Evaluation of an empowerment program for early adolescent girls.

LeCroy, Craig Winston

Adolescence, and certainly early adolescence, is a period of multiple, rapid, and profound changes and transitions. Over the past decade, there has been growing concern among researchers, clinicians, and policy makers about the overall health status of adolescents. Such concerns are particularly important for adolescent girls: "Girls today are much more oppressed. They are coming of age in a more dangerous, sexualized, and media-saturated culture. They face incredible pressures to be beautiful and sophisticated, which in junior high means using chemicals and being sexual. As they navigate a more dangerous world, girls are less protected" (Pipher, 1994, p. 12).

Dryfoos (1998) estimates that one third of U.S. 14-year-olds are at high to very high risk for future behavior problems. Consider the broad range of risky behaviors and pitfalls that await adolescents (Brindis, Irwin, Ozer, Handley, Knopf, & Millstein, 1997; Resnick et al., 1997).

Delinquency, for example, both major and minor, begins for most youth in early adolescence and reaches its height during later adolescence. Experimentation with alcohol, tobacco, and other drugs also begins for most children during adolescence, and it is during this time that they establish critical use patterns that extend into adulthood. Exploration of sexuality begins in early adolescence and for many girls leads to sexual intercourse, unwanted pregnancy, or sexually transmitted diseases. Beyond these problems, many adolescents begin to lose interest in school, which can often be further translated into school failure and dropout. Moreover, conflict with peers and parents can lead to dissatisfaction with one's self. All of these social problems can influence an adolescent's life course and have a long-term impact (Johnson & Millstein, 2003; Graber, Brooks-Gunn, & Petersen, 1996).

Adolescence is an ideal age group for preventive interventions to avert or delay the onset of problem behaviors. Indeed, too often preventive interventions target older youth when it is too late because problem behaviors have already started. Although the early adolescent years are a time of increased vulnerability and potential risk, they also constitute special opportunities for preventive interventions. As Hamburg and Takanishi (1989) argue: "One of the most neglected opportunities in disease prevention and health promotion has been the thoughtful exploration of how to reach large numbers of adolescents with preventive approaches" (p. 826). Many experts on adolescence have found public policy to be unresponsive to research evidence regarding when and how to intervene in ways that can produce positive impacts for young people and for the nation. The Carnegie Corporation launched the Carnegie Council on Adolescent Development to help bridge the gap between knowledge and practice to improve professional practice, program and policy development, and research for early adolescents (Carnegie Corporation, 1995). As representatives of the Council, Hamburg and Takanishi (1989) pointedly state: "It is essential that we move early adolescence much higher on the nation's agenda for scientific research and for youth and family policy" (p. 826). Current interest has grown in bridging practice and research and there is more optimism in the potential of prevention science (Cicchetti et al., 2000; Weissberg, Kumpfer, & Seligman, 2003).

This article describes an intervention for early adolescent girls taking into account recommendations that the development of prevention programs be based on the principles of prevention science (Coie et al., 1993). For example, developmental considerations were important in designing a program that was specific to a particular gender and age group. This allowed for the incorporation of dynamic developmental process variables and an understanding of systems of influence. Developmental theory was used in conceptualizing risk and protective factors. The notion of developmental tasks for the development of competence was critical in the intervention design. The research approach involved specifying a universal preventive intervention to be administered to a population in which there would be the likelihood of beneficial gains and no potential adverse effects for participants. Preliminary evaluation used short-term effects to test the impact of the intervention with the recognition that such analysis would lead to a further refinement of the program model.

Empirical or evidence-based prevention programs are needed to evaluate the potential of broad-based prevention efforts for adolescents. Denmark (1999) has specifically called for research on the development and evaluation of interventions for adolescent girls. Barber and Crockett (1993) make a similar call for broad-based prevention research with adolescents that might address multiple target areas. In addition, Kazdin and Weisz (1998) advocate for applied research in community settings. Thus, the need for this type of research is well established. With these research issues in mind, the purpose of this article is to describe a gender-specific prevention program for early adolescent girls and to evaluate the outcomes related to its implementation in a community setting.

