Effects of career counseling on French adults: an experimental study.

This research project is focused on theoretical and empirical works analyzing the effects of career counseling. This experimental and longitudinal study has been carried out in France; it compares a sample of individuals having undertaken competence assessment with a control group and deals with self-esteem, self-analysis, self-concept, and the participants' situations. The results correspond to the authors' main hypotheses: The effect size calculated for all the variables (+0.62) is similar to the one observed in the meta-analyses concerning the same topic. The stability of effects has also been demonstrated for 2 out of 3 categories of variables (effect size +0.44) in spite of a slight decrease in the production of self-concept. Finally, treatment intensity seems to have a lesser effect than the one expected, but it strengthens its role in the development of self-analysis. The discussion focuses on the prospects for counseling practice assessment, as well as theoretical directions meriting future development.

**********

The field of career counseling has generated numerous evaluative studies (Spokane, 1991) and reflections on the methods of counseling (Zytowski & Borgen, 1983). Work in this area was essentially directed toward analyzing the consequences of the feedback of counseling methods, particularly interest inventories (Atanasoff & Slaney, 1980; Fribush Cooper, 1976; Hansen, Gurwitz Kozberg, & Goranson, 1994; Hoffman, Spokane, & Magoon, 1981; Kivlighan, Hageseth, Tipton, & McGovern, 1981; Krivatsky & Magoon, 1976; M. J. Miller & Cochran, 1979; Rubinstein, 1978; Zener & Schnuelle, 1976; Zytowski, 1977).

The syntheses suggested by Goodyear (1990) and by Furnham (Furnham, 1989; Furnham & Varian, 1988) allowed us to examine the moderating variables involved in the feedback process--for instance, the variables related to the client or to the client-counselor interaction. Other studies highlighted psychological aspects in communication, such as the "Barnum effect" (Dickson & Kelly, 1985, p. 367). In these different studies, the evaluation criteria used are varied: perception of counselor (Barak & Lacrosse, 1975; Corrigan & Schmidt, 1983), career maturity, self-concept, self-esteem, and acceptance or assimilation of feedback results (Oliver & Spokane, 1988). The choice of these criteria is generally based on one or several theoretical models in vocational development; for instance, Donald Super's (1957) developmental theory is often cited, because it insists on the foreseeable aspect of development, describing the way the individual works in terms of "parts" played by an actor and placing self-concept at the core of the process. A more recent theory by Linda Gottfredson (1981), which introduced new notions, is also referred to when it comes to explaining the working out of vocational preferences and mechanisms producing the decisions in career direction. Spokane (1991) reported that the interventions in career counseling had a modest but durable effect. This effect was more pronounced when the counseling took place over a longer period or when it was individual, and the effect seemed to continue at least 6 months after the process had stopped. In a meta-analysis reviewing 58 studies and dealing with 7,311 participants, Oliver and Spokane (1988) observed that the effect size was equal to 0.48 for all the counseling procedures. The criteria that evolved most seemed to be career maturity, self-knowledge, and decision making. More recently, Whiston, Sexton, and Lasoff (1998), analyzing 47 studies conducted between 1983 and 1995 and involving 4,660 participants, confirmed these results and observed an effect size equal to 0.45.

Although these studies are essential for us to understand better the impact of practices, some limits need mentioning. First, many of these studies deal with samples of students, which do not allow us to evaluate genuine professional situations (problem of generalizibility). Second, these studies seldom focus on the stability of effects several months after the intervention has stopped. Yet, it is definitely over a certain period of time that durable changes may be observed regarding self-concept, vocational identity, or the evolution of career maturity. Last, intercultural aspects are often not considered, leading us to believe that the generalizibility of effects in other cultural contexts is yet to be studied.

The kind of career counseling studied here is called competence assessment (bilan de competences). It enables employees and job seekers to analyze their skills and to build up a career project. Its legal status was defined in France in 1991, the year of its setting up. Provided an employee has worked for a few years, he or she is entitled to a 24-hour leave to undertake the assessment with no loss of salary. The expenses are taken on by either his or her firm or by a mutual benefit society. The competence assessment may only be undertaken with the employee's consent and must take place in an official center (the Centre Interinstitutionnel de Competences, for instance). Finally, the person who takes the assessment is the sole recipient of the results. In 1992, the three steps of competence assessment were defined:

1. an individual preliminary phase, which aims at reinforcing the involvement of the clients in their project, defining and analyzing their needs, acquainting them with procedures and the methods applied;

2. an investigation phase, usually individual, to analyze the client's professional and personal motives; to identify his or her skills, professional and personal abilities, and general knowledge; and to determine different possibilities in career evolution;

3. a final phase, with personalized interviews, during which the results of the investigation are observed, the different factors likely to foster or hamper possible projects are reviewed, and the different steps of the project selected are defined. During this last phase, a written synthesis is made, summarizing circumstances, results, and recommendations.

