علاقه مندی، خوداثربخشی و انتخاب اهداف: دستکاری تجربی

An experimental design was used to test the hypothesis that vocational interests can be a precursor to the development of self-efficacy. Participants (n = 180) rated job descriptions for careers in the domains of information technology, sales, and teaching that contained information on activities and work values. Participants rated those job descriptions in terms of their interest, confidence, and choice intentions. An experimental manipulation of information on work values for each of the three work domains was used to create two levels of interests in each area. Participants expressed significantly more interest in job descriptions that contained desirable work value information in each of the three occupational domains relative to the corresponding descriptions that contained information on less desirable values. Obtained results suggest that this manipulation of level of expressed interests in each occupational domain had both a direct effect on self-efficacy ratings and also an indirect effect on self-efficacy through choice goals. These findings have theoretical implications concerning the relation between interests and self-efficacy specified by social cognitive career theory ( Lent, Brown, & Hackett, 1994).

Social cognitive career theory (SCCT; Lent et al., 1994) is an expansion of Bandura, 1982 and Bandura, 1986 social cognitive theory that provides a model for different aspects of vocational behavior. More specifically, SCCT addresses the questions of how career interests develop from self-efficacy, how career related choices are made, and how performance outcomes are achieved. The theory postulates a direct causal link between self-efficacy and interests, where changes in confidence lead to changes in interests. Although there is ample evidence that these two constructs are related (Rottinghaus, Larson, & Borgen, 2003), this correlational research does not address the underlying issue of causality, that is, whether changes in self-efficacy effect changes in interests, whether the reverse is true, or whether the relation is bidirectional. Experimental manipulations of confidence have resulted in changes in interests, providing some support for the SCCT model’s causal links (e.g., Betz and Schifano, 2000, Campbell and Hackett, 1986 and Luzzo et al., 1999). However, alternatives to the SCCT model, including the possibility that changes in interests may produce changes in confidence, have not been investigated. The present study will use an experimental manipulation of interests to evaluate causal links between interests and self-efficacy. 1.1. Key constructs in the SCCT model Career-related self-efficacy, outcome expectations, interests, and choice goals are among the key constructs hypothesized to play a major role in guiding vocational behavior in SCCT. Self-efficacy, which is defined as “people’s judgment of their capabilities to organize and execute courses of action required to attain designated types of performances” (Bandura, 1986, p. 391), is the mechanism emphasized the most in SCCT-based research. Self-efficacy is linked to the perception of having the power to produce effects based on one’s actions. Outcome expectations are defined in SCCT as a person’s beliefs about the probability of response outcomes, or the imagined consequences of a particular behavior. These expectations can be tangible (e.g., monetary rewards), but they can also be intrinsic or self-evaluative (e.g., work-related needs and values). From this perspective, values are seen as preferences for certain reinforcers in the environment that serve as personal standards of behavior a person acquires over time through basic social learning mechanisms. If the environment provides the kinds of reinforcers that are congruent with what a person values most, this leads to self-satisfaction. The anticipated outcome of self-satisfaction in turn is a powerful intrinsic motivating force toward the development of interests and the choice of activities. In the present study the focus was on self-evaluative outcomes in the form of work values. Vocational interests are defined in SCCT as “patterns of likes, dislikes, and indifferences regarding career-relevant activities and occupations” (Lent et al., 1994, p. 88). Although many conceptualizations and determinants of interests have been identified in the literature (see e.g., Silvia, 2006, for an overview), SCCT primarily focuses on the role of learning experiences, other environmental influences, and cognitive factors in the development of vocational interests. The last key SCCT construct relevant to the present research is choice goals, defined as “the determination to engage in a particular activity or to effect a particular future outcome” (Lent et al., 1994, p. 85). Examples of goals are career plans and choice actions. Goals help people to organize and guide behavior, and to sustain it over extended time periods without the need for external reinforcement. 