METHOD

Participants

The Go Grrrls Program was presented to school officials who authorized their school to be an experimental site for the evaluation of the program. During a lunchtime period, girls were recruited to participate in the program. If girls agreed to participate, parents and adolescents completed an informed-consent agreement. This process yielded a pool of 55 volunteers from the school which was located in a suburban area of the Southwest that includes both a rural and a more urbanized population. Descriptive data collected from the participants revealed a mean age of 12.7 years. The sample was culturally diverse: 64.8% were Caucasian, 18.5% mixed race, 11.1% Hispanic, 3.7% African American, and 1.9% Asian American. Approximately one third of the participants had a mother who attended college. Approximately half lived in single-parent homes, and one sixth received a free school lunch. Nearly two thirds had experienced menarche.

Group Leaders

The intervention was delivered in small groups composed of 8 to 10 girls per group. Each group was lead by two female leaders who were primarily graduate level social work or psychology students. Group leaders were interviewed and hired based on previous work experience with adolescents, group leadership skills, and demonstrated interest in the program. The intervention was scripted in a detailed curriculum that described how each session was to be conducted (see LeCroy & Daley, 2001). Included in this manual are the verbal scripts and handout materials used, as well as the themes the curriculum was attempting to influence. All group leaders were supervised to ensure that the curriculum was being carried out in a standardized manner.

Procedure

Evaluation of the program was based on a quasi-experimental design. Those who volunteered for the program constituted the experimental group. A matched group of participants in a physical education class constituted the comparison group. The comparison group was a no-treatment group. Participants were pretested and posttested, but otherwise engaged in their regular physical education activities. In the experimental group, seven to eight participants were assigned to one of three groups. All participants completed questionnaires before and after the intervention to assess performance on the dependent measures. Testing was completed in classrooms where girls were given instructions by a trained research assistant. Participants were informed that the information gathered would be confidential.

The Go Grrrls Curriculum

The curriculum content for the program consists of six developmental tasks considered critical for the healthy psychosocial development of early, adolescent girls in contemporary society: being a girl in today's society (gender role identity), establishing a positive self-image, establishing independence, making and keeping friends, when it all seems like too much (using resources), and planning for the future (LeCroy & Daley, 2001). Development is conceptualized as a process whereby young people must learn to adapt to the necessary tasks placed on them by the social environment. Healthy development for early adolescent girls is defined by these tasks. The prevention program for girls was designed to emphasize competencies that help girls learn the information and skills necessary to successfully meet the demands being placed on them by peers, family, and society. The broad-based prevention program is gender specific because there are unique tasks that early adolescent girls need to master in order to transition successfully to adulthood.

These task areas each constituted a curriculum section in the Go Grrrls Program, and two sessions were devoted to each task for a total of 12 sessions. For example, the first task was to achieve a competent gender role identification. Early adolescence is a time for developing one's gender role identity. This task is particularly critical in today's society where social forces exert a powerful and early influence on how individuals view their gender identity. The second task was the development of a positive self-image. Adolescent girls need to develop acceptance of the self as a stable person of worth. However, many girls set themselves up for failure by aspiring to unattainable goals and self-criticism. In particular, adolescent girls are likely to develop a negative body image, which is related to measures of low self-esteem and depression. The third task was establishing independence. In early adolescence, girls come to a critical juncture where they need to find their own independent voice. Girls need to develop assertiveness and problem-solving skills so that they can establish their independence in a culture that is all too willing to relegate them to silence. The fourth task was making and keeping friends. Peer relations are critical to the healthy development of the adolescent. Membership in the peer group is a central process that must be successfully mastered by early adolescents. A fifth task involved learning how to find and obtain resources when pressures get to be too much. Since the social environment is often unfriendly to girls in many ways, an important contemporary developmental task is the ability to find and use needed resources. Lastly, planning for the future is a critical task for early adolescents. During early adolescence, young people begin to adopt a more serious attitude about the future. Sadly, during their early adolescent years, many girls experience a "crisis of confidence" that can seriously undermine their chances for educational and career success later in life. Since these developmental tasks formed the core of the program, two sessions were devoted to each task.

