Date of Award

1997

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Psychology

Abstract

The focus of this study wag to determine if women who report and women who do not report sexual harassment could be differentially described by a subset of the following variables: age, marital status, feminist ideology, frequency of behavior, perceived offensiveness, perceived outcomes, normative expectations. This study was based on a model for reporting sexual harassment by Brooks and Perot (1991) and extended the original test of this model by adding two variables, normative expectations and perceived outcomes, which were not tested by Brooks and Perot. The sample for this study consisted of 178 undergraduate juniors and seniors from upper level classes at a midwestern university during the fall 1993 semester. Of this sample, 53 reported and 81 did not report sexual harassment at level one of the Sexual Experiences Questionnaire. The null hypothesis was tested by a stepwise discriminant analysis. Tests of significance were computed, ascertaining the most parsimonious subset of discriminating variables. Tests of classification accuracy and total variance explained in the dependent variables were conducted. The criterion groups were significantly differentiated by four of the seven predictor variables. Reporters were found to have more feminist attitudes, experienced harassing behaviors more often, perceived harassing experiences as more offensive and perceived more negative outcomes for reporting. The variable contributing most to the separation of the reporter and non-reporter groups was frequency of behavior. Those variables that did not contribute to group differences were age, relationship status, and normative expectations. Conclusions give partial support to the Brooks and Perot (1991) model for reporting sexual harassment. The significance and non-significance of five variables tested in the original study were supported by the present study. The part of the model not tested in the original study was given partial support, perceived outcomes was significant, normative expectations was not. Replications of the present study with revisions should be conducted to further test the model.

Download

Share

COinS