The study presented here is the first quantitative experiment in France aiming to evaluate the effects of career counseling in adults through competence assessment and involving group comparison. It is based on three hypotheses:

Hypothesis 1: The procedure of career counseling will have an average effect similar for all criteria to what was observed in the American cultural context. The cultural difference between the two countries is rather limited and the procedures used, in career counseling as well as in the effect evaluation, appear very close. Therefore, effect sizes for all the criteria should approximate +0.50.

Hypothesis 2: The effects will continue 6 months after the end of the assessment (effect stability). We believe, like Donald Super, that vocational development is an irreversible process, which allows us to take for granted the progress made during the assessment.

Hypothesis 3: The duration of counseling will have a moderating influence on effects. As Oliver and Spokane (1988) showed, although Whiston et al. (1998) did not confirm it, the effect produced by duration seems to be the more determining criterion to explain why some individuals make more and faster progress than others in enriching self-concept or in establishing career projects. This moderating effect of the time spent in counseling is explained first by the fact that we resort to a greater number of methods in investigating self-concept and because there is a closer examination of the possible choices and better guidance in the decision-making process.

Method

Participants

The participants were clients of two counseling centers for the experimental group and clients of two training centers for the control group.

Experimental group. Only those having completed the whole procedure and taken the pretest and the posttest 6 months later were selected. The 158 participants selected had an average age of 33, with 31% men and 69% women; the majority (66%) were job seekers. Fifty-eight percent of them were graduates or postgraduates. Forty-two percent of the participants directly expressed the need for career counseling, whereas 52% expressed the need via an organization or firm.

Control group. The control group was composed of 80 people involved in professional training and not having undertaken any counseling action in the 2 previous years. Similarly to the experimental group, only questionnaires correctly filled out were selected. The participants' average age was 26, 69% of them were graduates, and there were 89% men and 11% women. The various branches of training ranged from data processing and technical training to administration, trade, and communication.

Instruments

The questionnaires included, in addition to the data above, questions regarding general demographic data such as gender, age, and academic and professional levels. The questionnaire was anonymous, with a code number allowing counselors to match the questionnaires with the various retests.

Self-esteem. The scale used was the French adaptation of Battle's (1992) CFSEI (Culture-Free Self-Esteem Inventory). This scale initially comprises 40 items with yes/no questions. It includes a CFSEI, Adult Form, and three specific aspects: General self-esteem is linked to the overall perception that the individual has of his or her own value (e.g., "Are you lacking in self-confidence?"); personal self-esteem is related to a more intimate perception of his or her value and mental health (e.g., "Do you worry a lot?"); and, finally, social self-esteem is linked to the quality of peer relationships (e.g., "Do you have only a few friends?"). The scale in the American version offers good psychometric properties, and numerous research studies have made use of it (Battle, 1990). In the French adaptation, only 33 items were selected according to a factor analysis of the items (division into main components and oblimin rotation) and an examination of the internal reliability of dimensions. The alpha coefficients on the French sample (N = 328) are good: .84 for the general scale, .81 for the Personal Scale, .64 for the Social Scale and .87 for the Global Scale.

Activity level. This scale was made-to-measure for the experiment to evaluate the participant's activity in information research about a job. The evaluation consists of an 18-item questionnaire and aims to measure individual mobilization in this area. The items concern preliminary steps to the working out of career projects (e.g., "visit a firm") or to professional integration (e.g., "write or rewrite my resume"). The participants had to indicate whether they had taken these steps during the 4 weeks preceding the assessment. The questionnaire score corresponded to the number of steps taken. The coefficient K-R 20 obtained with 238 participants gives a significant value of .81.

Self-analysis. This scale, comprising 22 items measured on a four-level scale, was also designed specifically for the experiment. It is composed of three oblique factorial dimensions. The first factor (8 items), called Thinking About Oneself, evaluates to what extent the participant questions himself or herself (e.g., "I can analyze may own behavior"). The second factor (6 items) measures the participant's impression of being well aware of his or her personal characteristics and abilities (e.g., "I can identify my own abilities"). The third factor deals with the participant's determination in pursuing his or her career project and his or her ability to make a decision (e.g., "I can establish my own career project"). The alpha coefficients (N = 238) are significant for the three factors with respective values of .76, .69, and .73.

