Career choice anxiety, coping, and perceived control. (Articles).

Extrapolating from D. H. Barlow (2000), the authors explored whether perceived control moderated the relation between coping with career indecision and choice anxiety among 126 women in low-level jobs. Analyses of the women s career indecision, coping, perceived control, and career choice anxiety scores through regression identified the moderator effect. Perceived control interacted with problem-focused coping to increase accountable variance in choice anxiety (p < .05). Women perceiving high control and doing more problem-focused coping reported lower anxiety than did women doing comparable coping but perceiving lower control. Implications are discussed for interventions with women in low-level jobs.

**********

Studies confirm the association of choice anxiety and career indecision (Fuqua & Hartman, 1983; Fuqua, Seaworth, & Newman, 1987; O'Hare & Tamburri, 1986). These studies, however, are primarily of high school and college students. The association of career indecision and choice anxiety among out-of-school adults between ages the ages of 22 and 32 years in low-level jobs remains virtually unexamined. Yet, such adults are likely to make multiple occupational changes in the first several years after they leave school (Bureau of Labor Statistics, 1998).

The relation of career indecision and choice anxiety is important to counselors. Spokane (1991) has noted that career choice anxiety may lead some prospective clients to eschew career counseling and other clients to discontinue counseling prematurely. In other words, their anxiety may prevent them from the very actions that would relieve their indecision and reduce their anxiety. Inclusion of anxiety reduction in career decision-making interventions seems to enhance client career decision-making efforts but has not been shown to lower career choice anxiety in itself (Mendonca & Siess, 1976). Because choice anxiety seems to constrain the coping of some clients with career indecision, it is important to identify factors that may contribute to the relation between coping with career indecision and choice anxiety.

Barlow's (2000) model of vulnerability for anxiety and depression suggests one factor--perceived control of one's career choice--that may elucidate the relationship. He proposed that repeatedly experiencing a low sense of control of an anxiety-invoking stimulus creates vulnerability for a heightened anxiety reaction to the stimulus. With repetitions, such low perceived control becomes an amplifier that acts as a moderator of the process occurring between exposure to the stimulus and ensuing anxiety, implicitiy through coping and other variables. Extrapolating from Barlow, one would hypothesize that persons coping with career indecision who perceive control over their career-choice context, both the personal and environmental facets, are likely to experience lower choice anxiety as they increase their level of coping. Individuals who perceive low control, in contrast, would not be expected to reduce their choice anxiety as much, if at all. A participant's comments about perceived control to the first author weeks after the study suggest what it meant for the participants. She said,

I get stressed not from deciding but from everything around it. I have so much stress from having to think of rent, furniture, and food. I need this and that. Of course it would be easier to decide if everything else were taken care of.

Barlow's (2000) model is consistent with the conclusion of Taylor (1982). From her review of studies of the correlates of career indecision among college students, Taylor concluded that students' internal locus of control--an enduring disposition to feel control over one's various contexts--moderates their efforts to resolve career indecision. Moreover, it seems consistent with the recent findings of Luzzo, James, and Luna (1996). They reported that strengthening the perceived control of students over their career choice increased the level of their career exploration relative to students whose perceived control was not strengthened.

Some of the apparent advantage in reducing choice anxiety for persons who perceive high control may stem from their use of problem-focused, rather than emotion-focused, coping. Lazarus and Folkman (1984) and Lazarus (2000) proposed that persons with high perceived control over a stressor, such as career indecision, are likely to use more problem-focused coping. In other words, they are more likely to take actions such as acquiring the information and skills needed to resolve the problem by such means as asking a respected person for advice and analyzing the problem. Individuals with lower control, in contrast, are expected to use more emotion-focused coping, that is, actions to distance themselves from the stressor, such as fantasizing about how things might turn out or telling themselves things to feel better. Although Lazarus and Folkman (1984) posited that problem- and emotion-focused actions complement one another in reducing anxiety, they allowed that problem-focused actions are more likely to lessen the anxiety caused by a chronic stressor.