Measures

Basic demographic information and all measures were contained in a self-report booklet that the girls were asked to complete. Because the curriculum content was diverse, a multiple measurement strategy was needed. It was not clear to what extent the different program components might have influenced different types of outcomes. Most of the measures were based on previously standardized scales that have been subjected to reliability and validity testing. The measurements were selected based on age appropriateness for an adolescent population as well as ease and simplicity in completing the items. Since the school allowed only a limited amount of time for pretesting and posttesting the participants, the measurement model was shaped by pragmatic concerns.

Concern with Body Image Scale. This five-item scale measures satisfaction with body image (Simmons & Blythe, 1987). Items include: "How happy are you with your overall figure?" and "How happy are you with how much you weigh?" Items are scored on a four-point scale, ranging from not at all to very much.

Gender Role Attitudes Scale. This three-item scale measures attitude toward being a girl (Simmons & Blythe, 1987). Sample items include: "Would you say you feel good about being a girl?" and "How important is it for you not to act like a girl?" Items are scored on a four-point scale, from very much to not at all.

Peer Self-Esteem Scale. In this 10-item scale, self-esteem is measured by asking subjects to assess their friendships (Hare, 1985). Sample items include: "I have at least as many friends as other people my age," and "Other people think I am a lot of fun to be with." Items are scored on a four-point scale, from strongly disagree to strongly agree.

Common Beliefs Inventory. This 44-item scale measures common irrational beliefs (Hooper & Layne, 1983). A shortened version of 24 items was used in the present study. Sample items include: "If a person doesn't have any friends, that means that nobody likes him," and "I believe I should be a better person." The items are scored on a five-point scale, from never to always.

Depression Self-Rating Scale. This is an 18-item scale that measures extent and severity of depression in children between the ages of 7 and 13 (Birleson, 1981). Sample items include: "I feel like crying" and "I am easily cheered up." The items are scored on a 3-point scale that includes never, sometimes, and most of the time.

Help Endorsements Scale. The scale is a list of 15 possible sources of help (e.g., "hotlines and crisis centers" and "friend your age"). Subjects are asked to circle all the sources they might use if they need help. This scale was specifically designed by the author for use in the present study.

RESULTS

A total of 55 girls agreed to participate in the program: 23 were in the treatment group and 32 were in the control group. The analysis of measurement results was based on the responses of 48 to 54 girls (one girl dropped out of the program and some sections in the questionnaire were not properly completed, therefore not all of the individuals were included in the analysis). Fidelity of the program components was monitored by the degree to which each session occurred as described in the curriculum program. Ongoing supervision was provided to minimize problems with program fidelity. Group leaders were trained to follow a detailed script of the program to ensure this fidelity. The complete program curriculum or treatment manual is available for review (see LeCroy & Daley, 2001).

The results of the reliability analysis for the measures at the pretest time period were: Depression = .80; Concern with Body Image = .67; Gender Role Attitude = .36; Peer Esteem = .80; Common Beliefs = .83. Because the Gender Role Attitude Scale's reliability was .36, this measure was not used in the study. The remaining measures demonstrated low but adequate reliability.

Preliminary analyses to test for differences between the intervention and comparison groups were conducted to identify variables for potential use as covariates. On all major demographic variables (age, ethnicity, school lunch, single-parent families, and menarche), there were no significant differences between the groups at the pre-intervention stage. Because there were no differences between the groups and these variables were not predicted to interact with the program mediators, they were not included in any additional analysis.

The primary analysis compared mean scores of the dependent measures between the intervention and comparison groups using a one-way ANCOVA that used the pretest scores as a covariate and posttest scores as the dependent variable. Univariate analysis was conducted for the outcome variable because it was theoretically important to examine the effectiveness of the different components. Additionally, in some instances different components of the program were related directly to specific outcome measures. For example, the peer esteem variable was one component of the overall program. The measures showed some moderate correlation, but overall there was fairly good independence between the measures. The highest correlation was between common beliefs and depression, r = .41. Table 1 presents the means and standard deviations for pretest and posttest scores across the intervention and comparison groups.

