Family factors associated with sixth-grade adolescents' math and science career interests.

This study tested a causal model, based on social cognitive career theory (SCCT; R. W. Lent, S. D. Brown, & G. Hackett, 1994, 2000), of math and science career interests among 6th-grade adolescents (N = 318). Consistent with SCCT, it was found that career gender-typing, mother's and father's support for pursuing math and science careers, as well as the structure of the family itself (2-parent intact vs. single parent) predicted young adolescents' math self-efficacy; career gender-typing and mother's support predicted math outcome expectations; and math self-efficacy and outcome expectations predicted math and science career interests. Counselor recommendations based on these findings are included.

**********

Investigations of factors that promote mathematics achievement and interest in math and science careers were precipitated by the recognition that mathematics acts as a "critical filter" affecting entry into a wide range of scientific and technical careers (Betz & Hackett, 1983). Extensive research has demonstrated that mathematics self-efficacy, outcome expectations, and gender role socialization are highly relevant to a variety of math- and science-related pursuits. For example, math self-efficacy has been found to be predictive of mathematics performance (Lent, Lopez, & Bieschke, 1993), achievement (Randhawa, Beamer, & Lundberg, 1993; Siegel, Galassi, & Ware, 1985), mathematics-related career choices (Hackett & Betz, 1989; Lent, Lopez, & Bieschke, 1991), and the consideration of math and science careers (Post, Stewart, & Smith, 1991). Mathematics self-efficacy and outcome expectations have been shown to predict subject matter interests in math and science courses (Lopez, Lent, Brown, & Gore, 1997), and mathematics self-efficacy, math outcome expectations, and gender role socialization have been shown to predict math career interests, perceived career options, and academic and vocational choices among college students (Lapan, Shaughnessy, & Boggs, 1996; Lent et al., 1993), high school students (Lauver & Jones, 1991; Lopez & Lent, 1992; O'Brien, Martinez-Pons, & Kopala, 1999), minority high school equivalency students (Church, Teresa, Rosebrook, & Szendre, 1992), and middle school students (Fouad & Smith, 1996; Lapan, Hinkelman, Adams, & Turner, 1999; Lapan & Jingeleski, 1992).

Several studies within the rubric of social cognitive career theory (SCCT; Lent, Brown, & Hackett, 1994, 2000) have supported the relationships that SCCT predicts among math efficacy, outcome expectations, interests, and other associated variables in samples composed of both college (Gainor & Lent, 1998) and middle school students (Fouad & Smith, 1996). SCCT hypothesizes that both contextual and person factors are mediated by math self-efficacy and outcome expectations as they predict math and science career interests. Person factors are biological attributes (e.g., gender, race-ethnicity) that, because of their socially constructed meanings, result in the internalization and expression of differential socialization processes (e.g., the gender-typing of math and science careers as more appropriate for men than for women). Contextual factors are those environmental barriers and supports (e.g., family structure, mother's and father's support for pursuing math and science careers) that affect adolescents' learning experiences through which their career-relevant efficacy and outcome expectations develop. According to SCCT, contextual and person factors, and the variables they predict, influence each other in complex, reciprocal ways. Further, SCCT provides a robust framework for studying and modifying potential support and barrier effects.

One of the contextual factors found to be uniquely relevant to the development of adolescents' career interests is parent support (Ferry, Fouad, & Smith, 2000; Lapan et al., 1999). For example, research has shown that perceived support from fathers relates to the educational plans and career expectations of Mexican American high school girls (McWhirter, Hackett, & Bandalos, 1998); and parental encouragement has significant direct effects on self-efficacy, outcome expectations, and math and science career interests among middle school adolescents (Ferry et al., 2000).

