Self-regulation and the transition to adulthood.

Abstract

Longitudinal interview data from 50 adolescents reveal that an early sense of realism, or knowledge about requirements of one's future goals, contributes to adolescents' self-regulation, or willingness to regulate actions to achieve those goals. Self-regulation, in turn, is associated with adolescents' engagement in career-related activities. These internal resources, rather than background characteristics, most robustly predict engagement in activities related to future goals.

Introduction

Adolescents' future plans are often not accompanied by specific strategies for achieving these educational and occupational goals. The majority of adolescents today expect to attend college; in fact, adolescents attend college in higher numbers than ever before (Schneider & Stevenson, 1999). Because of the highly ambitious nature of young people today, competition is more intense than ever before, and adolescents must learn how to develop a pathway to their goals (Csikszentmihalyi & Schneider, 2000). Encouraging adolescents to develop high ambitions is no longer the most critical issue; the more pressing question is how adolescents translate ambitions into accomplishments.

Adolescents rely on both external and internal resources for future success. From parents, teachers, and peers, adolescents need to obtain an educational foundation, information about college and careers, and financial assistance to attend college. Internally, adolescents need to develop a sense of realism and self-regulation about their abilities, goals, and plan of action. Realism refers to a person's career-specific knowledge about their desired occupation, and it develops through access to information from parents, guidance counselors, and peers about the skill and educational requirements of a job. A person with a high sense of realism is likely to develop aligned ambitions; that is, knowledge about the specific and appropriate path for their stated goals. Self-regulation measures adolescents' ability to monitor their activities, assess their performances, motivate themselves, and maintain resiliency while learning from academic and social disappointments (Zimmerman & Martinez-Pons, 1988, 1990). Self-regulation draws on social cognitive functioning, and, therefore, adolescence is a particularly salient developmental period for the creation of self-regulation. Because self-control is one element of self-regulation, it might develop with maturity as adolescents are better able to control their impulses and act appropriately to their situations.

Research on self-regulation's role in education has generally focused on academic achievement; students who display higher levels of self-regulation perform better on academic tasks (Bandura, 1993; Zimmerman & Bandura, 1994). Here, self-regulation of learning is extended and applied to academic success in terms of taking the steps to transitioning to postsecondary education or gainful employment. Being self-regulated aids students in preparing for immediate academic success by facilitating good study habits, but self-regulation may also assist students in achieving long-term goals by enabling them to see the "big picture" and plan their current actions according to the needs of their future.

Researchers have argued that self-efficacy, or judgment of one's own abilities (Bandura, 1986), informs self-regulation; that is, students who believe in their abilities are much more likely to be highly motivated and appropriately apply themselves to their tasks (Zimmerman & Bandura, 1994). While these researchers have argued that knowledge of one's own abilities predicts later success (Bandura, 1977; Lent, Brown, & Larkin, 1984), we hypothesize here that knowledge with respect to the demands of one's future career aspirations, i.e. realism, informs self-regulation, which then serves as a motivational and monitoring tool for adolescents. Knowing the pathway to one's goal is likely to increase one's willingness and ability to regulate actions to achieve that goal. Without knowledge about the process to successfully achieve one's goal, students might feel uncertain about the future and be less likely to commit to regulating their actions. In this paper, self-regulation in later adolescence is predicted to determine the influence realism about the future and self-regulation in earlier adolescence has on its development. Self-regulation as both a state informed by outside information and as a developmental concept is therefore explored. The impact of self-regulation on students' engagement in activities related to their future goals is assessed to determine the importance of self-regulation as a tool for transitioning to the future.

Data and Methods

Sample

This paper examines longitudinal data from the Alfred P. Sloan Study of Youth and Social Development. This study was designed to examine how adolescents transition from high school to postsecondary activities.[1] Data were collected in four waves: 1992-1993, 1993-1994, 19941995, and 1996-1997. Participants from twelve sites across the country completed surveys regarding academic performance, peer and family relationships, and ideas about the future; the Experience Sampling Method, a time diary instrument requiring respondents to report both their activity and subjective evaluation of that activity; and in-depth interviews, which included questions about the student's family life, friendships, future expectations, and perception of adult work. Fifty participants (the Sloan 50) were selected for further follow-ups by block randomizing the larger sample based on race/ethnicity, academic ability, and socioeconomic status. In Year 1, the Sloan 50 included eight 8th graders, twenty-nine 10th graders, and thirteen 12th graders. Here, data from 3 years of the study (Years 1, 3, and 5) are examined to understand how the adolescents' perceptions of and plans for the future changed over time. These years were selected because each represents a transitional point in adolescence or early adulthood: middle to high school, high school to college, or college to career. Examining three years of interview data from each participant allows for consideration of the developmental aspect of characteristics.

Variables

Three interviews (Years 1, 3, and 5) with each Sloan 50 participant were coded to rate characteristics relevant to transitions to the future. Variables were rated on a 1 to 5 scale. The interviews were independently read and coded by six researchers with a Cronbach's alpha measuring inter-rater reliability of .76.