Three of the five dependent measures produced a significant group by time effect. The intervention group reported greater increases in peer esteem, F(1, 53) = 4.74, p < .01; help endorsements, F(1, 52) = 3.49, p < .03, and beliefs F(1, 48) = 5.19, p < .01, in contrast to the comparison group. One measure, depression, would be significant if the significance standard was set at the .15 level. Thus, on this outcome, researchers may want to suspend judgment (Keppel, 1991) regarding significance. One variable that clearly did not show a significant group by time effect was body image. The effect sizes (eta squared) for the measures were, depression, .02; body image, .006; peer esteem, .08; to help endorsements, .03; and common beliefs, .09. Cohen (1977) states that a small effect is .01 and a medium effect is .06.

DISCUSSION

Adolescent girls face a difficult culture as they transition into adulthood. The changes they experience are defined by the social context and by their interaction and behavior in that context. The transition to adolescence is a significant issue when the cultural context is considered. For example, research by Caspi and Moffitt (1991) found that for some girls the adolescent transitional period resulted in a worsening pattern of poor adjustment. This may be the case because during this time period girls are adapting to new roles and contexts (Graber, Brooks-Gunn, & Petersen, 1996). Stewart (1982) believes that successful adaptations are difficult during a transitional period due to feelings of incompetence or of being overwhelmed. However, with adaptation and coping these feelings change, resulting in more integrated emotional responding and the ability to develop new courses of action. Unfortunately, few universal prevention program have been designed to address these issues. The outcome data from this experiment are interpreted as showing potential for the development of an effective program like the Go Grrrls preventive intervention. Dependent measures subjected to a time by group analysis showed increases in the intervention group above those in the comparison group on three of the five outcomes: peer esteem, common beliefs, and help endorsements, and the fourth measure, depression, approached significance.

Peer esteem is a key variable because of its importance to this age group. Previous research has found that as peer relationships become more salient during adolescence, attachment to friends can have a positive effect on mental health (Cauce, 1986; Kenny & Donaldson, 1991). More recent research has found that adolescents with low levels of friendship had greater levels of anxiety and depression, lower self-esteem, and less effective coping styles than did adolescents with higher friendship levels (Notaro, Miller, & Zimmerman, 1998). Notaro et al. also found friendship attachment was a stronger predictor of mental health outcomes for females than for males. The results of the present study suggest that peer friendship esteem can be changed during this critical adolescent developmental period for a broad group of adolescent girls.

The Common Beliefs scale also showed change for the intervention participants. This measure tapped into a main part of the intervention that focused on the development of a positive-image for girls. In their classic analysis of self-image and adolescent development, Simmons and Blythe (1987) postulated that a primary task of early and middle adolescence is to achieve a positive sense of self in response to the changes that occur in adolescence. Unfortunately, this often does not occur for adolescent girls. Several studies (Conger, Peng, & Dunteman, 1977; Rosner & Rierdan, 1994; Simmons & Blythe, 1987) have continued to document differences between boys' and girls' self-image, self-confidence, and self-esteem. More significantly, longitudinal research (Bush & Simmons, 1988; Eccles et al., 1993) has shown that girls who transition from the elementary to middle school show a precipitous drop in self-esteem and self-confidence. Preventive interventions can have a positive effect on girls' mental health and help prevent the downturn of girls' self-esteem as they make these transitions. This study suggests that aspects of a girl's self-image can be changed by the proposed preventive intervention.

Help endorsements were also found to change more for Go Grrrl participants than for those in the comparison group. Most adolescents do not have access to the resources and support structure they need (Millstein, 1988). It is therefore critical to enhance opportunities that can help facilitate healthy development and proper preparation for adulthood (Johnson & Millstein, 2003). The ability to recognize and ask for help is becoming increasingly critical for young people growing up in a society that poses many risks. Furthermore, girls are often overrepresented regarding problems that are more "invisible" or "inner directed" such as negative self-image, depression, eating disorders, and poor body image (LeCroy & Daley, 2001). Because these problems are easily hidden, girls who require assistance may successfully conceal them until they have reached a precarious level. Also, too often, middle school girls do not know where to go or whom to trust. Locating help for adolescents is not only problematic for youth themselves but is also a major challenge for professionals, parents, and communities. The positive finding on help endorsements suggests progress can be made in solving the serious underutilization of care systems for adolescents (Millstein, 1988). This is particularly important because adolescents seek care less frequently than do any other age group (Cypress, 1984).