Research has also suggested that children from divorced families may need even more support to succeed in their educational and career endeavors. For example, researchers have shown that family divorces occurring during a child's high school years have significant detrimental effects on the child's graduation from high school and entry into post-high school training (Sandefur, McLanahan, & Wojtkiewicz, 1992); and children from single parent homes express significantly less interest in any career than children from intact families (Trice, Hughes, Odom, Woods, & McClellan, 1995). Studying the effects of family structure on adolescents' math and science career interests becomes even more important in light of the rising divorce rate in the United States. For example, the U.S. Census Bureau (1995) reported that approximately 27% of American households with children under 18 are headed by a single parent. This is an increase of 11% in the rate of single-parent families in the United States since 1970 and equates to nearly 25 million adults and children who live in single-parent family situations.

Therefore, the purpose of this study was to investigate the interactional effects of contextual factors (mother's support, father's support, family structure in terms of single-parent or two-parent intact families) and person inputs (the gender-typing of math and science careers) on the math self-efficacy, math outcome expectations, and math and science career interests of young adolescents. We tested a causal model of these interests for sixth-grade adolescents as they responded to the effects of these person and contextual factors. We hypothesized that these relationships could be successfully modeled according to SCCT.

Method

Participants

Participants were 318 sixth graders. Of these participants, 73.6% (n = 234) were Caucasian Americans, 20.5% (n = 65) were African Americans, 2.8% (n = 9) were Asian Americans, 0.9% (n = 3) were Hispanics, and 2.2% (n = 7) self-identified as "Other." All participants were recruited by their math teachers from seven different sixth-grade math classes in two integrated public schools in one midwestern state. Both schools were in middle-class neighborhoods in communities of approximately 100,000 people. All math teachers used standard sixth-grade math curriculum. Of the participants, 77.4% (n = 246) lived with their biological parents in two-parent intact families, and 22.6% (n = 72) lived with one of their parents in single-parent families. Only two of the single-parent families were headed by a single father.

Data collection was done during math class by the first and third authors and one research assistant during the last 2 months of the school term (April and May) in 1 school year. Stability data were collected during the previous November of the same school year. All research materials were completed in the classroom by all participants. No research incentives were provided. To ascertain if the participants' scores between the two schools were commensurate, we conducted a one-way analysis of variance (ANOVA) with each of the test variables. Our results confirmed that there were no significant differences in participants' scores between the two schools: Interests, F = 3.64, p < .06; Efficacy, F = .02, p < .89; Outcome Expectations, F = 1.66, p < .21; Gender-Typing, F = 3.39, p < .07; Father Support, F = .17, p < .61; Mother Support, F = .07, p < .81; and Family Structure, F = .00, p < .98.

Measures

Two instruments were used in this study, the Science scale of the Revised Unisex American College Testing Interest Inventory (UNIACT; Lamb & Prediger, 1995) and an abbreviated version of the Fennema-Sherman Mathematics Attitude Scales (FSMA; Fennema & Sherman, 1976), which contains items from five of the nine original scales. Family structure was a status variable with two-parent intact families coded 3, and single-parent families coded 1. All other variables were created by taking the sum of scores for each scale divided by the number of items in that scale (mean of each scale) for each participant.

UNIACT. Math and science career interests were measured using the Science scale of the UNIACT. This scale corresponds to Holland's Investigative Occupational theme. Job clusters associated with this theme are interests in engineering and other applied technologies, medical specialties and technologies, natural sciences and mathematics, and social sciences, such as marketing research and anthropology (Lamb & Prediger, 1995). The Science scale consists of 15 items that participants are asked to rate on a 3-point Likert scale (2 = like, 1 = indifferent, 0 = dislike). Sample items for this scale are the following: "Explore a science museum," and "Solve math puzzles." For this scale, M = .98, SD = .53, [alpha] = .90, skewness = .1, and kurtosis = -.8. The internal consistency estimate in the original norming sample of sixth-grade students (n = 1,732) was [alpha] = .91 (Lamb & Prediger, 1995).