Dependent Variables. Self-regulation is measured as the level of control one has over his or her actions, the ability to keep actions in line with goals while acting on self-selected goals, and knowledge about how one's own skills could be best applied to future goals. Self-regulation also includes indicators of level of motivation and self-discipline. In Years 1 and 5, the median score was (3); in Year 3, it was (4), indicating that self-regulation varies somewhat according to age. An interview in which an adolescent acknowledges needing to study more in order to get good grades or needing to balance his or her social life with schoolwork would receive the median selfregulation score of (3); the adolescent understands the need for self-control, but is not always practicing it effectively. To measure how self-regulation influences the actions adolescents take toward achieving their goals, a measure of current actions related to future goals is also included in analyses. This variable was defined as the degree to which the adolescent was actively pursuing a course of action likely to lead to his or her expressed goals. The median score for current actions related to future goals in Years 1 and 3 was (3); in Year 5, it was (4), indicating that students became more engaged in career-related activities as they neared the end of high school or entered college. The median score of (3) meant that the respondent was focusing on achieving good grades and taking courses in their area of interest. A higher score would indicate the adolescent's participation in an internship or research position related to future goals.

Independent Variables. To predict self-regulation, both self-described characteristics from interview data and demographic traits were considered. Realism was assessed by evaluating the level of knowledge expressed in the interviews about the educational and training requirements of stated goals. The median score in Years 1, 3, and 5 was (3), which implied that the adolescent was generally aware of the amount of schooling required for the job or of what the job itself entailed. For example, an adolescent aspiring to be a teacher who mentioned attending college but didn't know the specifics of student-teaching would receive a (3). While rating the adolescent on realism, the coders looked at their level of realism in terms of their knowledge of their own skills. A respondent who hoped to be a professional athlete but had not made the varsity team would be scored lower than a respondent hoping to be a chef who had done well in a culinary arts course.

Measures of social support were analyzed to determine how social relationships influenced self-regulation. The social relationships students develop with family members and peers as they begin to plan for the future can provide important information, opportunities, and financial resources for later aspirations. Complex family background measured the level of both support and challenge a respondent's family provided with respect to their plans for the future. An interview rated as a (3) (the median score in Years 1, 3, and 5) indicated an adolescent who spoke about one aspect of family background but not both. Parents who provided only support might allow a lot of freedom or encourage every route the adolescent proposed for his or her future but not hold high expectations, whereas parents who only provide challenge might pressure their children to succeed in school and get an early start in future planning but not provide emotional support or nurturing. Interviews were assessed for degree of sociability, which measured adolescents' peer networks. The median score of (3) in all three years indicated a respondent who had friends outside of school. Coders rated adolescents higher if they mentioned close friends with whom they had supportive relationships that facilitated discussion about future plans and support for goals. While many students reported a broad social circle, those whose friends appeared to be particularly engaged in helping each other pursue goals were rated more highly.

The demographic traits included in analyses were parents' education level, students' and parents' educational expectations, and race. Parents' education level measured the higher of either parent's educational attainment on a 1 to 4 scale (1: high school diploma or less, 2: some education beyond high school; 3: college degree; 4: advanced degree). The mean level of education among parents was 2.88--the average parent had pursued some education beyond high school. Educational expectations were assessed on the same scale as education level, and the higher of either parent's expectations are included. Both students and parents expressed an average expectation level of over 3, at least a college degree. Forty-one students in the sample were white, two were African American, two were Asian American, four were Hispanic, and one was Native American. Thirty-five girls and 15 boys made up the Sloan 50 subsample.

Results

Because longitudinal data were analyzed, correlations between the variables at each year were examined to determine which variables were dispositional and unchanging and which were affected by other variables. Results are displayed in Table 1. See issue website http://rapidintellect.com/AEQweb/win2005.htm Some variables are not perfectly correlated within the same subject across three years, which indicates that they are not only influenced by one's developmental cycle, but that intervening variables can also affect development of these traits. Realism in Years 1 and 3 is associated with realism in Year 5, indicating that realism might be sustained over time. Self-regulation in Year 1 is positively correlated with self-regulation in Year 3 but not Year 5, suggesting that self-regulation may not be stable, but rather developed and sustained by other factors. Looking between variables, realism in all three years and complex family background in Year 1 are associated with self-regulation. Self-regulation in all years is associated with current actions related to goals within that year. Parents' education and expectations are associated with current actions in Year 1 but not later years, suggesting that as adolescents mature and depend less on parents, they rely on internal resources to keep on track. Race is not significantly associated with any of the variables, indicating no fundamental differences with respect to these characteristics.

Predicting Self-Regulation

To understand how self-regulation develops, a series of OLS regressions were conducted to determine if self-regulation is a constant state or one informed by outside sources such as realism and social support. See issue website http://rapidintellect.com/AEQweb/win2005.htm As seen in Table 2, self-regulation in Years 1 and 3 were included in analyses to test the extent to which self-regulation is sustained regardless of differences between individuals in terms of realism or student background traits. When only self-regulation in earlier years is included in the model, self-regulation in Year 3 is a significant predictor of later self-regulation. Once realism is included in the model, early self-regulation is not a significant predictor, suggesting that self-regulation is not a dispositional constant, but a resource that emerges through outside influences.