Two variables that showed no statistically significant (.05) between-group changes were body image and depression. The lack of a stronger effect on these two factors is interesting and may reflect the preventive nature of the intervention. Body image may have been poorly measured, making it unclear if this possibility may have caused the nonsignificant results, or if perhaps the intervention in fact produced no effect. With regard to depression, changes in more serious psychological adjustment might not be expected from a preventive intervention. On the other hand, many adolescent girls did report feelings of depression, so perhaps this aspect of the program needs refinement. Also, there may be developmental considerations at play because many girls do not develop depressive symptoms until they reach middle or late adolescence. One study found pronounced increases in depressed moods from the ages of 13 to 15 years, a peak at approximately 17 to 18 years, and then a decline (Radloff, 1991). Furthermore, it is noteworthy that studies on the effectiveness of treatment for depression with adolescents have found stronger effects with older adolescents when compared with middle school adolescents (Weisz, Rudolph, Granger, & Sweeney, 1992). Depression remains an important gender-specific issue that preventive interventions need to address because by age 15, girls are twice as likely as boys to be depressed, a figure that remains unchanged into adulthood (Nolen-Hoeksema & Girgus, 1994).

Overall, this study found positive results and may demonstrate that early adolescent girls can benefit from prevention programs like the Go Grrrls Program in the short run. However, these positive results still represent a modest outcome. There may be a number of reasons for this. The quasi-experimental design always leaves open the possibility that the two groups were not comparable on an important but overlooked factor. Another limitation with this design is that the experimental and control conditions were implemented at the same site. This could introduce threats to the internal validity of the study (Cook & Campbell, 1979). Indeed, a certain amount of "contamination" could have occurred since the girls in the treatment group had opportunities to interact with girls in the comparison group. This may be a realistic threat given the population--middle school early adolescent girls--known for being interpersonally connected. This would have weakened the differences detected between the groups and caused an underestimation of the intervention's impact.

Important methodological limitations may also have played a role in the outcome results. The dependent measures had low reliabilities, thereby reducing some of the power to detect differences between the two groups. Indeed, the body image scale had a low reliability score of .67, which would explain the failure to find any significant differences on this measure. Furthermore, the available power in the experiment was low. Many researchers recommend a power of .80 as a realistic value for the behavioral sciences (Cohen, 1977). The observed power on the Concern for Body Image Scale was a dismal .07, and on the Depression Scale it was .17. One way to cope with this issue is to relax the control of Type I errors as a means of increasing the power of the experiment. Increasing the significance level from .05 to .15, for example, increases the probability of the null hypothesis being rejected because the rejection region is expanded and the power is increased as a consequence (Cohen, 1977). This is why it was suggested that the results on the depression measure might be meaningful even at p < .15. Keppel (1991) notes that researchers may want to "suspend judgment" in such instances, allowing for either a more conservative or liberal judgment on the outcome depending on one's perspective. The effect size, which is less sensitive to small numbers, was .02. While small, this is still considered a meaningful effect size according to Cohen (1977). While this study has found positive results, a lack of follow-up data to determine the persistence of these outcomes over time leaves the longer-term effects of the intervention unknown. In a similar manner, it would be desirable to measure some "hard" outcomes at a long-term follow up, such as dieting behavior, depression rates, career aspirations, and college attendance. Essentially, further research is needed to test the boundaries of the intervention. A more complete measurement package is also needed to further evaluate the potential effectiveness of the intervention. For example, the gender identification measure could not be used because of its low reliability, but the researchers have developed a girls' self-efficacy scale that may be a better outcome indicator for the program (LeCroy & Daley, 2001).

In summary, this study presents an initial evaluation of a gender-specific program for early adolescent girls. The results provide some encouragement for continued study of the potential of this type of intervention; however, a more rigorous design is needed. A larger randomized experiment with a more extensive set of dependent measures has recently been conducted, and these results should provide additional answers to important questions about the effectiveness of a universal gender-specific intervention for early adolescent girls. Future research is needed to assess the long-term and differential impacts of such gender-specific prevention interventions that aim to address a wide range of issues in the adolescent girl's successful transition to adulthood.