FSMA. The original Fennema-Sherman Mathematics Attitude Scales consisted of nine separate 12-item scales measuring nine different math attitudes. These scales, which have been used in over 35 research studies, have been adapted previously for special populations by reducing the number of items used to measure each construct. In each instance, the abridged FSMA scales were found to have factor structures comparable to the original scales, as well as strong internal consistency estimates (Betz & Hackett, 1983; Mulhern & Rae, 1998; Vezeau, Chouinard, Bouffard, & Couture, 1998).

For the purposes of this present study, given the age of the students and the need to keep the research questionnaire brief, five FSMA scales that bore directly on our research questions were chosen. In addition, we further reduced the length of these scales by eliminating items with the lowest factor loadings in the original norming group (Fennema & Sherman, 1976). All FSMA items were scored on 5-point Likert scales with positively worded items scored from 5 (strongly agree) to 1 (strongly disagree) and negatively worded items scored from 5 (strongly disagree) to 1 (strongly agree).

Psychometric properties of the abbreviated FSMA were assessed in our sample using a principal components analysis with a varimax rotation. As predicted, five factors with eigenvalues greater than 1.00 were found (total variance explained = 67.77). Each item of the FSMA that was used for this study loaded more highly on its predicted factors than on any other factors, with factor loadings ranging from .44 to .82 for the Math Confidence Scale, from .59 to .86 for the Usefulness of Mathematics Scale, from .64 to .75 for the Mother Scale, from .61 to .82 for the Father Scale, and from .65 to .81 for the Mathematics as a Male Domain Scale. These scales are further described below.

Perceived mother support. Perceived mother support was measured using the abbreviated FSMA Mother Scale. The scale consists of four items (three scored positively and one scored negatively). Sample items are the following: "My mother has strongly encouraged me to do well in mathematics" (scored positively) and "My mother wouldn't encourage me to plan a career which involves math" (scored negatively). For this scale, M = 4.18, SD = .72, [alpha] = .75, skewness = -.1, and kurtosis = 1.3.

Perceived father support. Perceived father support was measured using the abbreviated FSMA Father Scale. This scale consists of four items (three scored positively and one scored negatively). Sample items are the following: "My father has always been interested in my progress in mathematics" (scored positively) and "My father wouldn't encourage me to plan a career which involves math" (scored negatively). For this scale, M = 4.05, SD = .89, [alpha] = .83, skewness = -1.1, and kurtosis = 1.2.

Career gender-typing. Career gender-typing was measured using the abbreviated FSMA Mathematics as a Male Domain Scale. This scale consists of three items, all scored negatively. Sample items for girls are the following: "Girls can do just as well as boys in mathematics" and "Studying math is just as good for girls as for boys." Sample items for boys are the following: "Boys can do just as well as girls in mathematics" and "Studying math is just as good for boys as for girls." For this scale, M = 1.79, SD = .89, [alpha] = .72, skewness = 1.1, and kurtosis = .9.

Math self-efficacy. Math self-efficacy, defined as "beliefs in one's capabilities to organize and execute a course of action necessary to produce a given attainment" (Bandura, 1997, p. 3), was measured using the abbreviated FSMA Math Confidence Scale. This scale consists of eight items. Five items are scored positively, and three are scored negatively. Sample items from this scale are the following: "I am sure that I can learn mathematics" (scored positively) and "I don't think I could do advanced mathematics" (scored negatively). For this scale, M = 3.85, SD = .70, [alpha] = .88, skewness = -.9, and kurtosis = .9

Math outcome expectations. Math outcome expectations, defined as the expectations that math will be important to one's future career, were measured using the abbreviated FSMA Usefulness of Mathematics Scale. The scale consists of four items. Three items are scored positively, and one is scored negatively. Sample items from this scale are the following: "Knowing mathematics will help me earn a living" (scored positively) and "In terms of my adult life, it is not important for me to do well in mathematics in high school" (scored negatively). For this scale, M = 3.92, SD = .88, [alpha] = .86, skewness = -.9, and kurtosis = .5.