Including social support and student background variables, realism in Year 1 remains the most robust predictor of self-regulation in Year 5. Having clear and early knowledge about what one must do to achieve one's goals appears to be a critical resource for becoming self-regulated. For adolescents, it is much easier to stay motivated and keep one's behavior in check if a path towards success has been worked out. The standardized coefficients for realism are quite large, about .500 for all models, with an R2 of .625 in the final model, revealing a strong relationship.

The social context variables of complex family background and sociability are not significant predictors in the models shown and further analyses not shown. However, family and friends are often important sources of information about career and college and contribute to self-regulation by influencing adolescents' realism. Students' demographic characteristics were also not significant predictors, with the exception of students' expectations becoming marginally significant when race, parents' education, and parents' expectations are controlled for. Students with higher educational expectations appear to be more self-regulated, likely because of their knowledge about the higher demands of their future goals. However, background characteristics are not as important to adolescents' development of self-regulation as realism.

The Impact of Self-Regulation on Actions

Being realistic appears to encourage adolescents to regulate their actions in accordance with future goals. To test the strength of self-regulation as a predictor of taking action towards these goals, OLS regressions were conducted to predict current actions related to goals in Year 5 (See Table 3). See issue website http://rapidintellect.com/AEQweb/win2005.htm Including only self-regulation in the model produced no significant predictors. When early realism is included, self-regulation becomes a significant predictor. Student background characteristics were not significant predictors of current actions related to goals, and self-regulation in Year 3 was the strongest predictor of participating in activities relevant to future goals in these models. [2] Models including realism had the lowest predictive power, suggesting that realism alone is not enough to facilitate action, but that it must be accompanied by self-regulation. Self-regulation was only borderline significant in most models, but a larger sample size might provide stronger results about the importance of self-regulation in engaging in career-related activities.

Discussion

Realism, or early and specific career knowledge, is integral in the development of later self-regulation, which influences students' engagement in actions relevant to their future goals. These two internal resources appear to be more critical in facilitating early participation in career-relevant activities than background characteristics. Research has indicated that students who do not have access to early career information often turn to the media to create ideas about future occupations (Schneider and Stevenson, 1999; Csikszentmihalyi and Schneider, 2000). If parents and teachers do not provide adolescents with a realistic sense of their world and their future, adolescents are likely to form unrealistic aspirations and not know how to successfully achieve their goals. By providing adolescents with early knowledge, parents and teachers can contribute to the development of a strong sense of self-regulation, which facilitates participation in goal-relevant activities.

Results are limited by the small sample size of students with whom longitudinal, in-depth interviews could be conducted. Analyzing 150 interviews from 50 students across three years does not guarantee trends would hold in a larger dataset. Additional analyses such as multi-level growth curve models to examine developmental constructs are not possible because of the sample size. Another limitation of the dataset is that the measures are imprecise as they are based on qualitative data. However, when examining intrapersonal resources, adolescents are unlikely to be able to assess their own levels of realism or self-regulation accurately, making the use of interview data appropriate. While analyses are imperfect, the strong relationship between realism and self-regulation indicates a general pattern that early knowledge and access to information facilitates the development of internal resources like self-regulation. These preliminary results suggest that realism and self-regulation are two internal resources that adolescents can draw on to stay motivated and engaged in activities relating to their goals.

References

Bandura, A. (1977). Social learning theory. Englewood Ciffs, N.J.: Prentice-Hall Inc.

--. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall, Inc.

--. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist, 28, 117-148.

--(1999). Social cognitive theory: An agentic perspective. Asian Journal of Social Psychology, 2, 21-41.

Csikszentmihalyi, M., & Schneider, B. (2000). Becoming adult. New York: Basic Books.

Lent, R.W., Brown, S.D., & Larkin, K.C. (1984). Relation of self-efficacy expectations to academic achievement and persistence. Journal of Counseling Psychology, 31, 356-362.

Schneider, B., & Stevenson, D. (1999). The ambitious generation. New Haven, CT: Yale University Press.

Zimmerman, B., & Bandura, A. (1994). Impact of self-regulatory influences on writing course attainment. American Educational Research Journal, 31, 845-862.

Zimmerman, B., & Martinez-Pons, M. (1988). Construct validation of a strategy model of student self-regulated learning. Journal of Educational Psychology, 80, 284-290.

--. (1990). Student differences in self-regulated learning: Relating grade, sex and giftedness to self-efficacy and strategy use. Journal of Educational Psychology, 82, 51-59.

Notes

[1] For further sample information, see Csikszentmihalyi and Schneider (2000).

[2] Social context variables were not significant predictors (analyses not shown).

Ann Owens, University of Chicago, IL

Barbara Schneider, University of Chicago, IL

Owens is a research associate of the Sloan Center at the University of Chicago. Schneider is co-director of the Sloan Center at the University of Chicago.