1.2. SCCT link between self-efficacy and interests Much of the empirical research effort to validate the SCCT model has been devoted to the construct of self-efficacy. In particular, the link between self-efficacy and interests has been the focus of numerous studies. Based on a meta-analysis of 60 independent samples, Rottinghaus et al. (2003) concluded that self-efficacy and interests are distinct, albeit correlated constructs. SCCT postulates a direct link between self-efficacy and interests, where changes in self-efficacy lead to changes in interests. SCCT does not, however, explicitly specify a scenario in which changes in interests could directly cause changes in self-efficacy. Nevertheless, the model allows for an indirect causal path between the two constructs: interests in a particular domain lead to approach behavior and the intention to engage in an activity. Repeated engagement can lead to success experiences and goal attainment, which then influences the development of self-efficacy. This hypothesized indirect path is also congruent with theories that view interest as an emotional “approach urge” that promotes the continuous development of knowledge and skills necessary for survival (Silvia, 2006). A number of studies have focused on the experimental manipulation of self-efficacy (e.g., Betz and Schifano, 2000, Campbell and Hackett, 1986, Hackett and Campbell, 1987 and Luzzo et al., 1999). Early studies were based on the manipulation of the level of difficulty of tasks such as the solving of anagrams or incomplete number series (Campbell and Hackett, 1986 and Hackett and Campbell, 1987). As predicted by SCCT, performance success in these studies was associated with higher self-efficacy and task-related interest ratings, while performance failure lowered the participants’ rating of their level of self-efficacy and interest in the task. Another approach to the manipulation of self-efficacy was based on providing mastery and vicarious learning experiences as part of an intervention to raise self-efficacy in a specific academic domain. For example, Betz and Schifano (2000) developed an intervention designed to raise self-efficacy related to building, repair, and construction activities in women; participants randomly assigned to the intervention condition showed higher self-efficacy and an increase in interest in these types of activities compared to the control group. A similar study has been conducted by Luzzo et al. (1999) who investigated the separate and combined effects of performance mastery and vicarious learning on the mathematics and science self-efficacy, interests, and academic choices of a sample of college students. At a follow-up assessment four weeks later, the authors found that the largest increase in self-efficacy and subsequently interests had occurred in the participants in the combined mastery/vicarious learning condition. 1.3. Interests as a self-efficacy precursor Experimental studies have primarily focused on the manipulation of sources of self-efficacy to demonstrate that changes in self-efficacy based on performance attainment can effect changes in interests, hereby supporting the SCCT prediction with regard to a directional link between self-efficacy and interests. However, this support for the SCCT model is somewhat limited because researchers have not explicitly tested alternative models by using manipulations of interests to produce changes in self-efficacy. One challenge is the relative stability of the construct. Interests have been shown to be partially determined by genetics (e.g., Betsworth et al., 1994), they have been shown to correlate with stable personality traits (e.g., Barrick et al., 2003 and Larson et al., 2002), and they tend to fluctuate little over time (Low and Rounds, 2007 and Rottinghaus et al., 2007). Despite the lack of true experimental evidence to support a causal interests—self-efficacy link, recent longitudinal studies found preliminary evidence that vocational interests can lead to the development of self-efficacy (Nauta et al., 2002 and Tracey, 2002). For example, in a study by Tracey (2002), school children in fifth and eighth grade were assessed twice during a one-year period in terms of their interests and confidence in different RIASEC (Realistic, Investigative, Artistic, Social, Enterprising, and Conventional) domains. The results indicated that a reciprocal model with equally strong effects provided the best fit for the data: changes in self-efficacy effected changes in interests, but interests also led to confidence. A similar result was found in a study with college students (Nauta et al., 2002). At three points throughout an academic year, students completed RIASEC measures of confidence and interests. In the first half of the time period, initial interests were predictive of self-efficacy, but initial self-efficacy did not predict subsequent interests. During the second half, the effects of interests and confidence were reciprocal and equally strong. 1.4. Manipulation of interests using work values One potential mechanism for the manipulation of interests is the creation of different levels of interests based on work values. Work values, although not a primary focus of SCCT, are included under the construct of outcome expectations as described above (Diegelman and Subich, 2001, Fouad and Guillen, 2006 and Lent et al., 1994). Diegelman and Subich (2001) provided experimental evidence that vocational interests indeed were found to be a function of outcome expectations. In their study, non-psychology majors showed significantly increased interest and intent in majoring in psychology after participating in a discussion session in which positive outcome expectations with regard to a psychology degree were presented. The study not only showed that the manipulation of outcome expectations can be used to increase interests, but it also presented evidence that outcome expectations are independent from self-efficacy, an important consideration for the present research.