Split-half reliability estimates in the original norming sample ranged from .86 to .93 on these five scales (Fennema & Sherman, 1976). In our sample, Cronbach's alpha reliabilities ranged from .72 to .88. Six-month stability coefficients, which were estimated for a subset of the research sample (n = 278), ranged from high to moderate (Math Confidence Scale, r = .69; Father Scale, r = .57; Mother Scale, r = .50; Usefulness of Mathematics Scale, r = .46; Mathematics as a Male Domain Scale, r = .38). We judged each of these sample statistics to be within acceptable limits, and so proceeded to test our research questions.

Results

Structural equation modeling (SEM), using the maximum likelihood method of parameter estimation, was used to explore the hypothetical relationships among the variables. One of the multivariate assumptions underlying SEM is the use of continuous, normally distributed data. However, recent studies have explored the use of categorical data in the SEM procedure. For example, Muthen and Kaplan (1985) examined the performance of a categorical variable model estimator compared with estimators for continuous variables and found that the categorical model yielded a slight underestimation of chi-square but that the parameter estimates and sampling variability were well in line with expected values. Therefore, although one of our model parameters was measured using a categorical variable (family structure), we judged that the SEM procedure would yield adequate parameter estimates of our hypothesized model.

We used a confirmatory analysis approach and therefore imposed an a priori structure on the model that is congruent with the postulates of SCCT. We left all degrees of freedom unbounded. A raw data matrix was used to estimate our observed exogenous indicators of Perceived Mother Support, Perceived Father Support, Family Structure, and Career Gender-Typing and observed endogenous factors of Math Self-Efficacy, Math Outcome Expectations, and Math and Science Career Interests. Three unobserved exogenous indicators, which were associated with error variance, were also included in the model.

To assess model fit, we applied two goodness-of-fit indices. The chi-square likelihood ratio statistic, which is a measure of the overall fit of

the specified model to the data (Joreskog & Sorbom, 1993), yielded an index of 1.57 ([chi square]/df = 6.27/4, p < .181). If this ratio is less than 2, the model is considered a good fit. Because the chi-square statistic is sensitive to sample size, a second goodness-of-fit test, the root mean square error of approximation (RMSEA = [square root of (F 0/d)]; Steiger & Lind, 1980) was used to verify the original results. An RMSEA value of [less than or equal to] .05 indicates a close fit of the data to the model in relation to the degrees of freedom (Browne & Cudeck, 1992). The RMSEA for this sample yielded an index of .04. To control for data dependency, two bootstrap tests were performed using two independently drawn random samples (n = 200) from the full data set. Differences in standard errors of measure for all model paths in the two bootstrap samples ranged from .002 to .004. Covariances, path regression weights, and bootstrap estimates for the variables in this study are found in Table 1. The derived model, named the Family and Perceived Parent Support model, with beta weights, is shown in Figure 1.

Contextual factors (Family Structure, Perceived Father Support, Perceived Mother Support). In our constructed Family and Perceived Parent Support model, there was a significant negative relationship between Family Structure and Career Gender-Typing (-.11, p < .023), indicating that children from intact families had stronger perceptions that math and science careers were appropriate pursuits for persons of their own gender than did children from single-parent families. Family Structure covaried significantly and positively with Perceived Father Support (.19, p < .007) and Perceived Mother Support (.10, p < .05), indicating that children from intact families perceived significantly more reinforcement from their fathers and mothers to pursue mathematics education and math and science careers than did children from single-parent families. The significant positive path from Family Structure to Math Self-Efficacy ([beta] = .18, p < .007) also suggests that the structure of the family itself significantly affects young adolescents' math self-efficacy, with children from two-parent intact families experiencing greater positive family structure effects on their math efficacy than children from single-parent families. The path between Family Structure and Math Outcome Expectations was not significant ([beta] = -.08, p < .179).