Studies of the relation of problem- and emotion-focused coping with career decision-making variables suggest that problem-focused coping is more effective in alleviating choice anxiety. In their study of farmers in a career transition workshop, Heppner, Cook, Strozier, and Heppner (1991) found that the farmers' problem-focused coping correlated significantly with their perceived control and progress in their career transition, whereas their emotion-focused coping related significantly to their perceived stress and level of depression. The authors of the study did not consider the relation of perceived control to the other variables in the study: stress, depression, or progress; nor did they examine whether the interaction of control and coping affected the farmers' feelings. Earlier, O'Hare and Tamburri (1986) had reported that a form of coping, which included perceived controllability and primarily problem-focused strategies, related to low choice anxiety after partialling out the effects of trait anxiety.

Knowledge of whether a tendency to perceive low or high control over career choice relates to the career indecision, coping, and choice anxiety of subgroups of individuals with historically low control over their career choices is likely to be important in developing interventions and formulating policy for assisting them. One such group is young women in low-level jobs that are not part of a career ladder. Their circumstances may prevent many of them from believing that the issues related to their career choice can be resolved (McWhirter, Torres, & Rasheed, 1998). They are not receiving either formal or on-the-job training to equip them with skills to advance, and few of their coworkers have advanced. Certainly, the limited number of people like them who have entered management and professions may make many of them skeptical about their prospects (Betz, 1994). Although the career choice anxiety of college students has been studied extensively, there has been almost no attention given to the career choice anxiety of women in working-class jobs (Meara, Davis, & Robinson, 1997).

To explore the contribution of perceived controllability of career-choice context in understanding the career indecision, coping, and choice anxiety of women in low-level jobs, this study tested the following four hypotheses:

1. Women who perceive more control report more problem-focused coping than do women who perceive less control.

2. Women who perceive less control report more emotion-focused coping than do women who perceive more control.

3. More problem-focused coping is associated with less choice anxiety when perceived control is high but not when perceived control is low.

4. More emotion-focused coping is not associated with low choice anxiety regardless of whether perceived control is high or low.

Method

Participants

The participants were 126 single, female, unlicensed clinical health workers who were employed full-time in a chain of medical offices in California for at least 1 year (M = 2.43, SD = 1.47). These participants were recruited for the study because they were in a low-level, semiskilled position that was not part of a career ladder, so their jobs offered only a slight possibility for advancement. Participants ranged in age from 22 to 32 years (M = 25.79, SD = 3.55). The cutoff of 32 years was used because career development theories, such as Super's (1990) theory, suggest that persons older than 32 years old may focus on different career development concerns than those who are under 32. Neither years in the job nor age correlated at the p = .05 level with any of the variables described later in this article. Participants' self-reported ethnic backgrounds were 67% Caucasian, 14% Hispanic, 11% African American, 6% Asian American or Pacific Islander, and 1% Native American. In response to a question regarding years of schooling completed, they reported a mean of 14.04 (SD = 2.28). To ensure that career choice was an issue for them, participants rated the statement "At this point in my life I feel choosing a career is important." The 126 participants rated either exactly like me or very much like me rather than slightly like me or not like me. On a career time line on which they indicated their position and where they expected to be, all the participants indicated that they expected to be further along. The points on the line were trying out jobs, settling into a job other than my career, exploring my real career choice, training for my career, exploring a second career choice, and advancing in my career. Seventy-six percent did not list a career choice, whereas the other 24% listed occupations that were considerably more advanced than their current one, as indexed by the Nam and Powers (1983) scale.

Measures

Career Indecision. The 16-item Career Indecision subscale of the Career Decision Scale (CDS; Osipow, Carney, Winer, Yanico, & Koschier, 1987) was used to measure career indecision. Each item describes a state of uncertainty about an issue pertinent to career choice. Respondents indicate the degree to which each description fits their circumstances on a 4-point continuum from exactly like me (4) to not at all like me (1). High scores indicate more indecision. The scale has been used extensively in research and practice, and, in her review of it, Harmon (1994) opined that if one was looking for an overall measure of career indecision for use in research or practice, one could find no better measure. More recently, Osipow and Winer (1996) compared the total score with different factor scores found in several factor analyses and argued that the total score was the most stable overall measure of indecision. Osipow et al. (1987) reported test-retest reliability coefficients ranging from .70 to .90. The alpha coefficient for our participants was .89. According to the CDS manual, their mean of 30.23 (SD = 10.39) placed participants in this study at about the 37th percentile of adults seeking continuing education and at the 73rd percentile of women returning to college. The difference between the two adult norm groups made it difficult to interpret participants' scores. Their raw scores indicated that, as a group, the participants reported that they had resolved many of the issues involved in a career decision, even though they were not where they believed they should be in their career and believed that their career choice was important.