Table 1

Means and Standard Deviations for Each Assessment
Variable at Pretest and Posttest

 Pretest Posttest

Variable M SD M SD

Depression
 Intervention Group 31.60 4.9 30.52 4.4
 Control Group 29.93 6.3 28.80 8.4

Body Image
 Intervention Group 10.36 4.1 11.48 3.2
 Control Group 10.97 3.1 11.78 2.7

Peer Esteem
 Intervention Group 24.88 6.1 29.64 8.5
 Control Group 29.09 4.2 30.10 5.7

Seeking Help
 Intervention Group 5.56 2.1 7.08 3.3
 Control Group 4.66 2.6 5.21 2.9

Common Beliefs
 Intervention Group 23.38 4.7 24.50 4.9
 Control Group 26.81 4.7 27.04 5.6

Note. n = 48 to 53.

REFERENCES

Barber, B. L., & Crockett, L. J. (1999). Preventive interventions in early adolescence: Developmental and contextual challenges. In J. Schulenberg, J. L. Maggs, & K. Hurrelmann (Eds.), Health risks and developmental transitions during adolescence. Cambridge: Cambridge University Press.

Birlsen, P. (1981). The validity of depression disorderes in childhood and the development of a self-rating scale: A research report. Journal of Child Psychology and Psychiatry, 22, 73-88.

Brindis, C. D., Irwin, C. E. Jr., Ozer, E. M., Handley, M., Knopf, D. K., & Millstein, S. G. (1997). Improving adolescent health: An analysis and synthesis of health policy recommendations. San Francisco, CA: University of California.

Bush, D. M., & Simmons, R. G. (1988). Gender and coping with entry into early adolescence. In R. Barnett, L. Biener, & G. Baruch (Eds.), Gender, women and stress (pp. 185-217). New York: The Free Press.

Carnegie Corporation (1995). Great transitions: Preparing adolescents for a new century. New York: Carnegie Corporation of New York.

Caspi, A., & Moffitt, T. E. (1991). Individual differences are accentuated during periods of social change: The sample case of girls at puberty. Journal of Personality and Social Psychology, 61, 157-168.

Cauce, A. M. (1986). Social networks and social competence: Exploring the effects of early adolescent friendships. American Journal of Community Psychology, 14, 607-628.

Cicchetti, D., Rappaport, J., Sandler, I. N., & Weissberg, R. P. (Eds.). (2000). The promotion of wellness in children and adolescents. Washington, DC: Child Welfare League of America Press.

Cohen, J. (1977). Statistical power for the behavioral sciences (rev. ed.). New York: Academic Press.

Coie, J. D., Watt, N. F., West, S. G., Hawkins, J. D., Asarnow, J. R., Markman, H. J., Ramey, S. L., Shrue, M. B., & Long, B. (1993). The science of prevention: A conceptual framework and some direction for a national research program. American Psychologist, 48, 1013-1022.

Conger, A. J., Peng, S. S. & Dunteman, G. H. (1977). National longitudinal study of the high school class of 1972: Group profiles on self-esteem, locus of control, and life goods. Research Triangle Park, NC: Research Triangle Institute.

Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Chicago: Rand McNally.

Cypress, B. K. (1984). Health care of adolescents by office-based physicians: National ambulatory care survey (DHHS, NCHS Publication No. 99). Washington, DC: U.S. Government Printing Office.

Denmark, F. L. (1999). Enhancing the development of adolescent girls. In N. G. Johnson, M. C. Roberts, & J. Worell (Eds.), Beyond appearance: A new look at adolescent girls (pp. 377-404). Washington, DC: American Psychological Association.

Dryfoos, J. G. (1998). Safe passage: Making it through adolescence in a risky society. New York: Oxford University Press.

