Self-efficacy, perceptions of barriers, vocational identity, and the career exploration behavior of Latino/a high school students.

The present study examined the ways in which the social cognitive variables of career decision-making self-efficacy and perceptions of barriers may be related to the vocational identity and number of career-related activities undertaken in a sample of Latino/a adolescents. On the basis of the aforementioned literature, it was expected that lower levels of perceived barriers and a greater sense of career decision-making self-efficacy would predict a more integrated vocational identity and increased engagement in career exploration tasks. In addition, because of inconsistent study findings in the literature noted earlier, gender was included in the model to assess any potential gender differences. The present study used both psychological (vocational identity) and behavioral (career search activities) outcome measures to assess the influence of career decision-making self-efficacy and perceptions of barriers within this understudied population.

Method

Participants

Participants were 128 high school students at an urban high school in a large northeastern city in the United States who self-identified as Latino/Hispanic on the demographic questionnaire accompanying this study. Some 66.4% of participants were male adolescents (n = 85), and 33.6% were female adolescents (n = 43). All participants self-identified as Latino/a. The students' ages ranged from 15 to 18 years, with a mean age of 16.12 years (SD = 0.91). Most students indicated that they were in 10th grade (46.9%), followed by 11th grade (44.5%) and 12th grade (7.8%). (A total of 0.8% of the responses for the question about grade were missing.) The majority of students (85.9%) reported that they were born in the United States, followed by a small percentage of students who reported that they were born in the Dominican Republic (4.7%), Mexico (3.1%), various other countries in South and Central America (4.8%), or Puerto Rico (0.8%; percentages do not equal 100% because of rounding). Some 75.3% of the students attending this high school were eligible for participation in the U.S. Department of Agriculture free lunch program. Eligibility is determined by multiplying the year's federal income poverty guidelines by 1.30. The present study had a 95% response rate, suggesting that the sample was representative of the high school's Latino/a population.

Procedure

Packets of instruments were distributed throughout the day by the researchers (the authors of this article and a team of research assistants) to participants in the classrooms of teachers who had agreed to participate. Students were asked to complete the instruments during class time and were informed that their participation was voluntary and that their responses would be kept anonymous. The packets were counterbalanced and included the Career Decision-Making Self-Efficacy Scale-Short Form (CDMSES-SF; Betz, Klein, & Taylor, 1996), the My Vocational Situation (MVS; Holland, Daiger, et al., 1980), the Career Search Activities Index (Solberg et al., 1995), and a demographic questionnaire. The study was approved by the institutional review boards at the local board of education and at our home institution. Parental consent was obtained prior to administration.

Instruments

CDMSES-SF. The CDMSES-SF is a 25-item short form of the original CDMSES developed by Taylor and Betz (1983; see Betz et al., 1996). The scale measures respondents' self-efficacy expectations in relation to career decision-making tasks. Individuals rate their degree of confidence in successfully completing a variety of career-related tasks on a 5-point Likert scale ranging from 1 (no confidence) to 5 (complete confidence). Higher scores indicate higher levels of career decision-making self-efficacy. Validity of the CDMSES-SF has been established such that scores from the CDMSES-SF have been linked to career indecision (Betz et al., 1996). Internal consistencies ranging from .92 to .97 have been reported for the short form (Nilsson, Schmidt, & Meek, 2002). Reliability analyses for the present sample indicated an alpha of .89.

Career Search Activities Index. The Career Search Activities Index is a list of 24 career search activities developed by Solberg et al. (1995). Respondents are asked to indicate whether they had participated in each of the activities within the past 6 months. Items include such activities as "done a mock interview" and "made a list about what you want in a career." The total number of the activities completed is used as the score for this index. Using their sample of 426 college men and women, Solberg et al. reported an internal consistency coefficient of .73 using the Kuder-Richardson (K-R) 20 formula. The internal consistency coefficient for the sample used in this study was .73. Because the index is simply a checklist of behaviors, not a measure of a presumed psychological construct, no validity information is reported. Students either performed or did not perform the actions listed.

MVS. The MVS was designed to identify potential types of difficulties that people encounter in vocational decision making. The present study used subscales that assessed difficulties with vocational identity and perceived environmental or personal barriers.

The Vocational Identity subscale of the MVS consists of 18 true/false items intended to measure the extent to which respondents have a clear picture of their goals, strengths, and interests. Holland, Daiger, et al. (1980) reported an internal consistency of .86, using the K-R 20 formula, for a sample of 185 male high school students and 311 female high school students. The internal consistency coefficient for the sample used in this study was .81. Leong and Morris (1989) found that vocational identity was negatively related to social avoidance, distress, and intolerance of ambiguity and positively related to a high level of career maturity and the tendency to use a rational decision-making style. Holland, Daiger, et al. found substantial evidence for construct validity using a sample of 824 people from high schools, colleges, and businesses. Holland, Daiger, et al. found that the Vocational Identity subscale was positively correlated with age and negatively correlated with the quantity and variety of vocational aspirations that an individual lists on the MVS, suggesting that a more defined vocational identity is related to a smaller number and range of career aspirations. Furthermore, a subsample of participants who scored high on the Vocational Identity subscale was rated to be "well-organized," "self-confident," "not at loose ends," and "competent to handle their lives well" (Holland, Daiger, et al., 1980, p. 4). Finally, a survey of 2,343 high school students found that Vocational Identity scores were negatively correlated with the amount of career and personal help desired by the participants (Holland, Gottfredson, & Power, 1980).

The Barriers subscale of the MVS is composed of four yes/no items and is intended to measure the respondents' perceived external obstacles to occupational goals. The score for the Barriers subscale is the total number of no responses. Higher scores on the Barriers subscale indicated more perceived barriers. Holland, Daiger, et al. (1980) reported an internal consistency coefficient of .23 for both men and women using the K-R 20 formula. Holland, Daiger, et al. argued that the Barriers subscale is similar to a checklist; it does not often function as a homogeneous subscale but is useful in understanding this aspect of students' career development. It is also likely that the low number of items (four) adversely affects its reliability. However, this measure of perceptions of barriers was chosen to accommodate the time constraints and attention span of the sample. The internal consistency coefficient for the present sample was .42.

Results

The data were examined using a multivariate multiple regression. Career decision-making self-efficacy and perceptions of barriers were the predictor variables and vocational identity and career search activities were the criterion variables. Participant gender was dummy coded and included as a predictor in the analysis. Correlations, means, and standard deviations for the variables are noted in Table 1.

No main effect was found for gender, Wilks's [LAMBDA] = .97, F(2, 123) = 2.20, ns. However, both career decision-making self-efficacy, Wilks's [LAMBDA] = .81, F(2, 123) = 14.33, p < .01, [[eta].sup.m] = .19, where [[eta].sup.m] represents the multivariate effect size, and perceptions of barriers, Wilks's [LAMBDA] = .82, F(2, 123) = 13.57, p < .01, [[eta].sup.m] = .18, were found to be significant overall. Because multivariate significance was reached for these two predictor variables, univariate analyses were examined. These analyses indicated that career decision-making self-efficacy had a significant positive relationship with vocational identity, F(1, 127) = 12.08, p < .01, [eta] = .09) where [eta] represents the univariate effect size, and with career search activities, F(1, 127) = 20.40, p < .01, [eta] = .14. Perceptions of barriers was significantly positively correlated with vocational identity, F(1, 127) = 27.22, p < .01, [eta] = .18, but did not have a significant relationship with career search activities in this sample.

Discussion