Self-concept. The depth of self-concept was measured using five open-ended questions. The participants were asked to indicate freely what in their opinion were their own skills, personality features, vocational interests, individual characteristics useful for their professional project, and individual characteristics that they needed to develop in the future. Following the content analysis, the number of different elements in each question was taken into account.

Use of information concerning feedback and occupational changes. The follow-up 6 months after career counseling included individual interviews (for 93 participants) during which a standardized evaluation chart of 42 items was used. This chart made it possible (a) to evaluate the participant's situation--work, training, or unemployment; (b) to consider the use made of the written synthesis; and (c) for those working (n = 32), to measure the changes in the way their jobs were perceived, in the relationship between work/personal life, and in the satisfaction obtained at work. For the last two categories, the participants were directly interviewed in order to achieve an assessment on a two-mode ordinal scale (satisfied/not satisfied).

Content of competence assessment. Each of the 10 counselors from the two counseling centers completed a form describing precisely the assessment features for each participant: (a) his or her goals--7 modes (e.g., "working out a career project" or "evolving within a firm"; (b) the volume of the assessment--length in hours, period in months, number of individual interviews; and (c) the method used--personality survey or interest survey, "in-tray" exercises, job analysis, and so on.

Procedure

Recruitment of participants. For the experimental group, all the data collected for this study originated from adults having contacted the two counseling centers. In order to increase the external validity, the participants were not preselected; they were all people having started and completed a competence assessment between April 1996 and February 1998. They participated voluntarily and received no financial remuneration. They were asked to participate in a university research project. Few of them refused (less than 1%). For the control group, the participants were also nonpaid volunteers who were requested to complete a questionnaire on two occasions, collectively and during their training hours. Again, very few declined.

Testing phase. All the participants in the experimental group (n = 158) completed the questionnaires twice, and a small sample (n = 74) completed them a third time. The first session (T1) took place directly before the first individual interview, the second session (T2) took place immediately after the last feedback interview, and the third session (T3) 6 months after the assessment was over. During the first two testing sessions, the questions were always presented in the same order: exploration, career maturity, self-esteem, self-concept, and demographic questions (the latter were asked only in the first session). During the third testing session, the exploration questionnaire was irrelevant and therefore not included, and three questions instead of five were asked concerning self-concept to make the procedure quicker.

In the control group, the participants filled in all the questionnaires once in the 1st week of their training period and a second time 3 months later, this interval corresponding to the average duration of competence assessment. The questions followed the same order as for the experimental group. The participants were informed about the purpose of the experiment after the testing sessions.

Counseling homogeneity. In spite of inevitable variations in form, the approaches used in the career counseling process are similar whatever the center or whoever the counselor. As mentioned earlier, competence assessment is standardized by a decree that defines its content. The two centers and all the counselors work in accordance and use the same tests and methods. On average, competence assessment lasts 17 hours, is spread over a period of 3 months, and includes six individual interviews. The methods most used are personal story (100%), personality inventory (73.4%), interest inventory (67.8%), and skill analysis (46.5%).

Results

Data Processing

The statistical analysis of data was conducted in several steps using an SPSS program (Version 8.0) for Windows, and the analysis allowed us to produce different indexes. In the study of intergroup effect, the average results and standard deviations of the measuring scales were calculated for each group separately (experimental and control group) before the first intervention (T1) and 3 months later (T2). To evaluate the effect, two procedures were used. First, we performed a covariance analysis (Plan S <G2>*T2), with which we were able to test the degree of significance in the different intergroups. Covariance analysis offers numerous advantages compared with other inferring methods; it enables the researcher to work directly on the average differences in the posttest rather than on difference scores that are not as reliable. It also produces adjusted means by considering the initial level differences between the two groups. Second, the effect size was calculated for each variable, applying the procedure proposed by Lipsey and Wilson (1993).

ES = [Me-Mc]/[sigma]

Me = Mean of experimental group

Mc = Mean of control group

[sigma] = standard deviation for both groups or for the control group

The advantages of this index are the following: First, it calculates the effect regardless of sample size. The impact of counseling treatment can therefore be analyzed in a more intelligible way than with a level of significance. Second, this index makes it possible to compare the effects between one study and another, particularly when these studies deal with different criteria and/or different sample sizes or when they are conducted in different cultural contexts.