[FIGURE 1 OMITTED]

Perceived Mother Support and Perceived Father Support covaried both positively and significantly (.41, p < .007), indicating that there were significant positive interactional effects between Perceived Father Support and Perceived Mother Support as they predicted the endogenous variables in the model. There were also significant positive paths between Perceived Father Support and Math Efficacy ([beta] = .19, p < .036), Perceived Mother Support and Math Efficacy ([beta] = .21, p < .009), and Perceived Mother Support and Math Outcome Expectations ([beta] = .23, p < .010). The significance of these covariances and the magnitude of the paths between each of these sets of variables suggest that there is both direct and indirect correspondence between children's math efficacy and outcome expectations and their perceptions of parental support for pursuing both mathematics education and math and science careers. The path between Perceived Father Support and Math Outcome Expectations was not significant ([beta] = -.02, p < .753).

Person factor (Career Gender-Typing). There were significant negative covariances between Career Gender-Typing and Perceived Mother Support (-.15, p < .007) and between Career Gender-Typing and Perceived Father Support (-.21, p < .004), suggesting that greater father and mother support are associated with the perception that math and science careers are appropriate for persons of "my" gender. The significant paths between Career Gender-Typing and Math Self-Efficacy ([beta] = -.14, p < .017) and between Career Gender-Typing and Math Outcome Expectations ([beta] = -.15, p < .009) suggest that children who perceive that math and science careers are appropriate gender-equitable pursuits have greater math self-efficacy and greater expectations that math will be useful and valuable to them in their future careers.

Endogenous variables (Math Self-Efficacy, Math Outcome Expectations, and Math and Science Career Interests). In the Family and Perceived Parent Support model, the causal links between Math Efficacy and Math Outcome Expectations ([beta] = .35, p < .007), between Math Efficacy and Math and Science Career Interests ([beta] = .24, p < .013), and between Math Outcome Expectations and Math and Science Career Interests ([beta] = .18, p < .023) support Lent et al.'s (1994, 2000) propositions about the effect and direction of relationships among these variables. Taken together, the Gender-Typing of math and science careers, Perceived Father Support, Perceived Mother Support, and Family Structure accounted for a substantial proportion (36%) of the total unique variance in Math Self-Efficacy. Perceived Mother Support, the Gender-Typing of math and science careers, and Math Efficacy together accounted for 37% of the total unique variance in Math Outcome Expectations. Math Efficacy and Math Outcome Expectations accounted for 21% of the total unique variance in interests in math and science careers. These percentages were generated by computing the multiple correlation coefficients ([R.sup.2]) for the combined effects of all significant exogenous variables on the endogenous variables in the model.

Discussion

The purpose of this study was to investigate the interactional effects of contextual factors (mother's support, father's support, and family structure) and person inputs (the gender-typing of math and science careers) on the math self-efficacy, math outcome expectations, and math and science career interests of young adolescents. The results of this study indicate that adolescents' math self-efficacy positively affects their math outcome expectations and that math efficacy and math outcome expectations together affect their math and science career interests. These results are similar to the results described by Fouad and Smith (1996), except that the relative magnitude of the path between math efficacy and math outcome expectations was greater in the Fouad and Smith study (.55) than in the current study (.35).

Our results also showed that both mother's support and father's support positively affect adolescents' math self-efficacy and that children from single-parent families report significantly less math self-efficacy than children from two-parent intact families. However, the paths between mother support and math efficacy, between father support and math efficacy, and between family structure and math efficacy are similar in magnitude, suggesting that mother and father support can moderate the effects of living in a single-parent family on the math efficacy of adolescents.

Although both mother's support and father's support positively affected adolescents' math efficacy in this study, only mother's support affected their math outcome expectations. This finding suggests that the support and encouragement of mothers to explore math careers, as well as their help in connecting math and science courses to later career possibilities, is especially important in assisting adolescents to develop expectations that math can be important to their future careers. In addition, the results of our study show that adolescents who gender-type math as an appropriate pursuit for persons of their own gender have greater math efficacy and math outcome expectations than adolescents who do not and that mother's support and father's support are significant in developing these gender-typing perceptions.