Career choice anxiety. This was measured by a modification of the State Anxiety Scale of the State-Trait Anxiety Inventory (STAI), Form Y (Spielberger, Gorsuch, & Lushene, 1983). The modification asked respondents "to indicate how you feel right now, that is, at the moment when you think about being decided or undecided about your career." The scale contains 20 positive- and negative-feeling items that one rates as self-descriptive or not on a 4-point scale that ranges from not at all (1) to very much so (4). After recoding negative items, high scores equal higher anxiety. Fuqua and Hartman (1983) and O'Hare and Tamburri (1986) used a similar modification of the scale in their studies of college students. The manual reports alpha coefficients ranging from .91 to .93 for six samples of adults. In our study, the participants' alpha was .93. Their mean of 43.29 (SD = 11.55) placed them at about the 77th percentile of female workers (Spielberger et al., 1983). This suggests that they tended to feel higher anxiety about their career choice than the norm group felt about their current concerns. Their mean score put the group among the group of college students that O'Hare and Tamburri had characterized as highly anxious about career choice.

The Ways of Coping Checklist (WOCC; Lazarus & Folkman, 1988). This is a list of 66 coping strategies derived from interviews with people about how they had dealt with recent stressful events. As is generally done, the WOCC was adapted through oral and written instructions to gauge the participants' behaviors in coping with career indecision while they were experiencing choice anxiety. Participants indicated the ways of coping that they had used at the point of most stress in trying to cope with being undecided or with the issues arising after their decision. Tiedeman and O'Hara (1963) pointed out that the career decision process involves both a choice and an implementation phase.

There are 28 problem-focused strategies and 38 emotion-focused strategies listed in the WOCC. An example of a problem-focused behavior is "Tried to analyze the problem to understand it better." An example of an emotion-focused behavior is "I told myself things that helped me feel better." For participants, the two scales correlated r = .37, suggesting that they were not strongly related. Although some researchers divide the two scales into various subscales, we used total scores to encompass a broad range of coping actions. Our participants were not enrolled in a career intervention, so we did not anticipate particular subsets of actions. In Heppner et al. (1991), a shortened problem-focused scale correlated with perceived control and progress in deciding, and a shortened emotion-focused scale related to stress and depression. Folkman and Lazarus (1980) reported an alpha of .80 for the problem-focused coping scale and .81 for the emotion-focused scale. Because responses to the WOCC are expected to change with time and the target of coping, Lazarus and Folkman (1988) argued that test-retest reliabilities are not pertinent. For participants, the alphas for both scales were .80. Their means were 17.12 (SD = 5.19) on problem-focused coping and 20.52 (SD = 6.19) on emotion-focused coping. The use of more than half of each type of listed strategy suggests that the women were trying multiple ways of dealing with their career decisions. As Heppner et al. found for the farmers who were dealing with career transitions, the women in our study used a numerically higher percentage of the listed problem-focused than emotion-focused strategies.

Perceived control. Participants rated the two statements, "I had a great deal of control" and "I had very little control," on 4-point ratings ranging from exactly like me (4) to not at all like me (1). The directions asked them to indicate their overall state of mind regarding their control of their career choice at the point when its challenge was most intense. Ratings for the second statement were reversed and added to the ratings for the first statement, so that higher scores signified greater control. Controllability has often been assessed with one- or two-item ratings. For example, Heppner et al. (1991) used a 7-point rating of the question, "How much control do you feel you have in the process of making this career transition?" More recently, Long (1998) assessed controllability with a single 4-point rating. For participants, the two ratings correlated .76. The mean of 5.95 (SD = 1.98) suggests that, as a group, the participants felt a moderate degree of control.

