Psychosocial Factors Associated With Youth Involvement in Community Activities Promoting Heart Health

Psychosocial Factors AssociatedWith Youth Involvement in Community

Activities Promoting Heart Health

David G. Altman, PhDEllen Feighery, MS, RN

Thomas N. Robinson, MD, MPHK. Farish Haydel

Laura Strausberg, MEdKate Lorig, RN, DrPHJoel D. Killen, PhD

This study examined factors that influence youth participation in heart disease prevention activities among2,609 ninth graders in six inner-city public high schools. Constructs derived from social cognitive, empower-ment, and community development theories informed the conceptual framework employed. Study participantswere diverse with respect to gender, ethnicity, parent education, acculturation, and academic achievement.Perceived incentive value, self-efficacy, outcome expectancies, sense of community, and perceived policycontrol were all significantly associated with participation in community activities promoting heart health. Inmultivariate analyses, perceived incentive value, defined as the extent to which participants valued a heart-healthy environment, was most strongly associated with community participation, accounting for 11.9% of thetotal variance. These findings have implications for designing school curricula and after-school and communityprograms targeting adolescents' involvement in health advocacy activities.

David G. Altman, PhD, is a professor in the Department of Public Health Sciences, Section on SocialSciences and Health Policy, at Wake Forest University School of Medicine. Ellen Feighery is a research as-sociate at the Stanford Center for Research in Disease Prevention at Stanford University School of Medicine.Thomas N. Robinson is an assistant professor in the department of pediatrics at Stanford University School ofMedicine and an American Heart Association Clinician-Scientist. K. Farish Haydel is a statistical programmerin the Department of Medicine, Stanford University School of Medicine. Laura Strausberg is a social scienceresearch assistant at the Stanford Center for Research in Disease Prevention, Stanford University School ofMedicine. Kate Lorig is an associate professor at the Stanford University School of Medicine and the directorof the Stanford Patient Education Research Center. Joel D. Killen is an associate professor in the Departmentof Medicine, Stanford University School of Medicine.

Address reprint requests to David G. Altman, PhD, Department of Public Health Sciences, The Wake ForestUniversity School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1063; phone: (336)716-9556; fax: (336) 716-7554; e-mail: [email protected]

The study was supported by grants ROl NR03597-01 and ROI HL32185 from the National Institutes ofHealth and grant 95-19 from the California Wellness Foundation. This work was completed during the tenureof an American Heart Association Clinician-Scientist Award to Dr. Robinson. The authors thank the studentsand staff at the six schools for their participation in this project.Health Education & Behavior, Vol. 25 (4): 489-500 (August 1998)© 1998 by SOPHE

489

by guest on April 25, 2016heb.sagepub.comDownloaded from

http://heb.sagepub.com/

490 Health Education & Behavior (August 1998)

INTRODUCTION

Cardiovascular diseases (CVD) are the leading cause of death and disability in theUnited States. Epidemiological research demonstrates that tobacco use, high dietary fatintake, and physical inactivity contribute to the development of CVD and that thesebehaviors are acquired early in life.`13 Therefore, much effort has been focused ondeveloping school-based health promotion programs aimed at preventing or reducingCVD risk factors among children and adolescents. These programs are designed to helpyoung people acquire positive health skills that enable them to counter social-environ-mental factors promoting risky behaviors. Programs typically include multifacetedapproaches that combine health education and behavioral skills training.

Controlled evaluations indicate that school-based health programs can produce sig-nificant knowledge and behavioral change in the short term but that effects decay overtime.4` This is true for even the more recent school-based efforts that have included afocus on environmental and policy change.4 The lack of sustained effects may stem, inpart, from the fact that those targeted by behavior change efforts are seldom involveddirectly in changing school or community norms.8`0 To achieve durable behavior change,it may be important for youths to participate in activities that help shape the social contextthat influences their individual behavior."'12 Participatory programs afford young peoplethe opportunity to articulate their own interests and to play an active role in how anintervention addresses these interests.13

Recently, there has been renewed interest in fostering community participation incommunity activities and institutions related to health, although efforts to promote youthparticipation have largely targeted college students and young adults. For example, in1993, the Clinton Administration committed $100 million to start a major national serviceprogram, AmeriCorps, that facilitated college student participation in an array of com-munity institutions and activities.14 Similarly, religious institutions and communityservice clubs (e.g., Lions, Rotary) have a long history of sponsoring programs to increasecitizen participation in community activities. Less attention, however, has been given tosystematic efforts to encourage adolescents to participate in community activities, andthere are few empirical studies that identify factors that predict community participationby teenagers.'"6 In addition, most civic participation or service-learning programs focuson increasing community service (or attitudes toward service) as the primary goal ratherthan viewing community participation as a conduit to behavior change.'4"17'19