The results of this study suggest that the social-cognitive career model is a useful framework for conceptualizing the influences of parents' support, family structure, and math career gender-typing on the math efficacy, math outcome expectations, and math and science career interests of adolescents. Fouad, Smith, and Zao (2002) have suggested that "one of the strengths of the social-cognitive career model is that it is directly linked to interventions" (p. 170). Based on the findings of our study, we suggest a number of interventions for counselors to use to assist parents in supporting the math efficacy, math outcome expectations, and thus the math and science career interests of their children.

Counselors can help parents increase their adolescents' mastery of math-related skills by showing them how to help their adolescents choose and engage in educational activities related to math (e.g., advanced math courses, math summer camps). Counselors can help parents model math-related activities by showing their adolescents how they use math both in their careers and in their personal lives (e.g., how they use computers to do inventory, how they balance their checkbooks). Counselors can help parents manage their children's math-related anxiety by showing parents how to acknowledge this anxiety and assure their adolescents that these negative emotions are normal. Counselors can show parents how to encourage their adolescents when they successfully accomplish math-related tasks (e.g., offering concrete praise for their adolescents' accomplishments, giving their adolescents tangible rewards). Counselors can show parents how to encourage their adolescents' persistence in math when they encounter negative or gender-biased stereotyping (e.g., by assuring their adolescents that math is valuable and important for both men and women).

Counselors can also help parents increase their own efficacy to support their adolescents' math efficacy and outcome expectations (Bandura, Barbaranelli, Caprara, & Pastorelli, 2001). For example, counselors can help parents experience success in setting, implementing, evaluating, and adjusting explicit support goals (e.g., setting specific homework times for studying math with their children, engaging a math tutor when needed). Counselors also can assist parents in managing their own math anxiety as it relates to their children's math and science career development (e.g., helping parents find alternative ways to support their children's math career development if their own math skills or experiences are limited). Finally, counselors can encourage divorced parents to collaborate in the support of their adolescents' consideration of math- and science-based careers. For families in which one parent is not available, counselors can work with the custodial parent to find other adult female and male role models, such as grandmothers, grandfathers, teachers, or trusted friends, who can positively influence their middle school students' consideration of and preparation for math and science careers.

Limitations and Suggestions for Future Research

One limitation of this study is that we used a mixed (categorical vs. continuous) variable model, which violates one of the underlying assumptions of SEM. Although recent research has demonstrated the robustness of SEM with respect to categorical data, our results should be interpreted with caution. Another limitation of this study is that we did not distinguish between children who had experienced parental divorce and children who for other reasons live in single-parent families. Although in the research literature the effects of divorce and living with a single parent on the career aspirations of children are often difficult to disentangle (Barber & Eccles, 1992), future research may be able to separate out possible differences in these effects on children from these two types of families. An additional limitation of this study is that we did not measure the actual amount of involvement of the noncustodial parent. Future research could investigative how the magnitude of this involvement affects adolescents' math efficacy, outcome expectations, and math and science career interests.

TABLE 1 Covariances, Path Regression Weights, and Bootstrap Estimates

                                                      Boot-
Variable                            Estimate    SE    strap    SE     p

Covariance
Family Structure  Father Support      .187     .042    .187   .047  .006
Family Structure  Mother Support      .097     .040    .097   .046  .049
Family Structure  Gender-Typing      -.108     .043   -.104   .051  .022
Father Support    Mother Support      .413     .047    .412   .069  .006
Father Support    Gender-Typing      -.209     .045   -.204   .057  .003
Mother Support    Gender-Typing      -.147     .044   -.142   .046  .006

Regression weight
Math Efficacy     Gender-Typing      -.144     .044   -.148   .057  .016
Math Efficacy     Father Support      .193     .057    .205   .087  .035
Math Efficacy     Mother Support      .206     .055    .196   .074  .008
Math Efficacy     Family Structure    .181     .048    .179   .057  .006
Outcome Exp.      Math Efficacy       .347     .066    .341   .083  .006
Outcome Exp.      Gender-Typing      -.149     .053   -.148   .059  .008
Outcome Exp.      Father Support     -.024     .068   -.028   .076  .752
Outcome Exp.      Mother Support      .233     .066    .242   .075  .009
Outcome Exp.      Family Structure   -.081     .057   -.083   .058  .178
Interests         Math Efficacy       .236     .068    .242   .081  .012
Interests         Outcome Exp.        .181     .060    .178   .073  .022

Note. df = (4, 314); N = 318. Outcome Exp. = Outcome Expectations.

