Date of Award

4-1-1992

Document Type

Thesis

Degree Name

Master of Arts (MA)

Department

Psychology

First Advisor

Dr. James Thomas

Second Advisor

Dr. Lisa Scherer

Third Advisor

Dr. C. Raymond Millimet

Fourth Advisor

Dr. William L. Blizek

Abstract

Two potential moderating effects were investigated in a replication of a study investigating the saliency of applicant gender and information level upon preliminary employment decisions. The moderator of moral development was defined by Rest's Defining Issues Test, and the moderator of dogmatism was defined by Rokeach's Dogmatism Scale. These moderator variables were employed to explain the unusually inferior ratings given to female applicants in a low job-relevant information condition in the original study. Subjects were 60 undergraduate students in a pilot study and 244 undergraduate students in the main study. A 2 X 2 X 3 factorial design was used in the original study to evaluate four dependent variables: (a) the need to interview the applicant, (b) the perceived likelihood of the applicant's success on the job, (c) the perceived potential of the applicant's advancement within the company, and (d) the applicant's perceived managerial attributes derived from a composite score of five bi-polar managerial trait adjective pairs. The three factors were: (a) subject gender (male, female), (b) applicant gender (male, female), and (c) jobrelevant information (no, low, high). A pilot study indicated that a factorial design incorporating both moderator variables was not feasible. Therefore, two separate factorial designs were used in the main study, and dogmatism was chosen as the main moderator variable. A median split was made on this variable and included with the factors from the previous study to form a 2 X 2 X 3 X 2 factorial design with a reasonable balance in cell size achieved. A similar factorial design was made using the moral development measure, however cell size was much more unequal. A multivariate analysis of variance found significant effects for subject gender and information level but no interaction between applicant gender and information level as reported in the original study. The multivariate analysis was followed by univariate analysis of variance. The present study failed to replicate the original study; therefore, a valid assessment of moderator effects was not possible. However, no factor accounted for more than four percent of the total variance in the present study. The appropriateness of the statistical procedures and power of the statistical tests performed in the original and present studies were discussed.

Comments

A Thesis Presented to the Department of Psychology and the Faculty of the Graduate College University of Nebraska In Partial Fulfillment of the Requirement for the Degree Master of Arts University of Nebraska at Omaha. Copyright 1992, Konney Jay Larwood.

COinS