The primary objective of the present study was to investigate the utility of a work values based manipulation of job-related interests, and to assess how this experimental manipulation influences ratings of confidence and occupational choice. Previous research has established relations between SCCT’s key constructs of interests, confidence, outcome expectations, and choice. Most of this body of research, however, is based on correlational studies from which the conclusion of causality cannot be drawn. Of particular interest in the literature has been the link between vocational interests and self-efficacy. The theory specifies that self-efficacy is the precursor to the development of interests. Longitudinal evidence (Nauta et al., 2002 and Tracey, 2002) has indicated, however, that the direction of causality might also be reversed, with interests being the precursor to confidence, but only a study based on a true experimental design can fully answer this question. The present study extends previous research by using an experimental manipulation of vocational interests to assess whether interests can be a precursor to confidence. Three different occupational domains (information technology, sales, and teaching) were included in the study, and level of interest was manipulated by adjusting the desirability of work values embedded in a set of job descriptions. 5.1. Work values and interests Based on the SCCT model and previous experimental evidence (Diegelman & Subich, 2001), it was hypothesized that work values, which are included among the construct of outcome expectancies, would have an influence on the level of job-related interest. The hypothesis was supported across the three occupational domains of information technology, sales, and teaching. Individuals showed significantly more interest in the job positions that were associated with desirable work values as opposed to unappealing work values, accounting for 7–36% of the variance in job-related interest ratings. The success of the attempted manipulation is particularly notable since interests are considered to be a relatively stable construct (e.g., Low and Rounds, 2007, Rottinghaus et al., 2007 and Tracey and Robbins, 2005). However, the present study showed that individual differences in interests can be partly accounted for by the specific context in which these interests are being assessed. This finding is consistent with SCCT, which specifies that outcome expectancies, such as the degree of congruency with one’s work values, serve to refine the broader concept of interests as it applies to a more specific context. For example, an individual might have a general interest in investigative activities, but decides to not pursue this interest in a corporate environment that is incongruent with the individual’s value of independence and resentment of hierarchical structures. Despite having shared variance, the results obtained in the present study indicated that work values and interests are separate constructs that provide incremental information about vocational behavior. Although some theorists do not conceptually distinguish between values and interests (e.g., Macnab & Fitzsimmons, 1987), most authors treat them as more or less separate constructs (e.g., Dawis, 1991 and Savickas, 1999). Despite the scarcity of empirical research on the relation between interests and values, the available evidence supports the notion that interests and values overlap to some extent; the magnitude of the existing correlations is moderate and often domain-specific (e.g., Rottinghaus & Zytowski, 2006). 5.2. Interests, confidence, and choice goals Having assured that the manipulation of interests based on work values was successful, the effect of level of job-related interest on confidence and choice ratings was evaluated. Most importantly, since the present study was a true experiment as characterized by manipulation of an independent variable (job-related interest), random assignment of job descriptions to either the congruent or incongruent value-condition, and the use of a within-subjects design to control for possible confounding factors, conclusions about causality of the relation between the variables could be drawn. SCCT (Lent et al., 1994) specifies that interests, self-efficacy, and choice goals should be positively correlated with each other. Previous empirical research has confirmed these relations (e.g., Rottinghaus et al., 2003), and the results from the present study were consistent with these predictions. The direction of causality has also been specified by SCCT, with self-efficacy being a precursor to interests, which in turn directs an individual’s choice goals. Although SCCT does not explicitly specify the existence of a direct link from interests to confidence, it allows for the development of self-efficacy through an indirect pathway via choice goals: if people are interested in an activity they are more likely to choose to engage in it. Through repeated exposure to interest-relevant activities, individuals gain (or fail to achieve) a sense of mastery, which contributes to their sense of self-efficacy. The results from the present study support this pathway. Participants were significantly more likely to express willingness to pursue a job in the future when their interest ratings for this job were high. Importantly, this pathway of self-efficacy development based on interests via choice goals necessitates a time lag that is sufficient to allow for the buildup of experiences that shape the development of self-efficacy; this is consistent with longitudinal research (Nauta et al., 2002 and Tracey, 2002) that examined changes in interests and confidence over time. In addition to supporting an SCCT-consistent indirect pathway from interests to confidence that develops over time, the results from the present study indicate that the level of interest also has a more direct and immediate effect on confidence appraisal. Specifically, participants expressed significantly more confidence in being able to be successful at a job when their rated interest in the job description was high. This finding runs counter to SCCT, which does not specify a direct interest-confidence link. Due to the absence of a time lag between