|
Article Excerpt
Career indecision is a common presenting problem for students seeking counseling at high school guidance offices and college counseling centers. Recent research indicates that 10% of college-bound students are undecided about their college major and career selections (American College Testing Program, 1999). A considerable body of career indecision research has accumulated, driven by the fundamental goal of specifying distinct counseling interventions for different types of indecision. Conceptualization of career indecision has advanced from a simple dichotomous classification of decided-undecided (Holland & Holland, 1977), to a level model (Savickas, 1989), to interactional classification (Chartrand et al., 1994; Jones & Chenery, 1980), to multiple-type classification based on cluster analysis of career development and personality variables (e.g., Lucas & Epperson, 1990; Wanberg & Muchinsky, 1992). There is growing consensus that there are distinct career indecision types and that differential counseling interventions are appropriate for each type (Lucas, 1993).
Success in the development and implementation of differential interventions ultimately depends on the validity of the career indecision types. Researchers typically have gathered concurrent measures of career development variables to support the validity of the career indecision types generated by cluster analyses. However, there have been no studies documenting the subsequent academic, career, or student development outcomes associated with the different types. Although certain types of indecision consistently emerge from cluster analyses conducted by independent researchers, there is no evidence that distinct behavior patterns are associated with each type. For example, it remains to be demonstrated that a developmentally undecided student (Chartrand et al., 1994) is more likely than a chronically indecisive student (Callahan & Greenhaus, 1992) to respond positively to career counseling, to achieve good grades, to persist in attaining a baccalaureate degree, or to make a smooth transition to employment after graduation. The purpose of this article was to identify career indecision types and to evaluate the initial evidence for the validity of the types by studying their response to a standard career exploration intervention, with reduction in career indecision as the predictive validity criterion.
SUMMARY AND ANALYSIS OF CAREER INDECISION TYPOLOGY RESEARCH
We limited this summary to examination of multivariate studies of career indecision types. Correlational studies of career indecision measures were excluded from this review because multivariate studies have produced fewer spurious findings and have yielded more cogent descriptions of the career decision problems. Three indecision types have emerged consistently from independent investigations using different career and personality variables to form types. The first type has been labeled with terms that include "confident decided" (Wanberg & Muchinsky, 1992, p. 74), "decided" (Savickas & Jarjoura, 1991, p. 87), and "ready to decide" (Chartrand et al., 1994, p. 65). These individuals experience little career indecision and little or no career choice anxiety; they also have sufficient career information and well-developed vocational identity. They do not seem to need career counseling.
The second career indecision type has been labeled "confident but uninformed" (Larson, Heppner, Ham, & Dugan, 1988, p. 442), "developmentally undecided" (Callahan & Greenhaus, 1992, p. 227; Chartrand et al., 1994, p. 65), "well adjusted and needing information" (Lucas, 1993, p. 442), and "specifying a choice through advanced exploration" (Savickas & Jarjoura, 1991, p. 87). These individuals have a focal need for career information and have not begun career planning. However, they experience little career choice anxiety, are goal-directed, and have mature ego identity. Lucas (1993) conjectured that developmentally undecided individuals need brief, information-based career counseling.
The third career indecision type has been labeled "anxious undecided" (Wanberg & Muchinsky, 1992, p. 76), "planless avoiders" (Larson et al., 1988, p. 442), "chronically indecisive" (Callahan & Greenhaus, 1992, p. 227), "chronic indecision-impaired development" (Rojewski, 1994, p. 361), "anxious and unclear on goals" (Lucas & Epperson, 1988, p. 463) and "indecisive" (Chartrand et al., 1994, p. 67; Savickas & Jarjoura, 1991, p. 87). These students need career information, are highly anxious, lack decision-making confidence, perceive barriers to career decision making, and have low self-esteem and poorly developed vocational identity. In addition to acquiring career information, these individuals seem to need counseling to improve their problem-solving and decision-making abilities and to develop more positive self-esteem.
Although the initial research in this area has been informative, there are five significant limitations in the empirical attempts to create career indecision typologies. The first and most challenging problem is the dearth of predictive validity evidence for these types. Only Lucas (1993) examined predictive validity by studying various types in relation to subsequent number of counseling sessions; she found no differences in the number of sessions attended by different types of clients. Well-intentioned efforts to delineate a career indecision typology will remain an academic exercise until subsequent behaviors of different types are shown to differ meaningfully over time. Important predictive validity criteria include response to career counseling, academic adjustment, and quality of the transition to academic majors and work. The establishment of predictive validity data will strengthen the rationale for providing different career counseling interventions to different types of clients.
The second limitation is the failure to consider academic aptitude in identifying career indecision types. It is difficult to understand the exclusion of ability variables from attempts to create typologies because early research linked academic ability and achievement to indecision (Lunneborg, 1975, 1976). In addition, counselors have commonly observed that poor academic performance can precipitate doubt about one's career goals. For example, freshman engineering students who receive poor grades in calculus and physics courses have to quickly reevaluate the feasibility of pursuing an engineering degree. Inclusion of aptitude variables may provide more information about the causes of career indecision as well as the appropriate focus for counseling interventions. Consider the example of counseling a student with low academic achievement. It may be more important to work with this student to focus on available career options than it is to clarify career interests. Such students may have crystallized interests but struggle to find available options because of their low grade point averages.
The third limitation is that typology research has not been conducted exclusively with undecided students. Only Chartrand et al. (1994) and Lucas (1993) studied students who had declared themselves as undecided. The typical study has included convenience samples from readily available subject pools that have included many participants with crystallized career goals. If the goal of identifying types is to inform the development of appropriate counseling interventions for undecided clients, then there is a need to use exclusively undecided populations in the search for career indecision types.
The fourth limitation is related to the method used to interpret the outcomes of statistical analyses used to identify decision types. Cluster analysis (Borgen & Barnett, 1987), the most popular statistical technique for identifying decision types, uses career and personality variables to create and describe types. Cluster analysis leads to the formation of maximally homogeneous types that are maximally distinct from all other types on the cluster variables. This is an appropriate statistical analysis for identifying types. However, the interpretation of cluster analysis results can be problematic because the typical practice is to compare the resulting types using the means for each variable included in analysis. Consider the example of using the Career Decision Scale (CDS; Osipow, 1987) as a cluster variable. It is likely that one of the resulting types will have higher CDS scores and another will have lower CDS scores; this...
|