To investigate intragroup stability, variance analysis was used between T2 and T3 in order to determine whether the level differences were significant. We also calculated the effect size between T1 and T3. For the use of information concerning feedback and occupational changes, the data were summarized from the whole set of answers and allowed for a qualitative interpretation. Finally, for the impact of counseling duration, we took into account the length in hours, the total period in weeks, and the number of individual interviews. These parameters were correlated (Pearson's correlation) with the difference scores T2-T1 and T3-T1 so as to determine whether treatment intensity had an influence on the progress made by the clients.

The results in Table 1 show intergroup effects (i.e., an evaluation of the counseling effect, regardless of time effect, and the Hawthorne effect). Here the differences are significant for the four categories of variables and indicate a positive evolution in the experimental group participants, who made more progress than those in the control group. The average effect size for the four variables together equals +0.62, which is slightly above the one observed in Oliver and Spokane's (1988) meta-analysis and that observed by Whiston et al. (1998). The first of our hypotheses thus seems to be verified: The impact of career counseling is similar to that obtained in the American cultural context, which would indicate transcultural validity. As for the categories of variables, a hierarchy appears: Cognitive aspects of self-concept evolved more than emotional aspects or behavior descriptions. The production of self-concept descriptions (+1.19) positively increased, especially for vocational interests (+2.71) and personality features (+1.05). This may be particularly because specific instruments for assessing the components were resorted to frequently. Generally, the experimental group participants produced one to three additional descriptive items to talk about their personal characteristics. Self-analysis also improved significantly (+0.65). The results correspond to one of the main purposes of career counseling, that is, seeing the career project evolve (+0.89). A higher level of self-interest is to be noticed, which improves the explicitness of self-image (+0.72). Although the difference is significant, the effect in activity level is smaller (+0.47), two activities on average allowing the researcher to differentiate between the experimental group and the control group. This may be explained by the fluctuating aspect of these activities, which are limited in time. For instance, they may depend on the various opportunities that arise at different times during the counseling process. Finally, even though the difference in global self-esteem is significant (ES = +0.16), the effect is much more modest in self-esteem subitems, as only general self-esteem contributes to the effect. The relative weakness of these last results is not surprising. In the structure of the self, some aspects are inconstant and others are more stable. Self-esteem is intimately dependent on identity, and its ability to change is subject to certain conditions (Banaji & Prentice, 1994). The more significant evolution of general self-esteem may indicate the recovering of a self-confidence that had been destroyed by vocational questioning or by unemployment. A more limited effect may also derive from an approach more cognitive and analytic than evaluative and affective, which reduces the possible evolution of self-esteem.

The study of intragroup stability (see Table 2) consisted in comparing the experimental group means before the intervention (T1), after the intervention (T2), and 6 months after the intervention (T3). Here we deal in particular with the last phase (T3). With an average effect size equal to +0.44 between T1 and T3, Hypothesis 2 seems partly confirmed. The self-evaluating questionnaires resulted in nonsignificant differences between T2 and T3 in spite of a slight decrease in the means. We may therefore assume a global stability of gains regarding self-esteem and self-analysis. The decrease is more obvious and significant for 2 out of 3 self-concept categories: skills and personality features. In these two categories, there is on average one descriptive item fewer at T3 compared with T2, which is nearly a return to the initial stage (T1). Two hypotheses may be put forward to explain this phenomenon. First, the decrease may be due to a memory loss of some data concerning the self (also observed by Hansen et al., 1994). After the counseling session, the participant has in his or her memory the information items transmitted during the feedback phase. Six months later, the participant is employed again, and these items do not seem as useful to him or her because his or her decision in career direction has generally been made. Second, there may be an effect of weariness and lowered interest when the activity does not have the same importance; taking the same test for the third time, at a point in time when the test has less relevance, can have a demotivating effect and lead the participant to produce a more limited number of descriptive items in order to finish the test quickly.

Because the data shown in Table 3 are essentially qualitative and do not correspond to an initial hypothesis, we will merely perform a descriptive reading. Regarding occupational changes, we found that career counseling did not have a major direct impact on finding a job. Of the 40.9% initially employed people, 34.4% remained so 6 months after the assessment was over. However, these figures conceal various changes. Among people initially employed, an important turnover was noticed, indicating a period of change: 29% were unemployed after the intervention, while 26% were in training. Of the unemployed, 27% found a job by the end of the 6-month period, and 47% were in training. The career counseling process may thus be regarded as the groundwork for vocational training. Thirty-nine percent of all interviewees reached a point where clearly they redefined their career project. In fact, the proportion of inactive people fell from 59% to only 27%; therefore, career counseling seems to have helped the participants to be remotivated and to commit themselves to a social and professional life.