Procedure

During lunch or after hours at the work site, the first author explained to potential participants that the purpose of the study was to investigate issues in choosing a career. She then distributed packets that participants completed anonymously and returned by mail. The cover sheet asked for demographic information and was followed by the Career Indecision Scale, the Choice Anxiety Scale, and either the WOCC or the perceived control ratings in random order. Of the 275 women receiving packets, 175 (64%) returned them. However, 24 women were excluded because they were either not in entry-level positions or were over 32 years old, and 25 others were excluded because they rated the importance of career choice to them as only slightly like me or not like me.

Hypotheses 1 and 2 were tested by correlation. Following Holmbeck's (1997) suggestions about testing for moderator effects, Hypotheses 3 and 4 were tested by regression. For Hypothesis 3, we computed the full hierarchical regression on career choice anxiety of career indecision, perceived control, problem-focused coping, and the interaction of perceived control and problem-focused coping, and we termed the full model Step 2. It contained Step 1, which is the regression of career indecision, perceived control, and problem-focused coping on career choice anxiety without the interaction of perceived control and problem-focused coping. Similarly, for Hypothesis 4, we computed the full hierarchical regression model on career choice anxiety of career indecision, perceived control, emotion-focused coping, and the interaction of perceived control and emotion-focused coping. The full model contained Step 1, which was the regression of career indecision, perceived control, and emotion-focused coping on career choice an xiety without the interaction of perceived control and emotion-focused coping. It was necessary to center the variables because the interaction effects introduced unacceptable levels of collinearity with variance inflation factors (VIF) greater than 7.0. Centering, that is subtracting the mean of the variable from each observation, successfully reduced the degree of collinearity to acceptable levels, with VIF less than 1.2 for all variables (Pedhazur, 1997).

Results

To test Hypothesis 1 that women who perceived having more control will do more problem-focused coping in response to career indecision, the two variables were correlated, yielding an r = .27, p < .01. As hypothesized, women who perceived having relatively more control reported relatively more problem-focused coping. To test Hypothesis 2 that women perceiving less control will do relatively more emotion-focused coping, the two variables were correlated, yielding an r = -.20, p < .05. As predicted, the women who perceived having relatively less control reported relatively more emotion-focused coping.

We used hierarchical regression to test Hypothesis 3 that more problem-focused coping is associated with less choice anxiety when perceived control is high but not when perceived control is low. The variables problem-focused coping, perceived control, career indecision, and the interaction term consisting of the product of problem-focused coping and perceived control were regressed on choice anxiety, as reported in Step 2 of Table 1. Career indecision was entered to control for the participants' level of indecision. Then, Step 1 without the interaction term was examined. In Step 1 of the equation in Table 1, problem-focused coping did not relate to choice anxiety at a statistically significant level, but both career indecision and perceived control did. Comparison of the regression before and after entry of the interaction term in Step 2 showed that its entry changed the interpretation of variables. Specifically, its inclusion increased the portion of the variance accounted for by 3%, F(3, 122) = 8.67, p < .001. In Step 2, perceived control continued to have a direct effect, but problem-focused coping now also related to anxiety through its interaction with perceived control. Although the correlation of problem-focused coping and choice anxiety was only r = -.05, the beta of the interaction between coping and appraisal of control with choice anxiety was -.18. This suggests that the reason a stronger relationship between coping and anxiety was masked was because the effect of each of the predictors in the model varied as a function of the value of the other predictor. The beta of -.18 for the interaction indicates that increasing control by one standard unit changes the slope of coping by -.18 for the participants (i.e., as control increases, the association between coping and anxiety becomes clearer). More coping is associated with decreased anxiety. Without the interaction, the effect of coping was constant for all levels of control. Considering perceived control increased the amount of variance accounted for in the relationship between problem-focused coping and anxiety.

To ascertain whether the interaction of perceived control and problem-focused coping supported Hypothesis 3, we constructed categorical variables for perceived control and problem-focused coping, and we inspected the residualized means of choice anxiety (Tracey, 2000). Perceived control was split at the median, problem-focused coping was divided into quartiles, and career indecision was controlled. The residualized means of choice anxiety for the high perceived control, third- and fourth-quartile coping groups were -6.46 and -6.50, respectively. These were the lowest means of the groups. In contrast, the means for the low perceived control, third- and fourth-quartile coping groups were 6.97 and 5.78, respectively. These were the highest means of the groups. For the high perceived control, second-quartile coping group, the mean of 1.95 was relatively low; whereas for the low control, second-quartile coping group the mean of 3.35 was relatively high. For both high and low perceived control, first-quartile copin g groups the means were -4.12 and 1.73, respectively. The high perceived control, low-coping group was relatively low in choice anxiety, whereas the low-control, low-coping group was near the midpoint on choice anxiety. Inspection of these residualized means supports Hypothesis 3 and suggests that more problem-focused coping in the context of higher perceived control was related to lower choice anxiety, whereas more problem-focused coping with low control was related to higher anxiety. The residualized choice anxiety means of the four Low Control x Problem-Focused Coping quartile groups were all higher than the residualized means of the four High Control x Problem-Focused Coping groups, supporting Barlow's (2000) suggestion that less perceived control amplifies anxiety.