Eccles, J. S., Jacobs, J., Harold, R. D., Yoon, K. S., Arbreton, A. J. A., & Freedman-Doan, C. (1993). Parents and gender role socialization. In S. Oskamp & M. Costanzo (Eds.), Claremont symposium on applied social psychology, 1992: Gender and social psychology (pp. 59-84). Thousand Oaks, CA: Sage.

Graber, J. A., Brooks-Gunn, J., & Petersen, A. C. (1996). Adolescent transitions in context. In J. A. Graber, J. Brooks-Gunn, & A. C. Petersen (Eds.), Transitions through adolescence: Interpersonal domains and context. Mahwah, NJ: Erlbaum.

Hamburg, D. A., & Takanishi, R. (1989). Preparing for life: The critical transition of adolescence. American Psychologist, 44, 825-827.

Hare, B. R. (1985). The Hare General and Specific (School Peer, and Home) Self-esteem Scale. Unpublished manuscript.

Hooper, S. R., & Layne, C. C. (1983). The common belief inventory for students: A measure of rationality in children. Journal of Personality Assessment, 47, 85-90.

Johnson, S. B., & Millstein, S. G. (2003). prevention opportunities in health care settings. American Psychologist, 58, 475-481.

Kazdin, A. E., & Weisz, J. R. (1998). Identifying and developing empirically supported child and adolescent treatments. Journal of Consulting and Clinical Psychology, 66, 19-36.

Kenny, M. E., & Donaldson, G. A. (1991). Contributions of parental attachment and family structure to the social and psychological functioning of first-year college students. Journal of Counseling Psychology, 38, 479-486.

Keppel, G. (1991). Design and analysis: A researcher's handbook (3rd ed.). Englewood Cliffs, NJ: Prentice Hall.

LeCroy, C. W., & Daley, J. (2001). Empowering adolescent girls: Examining the present and building skills for the future with the Go Grrrls Program. New York: W. W. Norton.

Millstein, S. G. (1988). The potential of school-linked centers to promote adolescent health and development. Washington, DC: Carnegie Council on Adolescent Development.

Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender differences in depression during adolescence. Psychological Bulletin, 115, 424-443.

Notaro, P. C., Miller, A. L., & Zimmerman, M. A. (1998). Adolescent attachment to friends: Relationships with mental health and gender. Paper presented at the meeting of the Society for Research in Adolescence, San Diego, CA.

Pipher, M. (1994). Reviving Ophelia: Saving the selves of adolescent girls. New York: Ballantine Books.

Radloff, L. S. (1991). The use of the Center for Epidemiological Studies Depression scale in adolescents and young adults. Journal of Youth and Adolescence, 20, 149-166.

Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E., Harris, K. M., Jones, J., Tabor, J., Beuhring, T., Sieving, R., Shew, M., Ireland, M., Bearinger, L. H., & Richard, U. J. (1997). Protecting adolescents from harm: Findings from the National Longitudinal Study of Adolescent Health. Journal of the American Medical Association, 278, 823-832.

Rosner, B. A., & Rierdan, J. (1994, February). Adolescent girls' self-esteem: Variations in development trajectories. Paper presented at the meeting of the Society for Research on Adolescence, San Diego, CA.

Simmons, R. G., & Blythe, D. A. (1987). Moving into adolescence: The Impact of pubertal change and school context. New York: Aldine De Gruyter.

Stewart, A. J. (1982). The course of individual adaptation to life changes. Journal of Personality and Social Psychology, 46, 143-151.

Weissberg, R. P., Kumpfer, K. L., & Seligman, M. E. P. (2003). Prevention that works for children and youth. American Psychologist, 58, 425-432.

Weisz, J. R., Rudolph, K. D., Granger, D. A., & Swereney, L. (1992). Cognition, competence, and coping in child and adolescent depression: Research findings, developmental concerns, therapeutic implications. Developmental Psychopathology, 4, 627-653.

Craig Winston LeCroy, School of Social Work, Arizona State University.

Requests for reprints should be sent to Craig W. LeCroy, Tucson Component, School of Social Work, Arizona State University, 340 N. Commerce Park Loop, Tucson, AZ 85745. E-mail: craig.lecroy@asu.edu, web page: http:/w/ww.public.asu.edu/~lecroy/