The practical use of the written synthesis was quite moderate. Nearly half (46.5%) of the experimental group participants said they used the synthesis when getting ready for an interview or rewriting their resume. In 1 out of 4 cases (25.8%), the document was disclosed to someone else (e.g., adviser, social worker). The few participants who found a new job and who had already worked before said that their perception of work had not radically changed: 34.3% noticed changes in the relationships with their colleagues, 25% in the relationships with their superiors, whereas 15.6% had a different view of their firm, 34.3% of their job, and 6.2% of their work conditions. A quarter (25%) believed the commitment to their job had increased, whereas a third (34%) said that it was the involvement in their personal lives that had improved (e.g., family, leisure, associations). Finally, fulfillment at work seemed to be the most positive gain: 73% of the participants reported greater satisfaction at work since the career counseling process.

Table 4 shows the correlations between the three indicators of treatment intensity and the difference scores for each category of variable. The results do not corroborate Hypothesis 3 to the extent expected. The coefficient values are generally low and close to 0. No relation is significant regarding self-esteem and activity level. On the other hand, 2 items out of 3 in self-analysis (self-study, explicitness of self-concept) seem to be related to treatment intensity: The participants who benefited from more interviews and a longer period of assessment saw their ability in self-analysis improve. However, the effect seems to be inverted regarding the production of descriptive items in self-concept: The correlations are weak and negative, and they are significant for only two criteria--skills and personality features. In this respect, a longer assessment is likely to have restricted the expression of the numerous aspects of self-concept. At this stage, we can only conjecture about such a result and suggest a few ideas: The clients may have grown tired by the length of the process, and, even though they modified the image they had of themselves, they did not produce a great number of descriptive items. They may have also improved the quality of their reflection to the detriment of the quantity. Thus, the descriptive items may be scarcer but they would correspond more precisely to the clients' vision. The incomplete confirmation of the hypothesis is also due to the type of procedure: Career counseling in the form of competence assessment is, as mentioned in our introduction, highly formalized and standardized. It lasts from 10 to 28 hours, with 2 to 11 interviews. These ranges are much narrower than those taken in the meta-analyses, which include a greater variety of procedures--from a collective 3-hour session to a thorough counseling approach lasting several weeks.

Discussion

This study aimed at evaluating the individual effects of career counseling on employed or unemployed adults and better understanding how people assimilate information concerning their skills and their expectations. The interest of this study lies in the three following points: First, it is a longitudinal study, comprising three time markers and including a control group used as a reference point. Second, the research was based on multiple criteria, taken from theoretical as well as empirical works, and involving quantitative measures as well as qualitative ones, thus making it possible to establish comparisons with North American studies. Finally, the participants were involved in real situations and not artificially created ones, which gives the data some genuineness and ecological validity.

The first hypothesis was widely confirmed. The effects of career counseling appear positive for all categories of variables measured. Competence assessment enables individuals to enhance their self-concept and understand themselves better, and it provides them with a wider range of elements for reflection and decision making regarding their vocational horizon. The impact of the assessment can also be measured by an increased activity and a greater readiness to explore new domains, as well as a higher degree of self-confidence. We may therefore consider the data observed in France as similar to the meta-analytical studies carried out in the United States.

The second hypothesis is partly verified. The steadiness of developmental acquisitions 6 months after the end of the treatment is quite obvious for the two criteria of self-analysis and self-esteem. Nevertheless, the participants seemed to regress to the initial stage as far as self-concept description is concerned. It has been suggested that this lack of homogeneity in the results may be caused by certain behavior manifestations, such as weariness and a decrease in memory recall.

The third hypothesis cannot be regarded as confirmed: The correlations between treatment intensity and the progress made by the experimental participants were very slight. This study has certain limits: The sample sizes, appropriate for T1 and T2, appear insufficient for T3, particularly when we try to analyze the characteristics of subsamples (employed people, for instance). The qualitative measures made with the use of the interview chart at T3 must be considered as exploratory data in so far as we have little information concerning the reliability and the validity of these measures; moreover, the error in measurement is probably higher for this kind of data than it is for a traditional scale. The measures we used do not, in spite of their diversity, include the whole range of vocational dimensions; for instance, we still need to examine how effective the process of decision making is, how the participants' visions of the different occupations evolve, and how they plan the directing of their career. Finally, the control group cannot be strictly compared with the experimental group--in the latter, the participants were more advanced in vocational development, because they had already decided to become involved in training opportunities that fitted their wishes. The treatment effect thus cannot be as visible as we had hoped. However, it would have been unacceptable according to the code of ethics to set up a real control group composed of members who wished to benefit from career counseling but who were being asked to perform a task lacking purpose and meaning.