A hierarchical regression was also computed to examine Hypothesis 4 that more emotion-focused coping is not associated with low choice anxiety, regardless of whether perceived control is high or low. The variables of emotion-focused coping, perceived control, career indecision, and the interaction term of Emotion-Focused Coping x Perceived Control were regressed on anxiety in Step 2. Then, Step 1 without the interaction term was examined. In Step 1, emotion-focused coping did not relate to choice anxiety at a statistically significant level ([beta] = .01, p > .05). Its r = .10, p > .05 was not related. In Step 2, the interaction of emotion-focused coping and perceived control did not produce a significant increase in [R.sup.2]. This offered limited support for the hypothesized lack of relationship.

Discussion

The relationships among the participants' choice anxiety, coping, and perceived control are consistent with the hypotheses from Barlow (2000). They suggest that perceived control of career choice context may be useful in understanding the career decision making of women in low-level jobs. The women who perceived more control reported more problem-focused coping than did women who perceived less control, whereas the women who reported less control did more emotion-focused coping. Regardless of the magnitude of their career indecision, participants who perceived more control and reported more problem-focused coping felt relatively less choice anxiety, whereas those women who perceived less control but also did more problem-focused coping felt relatively more choice anxiety. Unless the interaction of perceived control and problem-focused coping was considered, a relationship between the problem-focused coping and choice anxiety would not have been detected. Regardless of the type or level of coping, perceived co ntrol related to choice anxiety. The findings from a cross-sectional study cannot establish causality or the direction of the variables' interrelationships. Yet the direct and indirect relations of perceived control and choice anxiety in the context of career choice suggest the need to monitor the level of perceived control before, during, and after a career intervention.

Recognizing clients' perceived control as important has implications for career counseling services and policy. Regarding career interventions, the finding that perceived control related to choice anxiety regardless of the degree of career indecision suggests that factors affecting a client's perceived control should be identified and addressed before, or along with, the traditional components of career decision making. Interventions such as those introduced by Luzzo et al. (1996) for dispelling dysfunctional attributions and by Mitchell and Krumboltz (1987) for countering obstructive beliefs might be combined and augmented with other strategies.

Such augmentation may be necessary to assist clients whose control of career choices has been historically low to strengthen their perceived control. Because Barlow (2000) postulated that a sense of low control develops over multiple experiences, correcting it may take longer than it would for students like those treated by Luzzo et al. (1996) and Mitchell and Krumboltz (1987). For example, to adapt the Luzzo et al. video intervention, one would first choose models with whom the clients could relate and portray the models discussing concerns such as locating and gaining access to training in which a mediocre student might succeed. To expand the intervention, one might have small groups of clients view the video together, describe similar kinds of information they need to obtain, identify ways of getting it, and obtain group support for trying the ways and updating the group at subsequent meetings. The Mitchell and Krumboltz cognitive restructuring approach might be expanded in a similar way. First, one might have a graduate of the program describe how identifying a belief that impeded her career development enabled her to use new strategies and overcome the obstacle. Then, in individual didactic counseling sessions like the one described by Mitchell and Krumboltz, one could show clients that they can switch to alternate coping strategies rather than persisting with ones that are not working by guiding them to try a new strategy to overcome one of their current obstacles.

Regarding policy, the findings call for provisions in career development programs to boost the perceived control of disadvantaged women as soon as they are recruited and to monitor their perceived control during a program. Both Krumboltz (1993) and Spokane (1991) proposed that increased choice anxiety may prevent people from initiating career counseling, and Spokane noted that rising choice anxiety during career counseling may lead clients to terminate prematurely. Inattention to perceived control may explain why well-intentioned, socially applauded programs fail to enroll disadvantaged clients or to assist them once they enroll.

