. Ratings of self and group peers were obtained at 2-wk. intervals; intelligence and class grades were used as validation criteria. It was found that internal consistency was low, but did not vary as exposure increased over time. The reliability of the four scales (Coordination and Organization; Ideas, Information and Resources; Effort; and Interpersonal Relations) did not vary appreciably. Factor analysis of the final ratings yielded Halo, Interpersonal Effectiveness, and Productivity factors. Several explanations for the high correlation (.77) between the Interpersonal Relations scale and class grades are offered.

The use of the position analysis questionnaire in selecting correlates of job performance among mentally retarded workers

Abstract The present study investigated the efficiency of various existing measures, mainly psychological tests, for predicting job performance of mentally retarded workers in a sheltered occupational shop. Twenty-two production-related jobs and one nonproduction-related job were analyzed with the Position Analysis Questionnaire resulting in two job attribute clusters. Correlations between 19 predictor variables and 5 criterion variables, which consisted of supervisor ratings of job performance, were then calculated separately for subjects performing preferred jobs in each attribute group. The results indicated that, contrary to past research, existing measures are predictive of performance on some but not all jobs in a sheltered workshop.