The critical analysis of this study also allows us to examine future prospects concerning theoretical reflection and empirical research in this area. A first project, using an empirical approach, could focus on the development of Aptitude Treatment Interaction patterns to gain deeper insight into the factors present in the interaction between the counselor's style, the client's style, and the type of method (Bernaud, 1998). Several studies (Kivlighan et al., 1981; Kivlighan & Shapiro, 1987) explored this field, but they seem insufficient to explain all the factors operating in the counseling process. Such a project would promote the development of a real individualization of counseling practices and aim to consolidate their effectiveness and increase the client's well-being and satisfaction. Although a few variables linked to personality were named (Furnham, 1989; Furnham & Varian, 1988; T. R. Miller, 1991), there remain many things to investigate regarding the client's representation and expectation system, the feeling of efficiency on both parts (see Bandura, 1997), or psychosocial factors such as attractiveness, confidence in others, and the beliefs that will help achieve the goals.

Another more theoretical project could question the psychosocial aspects of career counseling by distinguishing, for instance, between the notion of evaluation and the notion of analysis. As was observed, social practices easily confuse the two notions. Yet, if we want the client to be able to analyze his or her characteristics instead of judging them, it is necessary to establish the distinction between the processes. On the one hand, evaluation is related to the ideas of responsibility and arbitration. The mere presence of an observer is indeed likely to give the feeling that the situation is evaluative. On the other hand, one of the key elements that distinguishes analysis from evaluation is the explicit knowledge of the criteria at stake. In career counseling, we try to eliminate the evaluative aspect of the assessment by providing the client with precise criteria allowing him or her to focus his or her attention and to gather information on different aspects of himself or herself. It appears essential, therefore, to think over the foundations of counseling practices in order to facilitate the acquisition of new skills ("knowing how to analyze your own skills"). This purpose can be achieved by questioning the client's position, the counselor's role, and the way the methods are considered (Lemoine, 1998).

As far as the client is concerned, he or she should not be regarded as a "physical" object to evaluate but as an actor expected to react to a collection of data that affects him or her. Thus, we should try to find favorable conditions enabling the client to gather information about himself or herself and avoid making too many mistakes. In this approach, the client becomes an analyst who carries out his or her treatment without the counselor managing, influencing, or imposing. The counselor's role derives from that approach: He or she is no longer an expert producing evaluative categories but someone who plays the part of a guide and offers his or her client theoretical and methodological means. The counseling methods deserve to be challenged in the same way; they should be considered not as sets of measuring instruments independent of the user or the object but as links in the interactive communication between the two parts. Future studies in the field of career counseling, as Savickas and Walsh (1996) thought, will undoubtedly have to rely on a permanent relation between theory and practice.

References

Atanasoff, G. E., & Slancy, R. B. (1980). Three approaches to counselor-free career exploration among college women. Journal of Counseling Psychology, 27, 332-339.

Banaji, M. R., & Prentice, D. A. (1994). The self in social contexts. Annual Review of Psychology, 45, 297-332.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.

Barak, A., & Lacrosse, M. B. (1975). Multidimensional perception of counselor behavior. Journal of Counseling Psychology, 22, 471-476.

Battle, J. (1990). Self-esteem: The new revolution. Edmonton, Alberta, Canada: Battle & Associates.

Battle, J. (1992). Culture-free self-esteem inventories. Austin, TX: Pro-Ed.

Bernaud, J. L. (1998). The moderating effects of the competence assessment: A differential and situational approach. European Review of Applied Psychology, 48, 263-274.

Corrigan, J. D., & Schmidt, L. D. (1983). Development and validation of revisions in the counselor rating form. Journal of Counseling Psychology, 30, 64-75.

Dickson, D. H., & Kelly, I. W. (1985). The "Barnum effect" in personality assessment: A review of the literature. Psychological Reports, 57, 367-382.

Fribush Cooper, J. (1976). Comparative impact of the SCII and the Vocational Card Sort on career salience and career exploration of women. Journal of Counseling Psychology, 23, 348-352.

Furnham, A. (1989). Personality and the acceptance of diagnostic feedback. Personality and Individual Differences, 10, 1121-1133.

Furnham, A., & Varian, C. (1988). Predicting and accepting personality test scores. Personality and Individual Differences, 9, 735-748.

Goodyear, R. K. (1990). Research on the effects of test interpretation: A review. The Counseling Psychologist, 18, 240-257.

Gottfredson, L. S. (1981). Circumscription and compromise: A developmental theory of occupational aspirations. Journal of Counseling Psychology, 28, 545-579.