By highlighting the probable usefulness of the construct of perceived control, our study indicates the need for more information about perceived control and its measurement. The findings indicate that our two-item measure was internally consistent for the participants and related to the other variables, as expected. Still, more items that assess control are desirable to increase the scale's reliability. In addition, further study of the degree to which the measure captures the respondents' overall perceived control of their career choices, using interview procedures such as those described by McWhirter et al. (1998), is needed. Moreover, information is needed about whether strengthening perceived control during the decision-making process affects type or amount of coping.

Although participants' emotion-focused coping related to their perceived control, our study did not detect a role for emotion-focused coping in career choice anxiety. One possible explanation is that it is not the variety of emotion-focused behaviors but their frequency that affects choice anxiety. For example, if a woman told herself things to help herself feel better only once or twice, that may not have given relief over a period of distress. The WOCG completed by the participants operationalizes coping as the variety of coping behaviors rather than as the time given to them or their frequency during the period of distress. On their face, the emotionfocused behaviors in the WOCC seem relatively discreet, so that frequency of particular behaviors may be needed for effectiveness. Folkman and Lazarus (1980) noted that emotion-focused strategies offer only temporary relief. In contrast, many problem-focused behaviors seem to fit together into an integral decision-making process, so that their variety may refle ct engagement in such a process. Still another possibility is that the study's cross-sectional, intersubject design prevented the confirmation of Lazarus's (2000) theory that emotion-focused coping complements problem-focused coping in dealing with stress.

Our study is the first to suggest that perceived control of career-choice context is important in the career indecision and choice anxiety of women in low-level jobs. However, the study's findings must be interpreted cautiously because of its limitations. First, its generalizability may be limited because the participants worked for a single employer and aspired to higher level jobs. Other women in low-level jobs with different employers who have stopped aspiring may differ from the participants. Second, data collected at one point in time that suggest possible sequential and causal relationships require further studies with appropriate temporal sequences to confirm the sequential order among the variables. Third, the suggestions for counseling need to be tested in field studies of career interventions. Fourth, the study examined differences across participants rather than the interplay of coping, perceived control, and choice anxiety within each woman as she endured and dealt with indecision. As such, the findings do not answer the question, "What is happening within an individual?" Fifth, all measures depended on the participants' recall of their perceived control and coping with career indecision during the time of greatest choice anxiety rather than on real-time ratings, and these self-reports were not confirmed by external raters or behavioral measures.

TABLE 1

Summary of Hierarchical Regression Analysis for Variables Relating to
Career Choice Anxiety

Variable                   B    SE B    r     [beta]

Step 1
 Career indecision        0.21  .09   .26 **   .18 *
 Problem-focused coping   0.16  .19   .05      .07
 Perceived control       -2.03  .51   .38 **  -.35 **
Step 2
 Career indecision        0.23  .09            .20 *
 Problem-focused coping   0.06  .19            .03
 Perceived control       -2.18  .51           -.37 **
 Coping x Control        -0.18  .08           -.18 *

Note. [R.sup.2] = .18 for Step 1;

 [DELTA][R.sup.2] = .03 for Step 2 (p < .05).

* p < .05.

** p < .01.

References

Barlow, D. H. (2000). Unraveling the mysteries of anxiety and its disorders from the perspective of emotion theory. American Psychologist, 55, 1247-1265.

Betz, N. E. (1994). Basic issues and concepts. In W. B. Walsh & S. H. Osipow (Eds.), Career counseling for women (pp. 1-41). Hillsdale, NJ: Erlbaum.

Bureau of Labor Statistics. (1998). News: United States Department of Labor statistics (Publication No. 98-253). Washington, DC: Government Printing Office.

Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239.

Fuqua, D. R., & Hartman, B. W. (1983). A behavioral index of career indecision for college students. Journal of College Student Personnel, 24, 507-512.

Fuqua, D. R, Seaworth, T. B., & Newman, J. L. (1987). The relationships of career indecision and anxiety: A multivariate examination. Journal of Vocational Behavior, 30, 177-186.