References

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

Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (2001). Self-efficacy beliefs as shapers of children's aspirations and career trajectories. Child Development, 72, 187-206.

Barber, B. L., & Eccles, J. S. (1992). Long-term influence of divorce and single parenting on adolescent family- and work-related values, behaviors, and aspirations. Psychological Bulletin, 111, 108-126.

Betz, N., & Hackett, G. (1983). The relationship of mathematics self-efficacy expectations to the selection of science-based college majors. Journal of Vocational Behavior, 23, 329-345.

Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing model fit. Sociological Methods and Research, 21, 230-258.

Church, A. T., Teresa, J. S., Rosebrook, R., & Szendre, D. (1992). Self-efficacy for careers and occupational consideration in minority high school equivalency students. Journal of Counseling Psychology, 39, 498-508.

Fennema, E., & Sherman, J. A. (1976). Fennema-Sherman Mathematics Attitude Scales: Instruments designed to measure attitudes toward the learning of mathematics by females and males. Catalog of Selected Documents in Psychology, 6(31, Manuscript No. 1225).

Ferry, T. R., Fouad, N. A., & Smith, P. L. (2000). The role of family context in a social-cognitive model for career-related choice behavior: A math and science perspective. Journal of Vocational Behavior, 57, 348-364.

Fouad, N. A., & Smith, P. L. (1996). A test of a social cognitive model for middle school students: Math and science. Journal of Counseling Psychology, 43, 338-346.

Fouad, N. A., Smith, P. L., & Zao, K. E. (2002). Across academic domains: Extensions of the social-cognitive career model. Journal of Counseling Psychology, 49, 164-171.

Gainor, K. A., & Lent, R. W. (1998). Social cognitive expectations and racial identity attitudes in predicting the math choice intentions of Black college students. Journal of Counseling Psychology, 45, 403-413.

Hackett, G., & Betz, N. E. (1989). An exploration of the mathematics self-efficacy/mathematics performance correspondence. Journal of Research in Mathematics Education, 20, 261-273.

Joreskog, K. G., & Sorbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Hillsdale, NJ: Erlbaum.

Lamb, R., & Prediger, D. (1995). Technical manual: Revised unisex edition of the ACT Interest Inventory. Iowa City, IA: American College Testing Program.

Lapan, R. T., Hinkelman, J., Adams, A., & Turner, S. (1999). Understanding rural adolescents' interests, values and efficacy expectations. Journal of Career Development, 26, 107-124.

Lapan, R. T., & Jingeleski, J. (1992). Circumscribing vocational aspirations in junior high school. Journal of Counseling Psychology, 39, 81-90.

Lapan, R. T., Shaughnessy, P., & Boggs, K. (1996). Efficacy expectations and vocational interests as mediators between sex and choice of math/science college majors: A longitudinal study. Journal of Vocational Behavior, 49, 277-291.

Lauver, P. J., & Jones, R. M. (1991). Factors associated with perceived career options in American Indian, White, and Hispanic rural high school students. Journal of Counseling Psychology, 38, 159-166.

Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory of career and academic interest, choice, and performance. Journal of Vocational Behavior, 45, 79-122.

Lent, R. W., Brown, S. D., & Hackett, G. (2000). Contextual supports and barriers to career choice: A social cognitive analysis. Journal of Counseling Psychology, 47, 36-49.