Hansen, J-I. C., Gurwitz Kozberg, J., & Goranson, D. (1994). Accuracy of student recall of Strong Interest Inventory results one year after interpretation. Measurement and Evaluation in Counseling and Development, 26, 235-242.

Hoffman, M. A., Spokane, A. R., & Magoon, T. M. (1981). Effects of feedback mode on counseling outcomes using the Strong-Campbell Interest Inventory: Does the counselor really matter? Journal of Counseling Psychology, 28, 119-125.

Kivlighan, D. M., Hageseth, J. A., Tipton, R. M., & McGovern, T. V. (1981). Effect of matching treatment approaches and personality types in group vocational counseling. Journal of Counseling Psychology, 26, 315-320.

Kivlighan, D. M., & Shapiro, R. M. (1987). Holland type as a predictor of benefit from self-help career counseling. Journal of Counseling Psychology, 34, 326-329.

Krivatsky, S. E., & Magoon, T. M. (1976). Differential effects of three vocational counseling treatments. Journal of Counseling Psychology, 23, 332-339.

Lemoine, C. (1998). What is a "bilan de competences"? European Review of Applied Psychology, 48, 243-248.

Lipsey, M. W., & Wilson, D. B. (1993). The efficacy of psychological, educational and behavioral treatment: Confirmation from meta-analysis. American Psychologist, 48, 1181-1209.

Miller, M. J., & Cochran, J. R. (1979). Comparison of the effectiveness of four methods of reporting interest inventory results. Journal of Counseling Psychology, 26, 263-266.

Miller, T. R. (1991). The psychotherapeutic utility of the five-factor model of personality: A clinician's experience. Journal of Personality Assessment, 57, 415-433.

Oliver, L. W., & Spokane, A. R. (1988). Career-intervention outcome: What contributes to client gain? Journal of Counseling Psychology, 35, 447-462.

Rubinstein, M. R. (1978). Integrative interpretation of vocational interest inventory results. Journal of Counseling Psychology, 25, 306-309.

Savickas, M. L., & Walsh, W. B. (1996). Handbook of career counseling, theory and practice. Palo Alto, CA: Davies-Black.

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

Super, D. E. (1957). The psychology of career: An introduction to vocational development. New York: Harper & Row.

Whiston, S. C., Sexton, T. L., & Lasoff, D. L. (1998). Career-intervention outcome: A replication and extension of Oliver and Spokane (1988). Journal of Counseling Psychology, 45, 150-165.

Zener, T. B., & Schnuelle, L. (1976). Effects of the Self-Directed Search on high school students. Journal of Counseling Psychology, 23, 353-359.

Zytowski, D. G. (1977). The effects of being interest-inventoried. Journal of Vocational Behavior, 11, 153-157.

Zytowski, D. G., & Borgen, F. H. (1983). Assessment. In W. B. Walsh & S. H. Osipow (Eds.), Handbook of vocational psychology (Vol. 2). Hillsdale, NJ: Erlbaum.

Editor's note: This article is an English adaptation of the original French version that appeared in the journal l'Orientation Scolaire et Professionnelle. The editor of that journal, Jean Guichard, and the authors kindly provided permission to The Career Development Quarterly to publish this important contribution to the career development literature. There are many excellent articles in our field that are published in languages other than English, and Jean Guichard and I are, in a spirit of international collaboration, endeavoring to bridge this gap. This article is the initial foray into such a collaboration. This article is particularly important because it addresses the effects of career counseling on a sample of French adults. For those who wish to access the original article, it was published in 2001, volume 30, issue 4, pages 485-510 of l'Orientation Scolaire et Professionnelle. Many thanks to Jean Guichard and Jean-Luc Bernaud (who adapted the article to English) for their support and help in this project.

Jean-Luc Bernaud, Department of Psychology, University of Rouen, France; Jean-Philippe Gaudron, Department of Psychology, University of Toulouse, France; Claude Lemoine, Department of Psychology, University of Lille 3, France. Correspondence concerning this article should be addressed to Jean-Luc Bernaud, Departement de Psychologie, Universite de Rouen, Rue Lavoisier, 76821 Mont Saint Aignan Cedex, France (e-mail: [e-mail]).

