Holland's personality concepts as predictors of stability of choice



Journal Title

Journal ISSN

Volume Title



The purpose of this study was to test the validity of Holland's theoretical proposal that congruence of personality with chosen vocation, differentiation of personality and internal consistency of personality are predictors of vocational and educational behavior. To test this contention, it was hypothesized that the stability of a student's choice of college major would be directly related to: (a) the degree of congruence of major with personality, (b) the level of differentiation of personality, and (c) the degree of internal consistency of personality. It was also hypothesized that prediction of stability of choice could be improved by combining the independent variables—congruence, differentiation, and consistency—in a multiple relationship. It was further hypothesized that the relative importance of these variables in predicting stability are in the order given, i.e., congruence, first; differentiation, second; and consistency, third. This order of importance has been suggested by Holland. In order to ascertain if the validity of Holland's theory is related to either sex or age, it was hypothesized that there is a relationship between stability of choice and both age and sex. The subjects were 167 University of Houston students who were given the Self-Directed Search and a questionnaire designed to measure stability of academic major choice. Measures derived from these instruments, for each subject, included a three-letter code descriptive of personality type and the level of: differentiation, consistency, and stability of choice. Each subject's personality type was compared with the three-letter classification of the college major chosen by that student. This comparison resulted in a measure of congruence for each student. Pearson product-moment correlations between stability and the variables—congruence, differentiation, and consistency—were examined. A significant positive relationship was found between congruence and stability and between consistency and stability. No significant relationship was found between differentiation of personality and stability of choice. Multiple regression analysis was used to test the efficiency of predicting stability from congruence, differentiation, and consistency. When combined in an additive, linear relationship, all three constructs predict stability of choice. However, the efficiency of prediction is not improved by adding differentiation and/or consistency to congruence. As hypothesized, congruence is the most important predictor, followed in importance by differentiation and consistency, respectively. No significant relationship was found between stability and either sex or age. Findings supported most of Holland's theoretical proposals tested in this study, for both sexes and various age levels. It was unclear how useful differentiation and consistency measures are in predicting stability of choice. It was concluded that better measures of differentiation and consistency may be needed and that other variables should be identified that will improve prediction of educational and vocational behavior.