Lent, R. W., Lopez, F. G., & Bieschke, K. J. (1991). Mathematics self-efficacy: Sources and relation to science-based career choice. Journal of Counseling Psychology, 38, 424-430.

Lent, R. W., Lopez, F. G., & Bieschke, K. J. (1993). Predicting mathematics-related choice and success behaviors: Test of an expanded social cognitive model. Journal of Vocational Behavior, 42, 223-236.

Lopez, F. G., & Lent, R. W. (1992). Sources of mathematics self-efficacy in high school students. The Career Development Quarterly, 41, 3-12.

Lopez, F. G., Lent, R. W., Brown, S. D., & Gore, P. A. (1997). Role of social-cognitive expectations in high school students' mathematics-related interest and performance. Journal of Counseling Psychology, 44, 44-52.

McWhirter, E. H., Hackett, G., & Bandalos, D. L. (1998). A causal model of the educational plans and career expectations of Mexican American high school girls. Journal of Counseling Psychology, 45, 166-181.

Mulhern, F., & Rae, G. (1998). Development of a shortened form of the Fennema-Sherman Mathematics Attitudes Scales. Educational and Psychological Measurement, 58, 295-306.

Muthen, B., & Kaplan, D. (1985). A comparison of some methodologies for the factor analysis of non-normal Likert variables. British Journal of Mathematical and Statistical Psychology, 38, 171-189.

O'Brien, V., Martinez-Pons, M., & Kopala, M. (1999). Mathematics self-efficacy, ethnic identity, gender, and career interests related to mathematics and science. Journal of Educational Research, 92, 231-235.

Post, P., Stewart, M. A., & Smith, P. L. (1991). Self-efficacy, interest, and consideration of math/science and non-math/science occupations among Black freshmen. Journal of Vocational Behavior, 38, 179-186.

Randhawa, B. S., Beamer, J. E., & Lundberg, I. (1993). Role of mathematics self-efficacy in the structural model of mathematics achievement. Journal of Educational Psychology, 85, 41-48.

Sandefur, G. D., McLanahan, S., & Wojtkiewicz, R. A. (1992). The effects of parental marital status during adolescence on high school graduation. Social Forces, 71, 103-121.

Siegel, R. G., Galassi, J. P., & Ware, W. B. (1985). A comparison of two models for predicting mathematics performance: Social learning versus math aptitude/anxiety. Journal of Counseling Psychology, 32, 531-538.

Steiger, J. H., & Lind, J. M. (1980, May). Statistically based tests for the number of common factors. Paper presented at the annual meeting of the Psychometric Society, Iowa City, IA.

Trice, A. D., Hughes, M. A., Odom, C., Woods, K., & McClellan, N. C. (1995). The origins of children's career aspirations: IV. Testing hypotheses from four theories. The Career Development Quarterly, 43, 307-322.

U.S. Census Bureau: U.S. Department News. (1995, October 16). Single parent growth rate strabilized: Two-parent family growth renewed, Census Bureau reports. Retrieved August 31, 2004, from http://www.census.gov/Press-Release/cb95-186.html

Vezeau, C., Chouinard, R., Bouffard, T., & Couture, N. (1998). Adaptation et validation des echelles de Fennema et Sherman sur les attitudes en mathematique des eleves du secondaire [Adaptation and validation of the Fennema-Sherman Mathematics Attitudes Scales among secondary students]. Canadian Journal of Behavioral Science, 30, 137-140.

Sherri L. Turner, Department of Educational Psychology, Counseling and Student Personnel Psychology, University of Minnesota; Jason C. Steward, Department of Psychology, Counseling Psychology, University of Minnesota; Richard T. Lapan, Department of Educational, School, and Counseling Psychology, University of Missouri--Columbia. This research was funded by a grant from the National Science Foundation. Correspondence concerning this article should be addressed to Sherri L. Turner, Counseling and Student Personnel Psychology, University of Minnesota, 139A Burton Hall, 178 Pillsbury Drive SE, Minneapolis, MN 55455 (e-mail: [e-mail]).