TABLE 1 Intergroup Effects (Hypothesis 1)

                          Adjusted Means
                        Control   Experimental
                        Group     Group
Variable                (n = 80)  (n = 158)     F      p       ES

Self-esteem                                                    (+ 0.16)
  General               38.27     40.62         12.37  <.001    + 0.34
  Personal              15.02     14.85          0.16  ns       - 0.03
  Social                19.66     20.10          2.71  ns       + 0.17
  Total                 73.00     75.54          5.77  <.02     + 0.23
Activity level           6.48      8.52         14.15  <.0001   + 0.47
Self-analysis                                                  (+ 0.65)
  Thinking about        22.97     24.29          8.77  <.01     + 0.33
    oneself
  Self-knowledge        18.52     20.20         29.53  <.0001   + 0.72
  Career maturity       24.34     27.32         51.12  <.0001   + 0.89
Self-concept                                                   (+ 1.19)
  Skills                 2.81      4.52         32.02  <.0001   + 0.73
  Personality features   2.37      4.62         60.11  <.0001   + 1.05
  Vocational interests   1.55      2.72         25.87  <.0001   + 2.71
  Characteristics/       2.92      5.81         33.53  <.0001   + 0.78
    project
  Characteristics/to     0.82      1.61         23.39  <.0001   + 0.68
    develop

Note. ES = effect size. Values in parentheses indicate the ES mean for
each variable.

TABLE 2 Intragroup Stability (Hypothesis 2)

                                    Mean                        p
                          T1         T2         T3
Variable                  (n = 158)  (n = 158)  (n = 74)  T1/T2   T2/T3

Self-esteem
  General                 37.80      40.49      40.10     <.0001  ns
  Personal                13.63      14.58      14.19     <.001   ns
  Social                  19.77      20.20      20.47     <.02    ns
  Total                   71.20      75.27      74.77     <.0001  ns
Self-analysis
  Thinking about oneself  22.88      24.40      24.50     <.0001  ns
  Self-knowledge          16.32      19.99      19.57     <.0001  ns
  Career maturity         22.62      27.04      26.24     <.0001  ns
Self-concept
  Skills                   3.51       4.67       3.58     <.0001  <.01
  Personality features     3.03       4.68       3.58     <.0001  <.0001
  Vocational interests     2.00       2.73       2.48     <.0001  ns

                                ES

Variable                  T1/T2     T1/T3

Self-esteem               (+ 0.25)  (+ 0.25)
  General                  + 0.36    + 0.33
  Personal                 + 0.20    + 0.12
  Social                   + 0.18    + 0.31
  Total                    + 0.34    + 0.26
Self-analysis             (+ 1.06)  (+ 0.88)
  Thinking about oneself   + 0.40    + 0.40
  Self-knowledge           + 1.45    + 1.24
  Career maturity          + 1.32    + 1.01
Self-concept              (+ 0.54)  (+ 0.19)
  Skills                   + 0.48    + 0.03
  Personality features     + 0.71    + 0.24
  Vocational interests     + 0.42    + 0.29

Note. ES = effect size; T1 = Time 1; T2 = Time 2; T3 = Time 3. Values in
parentheses indicate the ES mean for each variable.

TABLE 3 Use of Information Concerning Feedback and Occupational Changes

                                Initial Situation
                          Employed  Unemployed  Total
Situation After 6 Months  n   %     n   %       n   %

Employed                  17        15          32  34.4
Vocational training       10        26          36  38.7
Unemployed                11        14          25  26.8
Total                     38  40.9  55  59.1    93  99.9

TABLE 4 Impact of Assessment Duration (Hypothesis 3)

                         T2-T1 Difference          T3-T1 Difference
                    PL       PL (Whole        PL       PL (Whole
Variable            (Hours)  days)      NI    (Hours)  days)      NI

General self-        .03     -.05        .00   .15      .14        .06
  esteem
Personal self-       .02     -.12        .02  -.03      .04       -.05
  esteem
Social self-esteem   .11     -.03        .08   .15      .14       -.13
Total self-esteem    .06     -.09        .02   .12      .15       -.10
Activity level       .05      .04        .06
Thinking about       .08      .17*       .06  -.01      .23        .02
  oneself
Self-knowledge       .18*     .24**      .06   .22      .32**      .10
Career maturity      .09      .11       -.03   .00      .09       -.14
Skills              -.13     -.18*      -.11  -.15     -.01       -.04
Personality         -.03     -.17       -.07  -.31**   -.17       -.24*
  features
Vocational          -.03     -.09       -.09  -.10     -.14       -.06
  interests
Characteristics/     .07     -.04        .00
  project
Characteristics/to  -.05      .09        .09
  develop

Note. T1 = Time 1; T2 = Time 2; T3 = Time 3; PL = Process length; NI =
Number of interviews. Boldfaced values indicate significance.
*p < .05. **p < .01.