Harmon, L. W. (1994). Career Decision Scale review. In J. T. Kapes, M. M. Mastie, & E. A. Whitfield (Eds.), A counselor's guide to career assessment instruments (3rd ed., pp. 258-262). Alexandria, VA: National Career Development Association.

Heppner, P. P., Cook, S. W., Strozier, A. L., & Heppner, M. (1991). An investigation of coping styles and gender differences with farmers in career transition. Journal of Counseling Psychology, 38, 167-174.

Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599-610.

Krumboltz, J. D. (1993). Integrating career and personal counseling. The Career Development Quarterly, 42, 143-148.

Lazarus, R. S. (2000). Toward better research on stress and coping. American Psychologist, 55, 655-673.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer.

Lazarus, R. S., & Folkman, S. (1988). Manual for the Ways of Coping Questionnaire. Palo Alto, CA: Consulting Psychologists Press.

Long, B. C. (1998). Coping with workplace stress: A multiple group comparison of female managers and clerical workers. Journal of Counseling Psychology, 45, 65-78.

Luzzo, D. A., James, T., & Luna, M. (1996). Effects of attributional retraining on the career beliefs and career exploration behavior of college students. Journal of Counseling Psychology, 43, 415-422.

McWhirter, E. H., Torres, D., & Rasheed, S. (1998). Assessing barriers to women's career adjustment. Journal of Career Assessment, 6, 365-480.

Meara, N. M., Davis, K. L., & Robinson, B. S. (1997). The working lives of women from lower socioeconomic backgrounds: Assessing prospects, enabling success. Journal of Career Assessment, 5, 115-135.

Mendonca, J. D., & Siess, T. F. (1976). Counseling for indecisiveness: Problem solving and anxiety management training. Journal of Counseling Psychology, 23, 339-347.

Mitchell, L. K., & Krumboltz, J. D. (1987), Effects of cognitive restructuring and decision-making training on career indecision. Journal of Counseling and Development, 66, 171-174.

Nam, C. B., & Powers, M. G. (1983). The socioeconomic approach to status measurement. Houston, TX: Cap and Gown Press.

O'Hare, M. M., & Tamburri, E. (1986). Coping as a moderator of the relation between anxiety and career decision making. Journal of Counseling Psychology, 33, 255-264.

Osipow, S. H., Carney, C. G., Winer, J., Yanico, B., & Koschier, M. (1987). The Career Decision Scale (3rd rev.). Odessa, FL: Psychological Assessment Resources.

Osipow, S. H., & Winer, J. L. (1996). The use of the Career Decision Scale in career assessment. Journal of Career Assessment, 4, 117-130.

Pedhazur, E. J. (1997). Multiple regression in behavioral research: Explanation and prediction (3rd ed.). Fort Worth, TX: Harcourt Brace College Publishers.

Spielberger, C. D., Gorsuch, R., & Lushene, R. (1983). Manual for the State-Trait Anxiety Inventory (2nd ed.). Palo Alto, CA: Consulting Psychologists Press.

Spokane, A. R. (1991). Career intervention. Englewood Cliffs, NJ: Prentice Hall.

Super, D. E. (1990). A life-span, life-space approach to career development. In D. Brown, L. Brooks, & Associates (Eds.), Career choice and development (2nd ed., pp. 197-261). San Francisco: Jossey-Bass.

Taylor, K. M. (1982). An investigation of vocational indecision in college students: Correlates and moderators. Journal of Vocational Behavior, 21, 318-328.

Tiedeman, D. V., & O'Hara, R. P. (1963). Career development: Choice and adjustment. New York: College Entrance Examination Board.

Tracey, T. J. G. (2000). Issues in the analysis and interpretation of quantitative data: Deinstitutionalization of the null hypothesis. In S. D. Brown & R. D. Lent (Eds.), Handbook of counseling psychology (3rd ed., pp. 177-198). New York: Wiley.

Faye M. Weinstein, Keck School of Medicine, University of Southern California; Charles C. Healy and Philip B. Ender, Graduate School of Education and Information Studies, University of California, Los Angeles. Correspondence concerning this article should be addressed to Charles C. Healy, 1028 Moore Hall, UCLA, Los Angeles, CA 90095-1521 (e-mail: [e-mail]).