participants’ ratings of job-related interest and confidence in the present study, there must be a second mechanism that does not necessitate the repeated exposure to the domain in question. One possibility that is consistent with SCCT that might provide an explanation for the immediate effect of interest on confidence is the existence of a cognitive or affective mediating factor since the operationalization of interests in terms of likes or dislikes inherently includes an affective component. Affective states, in turn have been described as one of the four basic mechanisms that determine ones self-efficacy beliefs (Bandura, 1982, Bandura, 1986 and Lent et al., 1994). The idea that there is a cognitive or affective mediator that accounts for the immediate influence of interests on confidence could not be tested in the present study; further research is needed to replicate the effect, and to clarify the exact mechanism by which changes in interests can cause changes in confidence in a direct and immediate way. 5.3. Implications for career assessment and counseling Structured inventories routinely used in career counseling tend to have their main focus on occupational interests, and, to a lesser extent, on confidence. With the exception of the Kuder Career Planning System, work values are often a neglected factor in the assortment of structured career assessment tools. However, the results from the present study suggest that work values are an important influence on a client’s interests and career choice, and should be systematically included in career assessment and the counseling process. Moreover, the results indicate that work values might be a precursor to the development of interests. Hence, it is important not only to determine what an individual’s interests are, but exploration should also include an understanding of how these interests developed. For example, it is important to distinguish whether interest development occurred based on the level of confidence (as specified by SCCT), or whether an individual becomes interested in an activity because of its congruence with the client’s value system. The determination of the mechanism of interest development is important because it directs the clinician towards different areas of exploration (including the client’s worldview, environmental and family influences, life experiences, etc.), which in turn dictates the choice of suitable counseling interventions. A second implication for career counseling relates to how various factors in the client’s life can act as a barrier to choosing a particular career path. It is often the case that people do not pursue a career path despite high levels of interest. One such barrier might be related to a client’s value system. A client might be very interested in a teaching career, but the career might be incongruent with the importance he or she places on status and income. Therefore, values can be used in conjunction with other variables to assess the practicality of different career options. 5.4. Limitations and future directions The design of the current study exhibits many strengths that increase the validity of the results. Among these are the use of a true experimental design characterized by manipulation of an independent variable, random assignment, and control of confounding variables through a within-subjects design. In addition, level of job-related interest was manipulated based on work values, which are thought to be independent from ability. A further strength of the present study was the use of four alternate forms in which the order of the appearance of the 30 job descriptions was varied. Generalizability was enhanced through the inclusion of three independent occupational domains, as well as the use of a total of five job descriptions in each condition. Moreover, the stimuli in the study were job descriptions modeled after actual job profiles from the O*NET occupational database. Nonetheless, there are several limitations that should be addressed in future research. First, there were three specific occupational domains included in the study, which were chosen based on their relative independence from each other in order to maximize the generalizability of the findings. Although the results were consistent across the three domains, future studies should include additional occupational domains in order to cross-validate the findings. Second, the population of interest in this study was college students; due to the fact that the stimuli used in the present study were job descriptions, it would be important to replicate the results with different populations such as working adults. 5.5. Summary and conclusions The present study makes a number of important contributions to the literature on the relations between interests, self-efficacy, and choice goals specified in SCCT. First, the manipulation of vocational interests using experimental methods represents one of the first of this type of study, thereby adding to the correlation-based research on the relations between these constructs. Second, the demonstration that changes in interests lead to changes in self-efficacy and/or choice goals has implications for the revision and expansion of the SCCT framework. The results from the present study indicated that changes in level of interests lead to an immediate effect on confidence, a finding that should be explored in future research. Third, the design enabled the systematic study of the relative importance of pre-existing activity-based interests and work values, and their interaction in relation to career choice behavior. Lastly, the findings might have ramifications for the career counseling process. For example, many current efforts focus on recruiting more women into science, engineering, and other occupations in which women have been traditionally under-represented ( Betz, 2006). These interventions mainly focus on the enhancement of self-efficacy beliefs in these areas ( Betz and Hackett, 1997 and Betz and Schifano, 2000). The results of the present study indicated that interests affect choice goals and confidence. Hence, counseling interventions that focus on the development of interests via a variety of paths could complement current career-related interventions.