Journal of Career Assessment

During the past 15 years the Career Decision Scale (CDS) has been used primarily to assess degree of indecision. Controversy has arisen around the issue of whether it should thus be considered a unidimensional measure or whether it is a multidimensional measure of several dimensions of career indecision. Although a number of studies reported significant empirical support for the multidimensionality of the CDS, a study by Martin, Sabourin, Laplante, and Coallier (1991) claimed to present evidence supporting its unidimensionality. The present study refutes those claims by demonstrating serious methodological and conceptual errors and omissions in the findings of Martin et al. (1991) and by presenting further empirical evidence in support of multidimensionality.

For more than 15 years, various writers have debated the underlying structure of the Career Decision Scale. Some are convinced it is best conceived to be a unidimensional measure, while others believe it is multidimensional. The present comment focuses on the paper by Shimizu, Vondracek, and Schulenberg (this issue) which promotes the multidimensional nature of the measure. An attempt to highlight some of the issues in this debate from the point of view of one of the Career Decision Scale's authors is made.

In two related papers, Shimizu, Vondracek, Schulenberg, and Hostetler (1988) and Schulenberg, Shimizu, Vondracek, and Hostetler (1988) proposed a four-factor model of the Career Decision Scale (CDS; Osipow, Carney, Winer, Yanico, & Koschier, 1976). In Martin, Sabourin, Laplante, and Coallier (1991), we tested the relevance of this model using more robust estimation techniques. Our conclusions contradicted those of Shimizu et al. (1988) and Schulenberg et al. (1988): We concluded that the CDS has no clear dimensional structure and recommended the use of the total score and not that of its subscales. Shimizu, Vondracek, and Schulenberg (1994) is a reply to our paper in which these authors raise doubts about our methodological choices and present results of the reanalysis of their original data using our methodology. Here we explain why most of the critics of our methodology and interpretations are mislead, and why the results they got with the use of our methodology proves exactly the opposite of what they claim. We also review other matters discussed by Shimizu et al. (1988): cross-cultural comparison and clinical issues. We conclude by reasserting that the CDS is a reliable instrument for the measurement of career indecision level, but not of anything else because it has no clear dimensional structure.

Our previous critique of Martin, Sabourin, Laplante, and Coallier (1991) included 10 serious criticisms of their methods and conclusions regarding the dimensionality of the Career Decision Scale (CDS). Although the quantity of the Laplante, Coallier, Sabourin, and Martin (1994) response was not bad (i.e., they attempted to address 7 of the 10 criticisms), the quality of their response was disappointing because it confused fact with fiction, and included a wealth of extraneous and irrelevant material. Nothing in their response convinced (or even tempted) us to retract any of our original criticisms. Most importantly, the equivalence of their French version of the CDS with the English version was still not demonstrated, our four-factor model was still not disconfirmed, and their model testing procedures remain suspect (e.g., for some undisclosed reason, Laplante et al. (1994) and Martin et al. (1991) are at odds as to whether they accepted their one-factor model). Based on our findings (as well as those of Martin et al. [1991]— despite their conclusions, their findings support our model), the fact remains that our four-factor model provided the best fit among the several models tested, indicating firm support for the multidimensionality of the CDS. We found ourselves in substantial agreement with Osipow (1994), and we urge him to refine the CDS by adding dimension-specific items and revising double-barreled items.

This study examined the relation between vocational interests and job congruence and (a) stability at work and (b) supervisors' evaluations. Congruence scores were based on a comparison between interest scores, as measured before employment, and occupational field after 1 year. Significant correlations (.19 to .31, n = 774) were found between congruence and job stability (persistence) and between congruence and performance (.09 to .27, for 90 males employed in the Business or Technology occupational fields), but not among 51 employed in the Organization field. Differentiation was found to make a negligible contribution in predicting stability and performance evaluations. The findings are discussed in light of the size of the correlation coefficients, the diversity of the Organization field, the Courses Inventory as a means for measuring vocational interests, and the method of measuring congruence differentiation.

A cluster analysis procedure was used to test a continuum versus an interactional view of career indecision. It was hypothesized that cluster groups would differ across cognitive and affective dimensions of career indecision. Clients experiencing career decisional problems were identified using a sample of college students from career planning courses (N = 325). Responses to the Career Factors Inventory (CFI; Chartrand, Robbins, Morrill, & Boggs, 1990), which measures both cognitive and affective dimensions of career indecision, were cluster analyzed using a disjoint nonhierarchical procedure. A four-cluster configuration, which revealed interactions between cognitive and affective dimensions of career indecision, was identified as the optimal cluster solution. This cluster solution was validated and results were discussed in terms of integrating career decisional typologies into a theoretical framework.

Stereotyping, prejudice, and discrimination serve as major barriers to women's career adjustment and advancement. To assess the current level of attitudes and the change in attitudes toward members of diverse populations in a variety of occupational roles, psychometrically sound assessment instruments are needed. The Women as Managers Scale is one such instrument. Exploratory factor and principle components analyses have been used in past research to challenge the dimensionality and reliability of the Women as Managers Scale. Confirmatory factor analyses revealed that a three-factor model was superior to a one-factor model, and was stable across subgroups. Evidence for discriminant validity among the factors was, however, weak. A possible explanation of the lack of discriminant validity is offered.

The purpose of this study was to investigate the relationship between task-specific self-efficacy as measured by the Task-Specific Occupational Self-Efficacy Scale (Osipow & Rooney, 1989) and career indecision as measured by the Career Decision Scale—3rd Rev. (Osipow, Carney, Winer, Yanico, & Koschier, 1976) for females classified as possessing high or low egalitarian sex-role attitudes. Subjects were 168 female college students who were classified as high or low egalitarian according to their scores on the Sex-role Egalitarianism Scale form KK (Beere, King, Beere, & King, 1984). Pearson-product moment correlations were computed between self- efficacy scores and career indecision scores for the two groups. The results provide minimal support for the hypothesis that there would be a significant relationship between career indecision and self- efficacy as moderated by sex-role orientation. These results are limited due to a negatively skewed distribution on the egalitarianism measure.