Indeed, some have argued that direct community involvement may lead to betteracquisition and maintenance ofhealthful behaviors than more traditional approaches builton didactic educational models.8'0 Direct involvement may influence individual healthbehavior by (1) increasing participants' commitment to the behavior change program (i.e.,ensuring intervention exposure); (2) increasing opportunities for the acquisition, rein-forcement, and maintenance of skills and behavior; and (3) providing opportunities forparticipants to make visible, public commitments to healthy lifestyles (i.e., increasingmotivation to comply with intervention recommendations).

Before the effects of community participation on individual health behavior can beadequately assessed, however, we must develop a better understanding of the factors thatinfluence direct involvement. To date, there has been only a small number of studies thathave examined these factors among adolescents. In this study, we undertook to identifythe factors that may influence youth participation in heart disease prevention activities.The adult literature on participation has identified some of the key factors that areassociated with citizen involvement in community activities.20'2 In part, these studies



Altman et al. / Factors Associated With Youth Involvement 491

have shown that community participation is associated with perceived self-efficacy inengaging in community activities and beliefs that community change is desirable and canbe achieved.2325 Likewise, perceived policy control, sense of community, and empower-ment have been found to be associated with adult involvement in community change.26-28The adult literature on community participation has influenced the current study in thatsimilar theoretical perspectives and constructs have been adapted for use with teenagers.

Conceptually, this study was guided by constructs derived from social cognitive theory(perceived incentive value, outcome expectancies, and perceived self-efficacy) and byconstructs from empowerment and community development theories (particularly per-ceived policy control and sense of community). We are unaware of other studies that havemeasured these constructs under the rubric of CVD prevention with high school students.

METHODS

Participants

Ninth graders attending six San Jose, California, public high schools were eligible toparticipate in a controlled trial of a school-based intervention to reduce cardiovasculardisease risk factor behaviors (N= 2,609). The present analysis used baseline data collectedfrom students in all six schools prior to the start of the intervention. Students notparticipating in mainstream classes (e.g., Special Education) and those with limitedEnglish language proficiency were ineligible, as all survey materials were in English.Trained research staff supervised the completion of self-administered surveys duringregular classroom periods over 2 days in each school. A third day was used for makeupsdue to absences and to finishing incomplete surveys.

Surveys in all six schools were completed between October 1994 and December 1994.Confidentiality was maintained by using unique identification numbers. Each surveycontained two cover sheets, the first with a label printed with the student's name andidentification number and the second with a label containing only the identificationnumber. Students were instructed to remove the top cover sheet once they received thesurvey so that only the cover sheet with the identification number alone remained. Parentswere informed of the study by mail and given the opportunity to withdraw their childfrom participation (i.e., passive parental consent). Only seven parents refused to allowtheir child to participate. Students were also given the opportunity to decline participationat any time before or during the assessment-eight students declined to participate. Theprotocol was approved by the Stanford University Committee for the Protection ofHuman Subjects in Medical Research.

Table 1 summarizes demographic characteristics of the study participants. Studentsreported their age by circling the month, day, and year of their birthday, and their genderby circling male or female. Ethnicity was reported by circling one of 21 descriptors,including "other" and an option for students who "strongly considered" themselves to

belong to more than one ethnic group. Students who considered themselves to belong to

more than one ethnic group circled up to three of the descriptors. For purposes of analysis,more specific ethnic categories were collapsed into five broader ethnic categories: Latino,Asian/Pacific Islander, white, African American, and Other (including Native American,Multiethnic, and all others). Acculturation was assessed by using three factors to estimateeach student's level of exposure and involvement in mainstream U.S. culture: primary




Table 1. Demographic Characteristics

Characteristic Percentage

Age (M= 14.69; SD = .51)Gender (female) 49Ethnicity

Latino (n = 1,003) 38Asian (n =784) 30White (n =461) 18African American (n = 153) 6Other (n = 185) 8

Born in the U.S. 77English as main language spoken at home with parents 65AcculturationLow 41High 59

Parent educationLess than high school 11High school graduate 17Post-high school education 61Don't know 11

Usually receives As and Bs in school 41

language spoken at home with parents, country of birth, and number of years living inthe United States. Students reporting English as the main language spoken at home withtheir parents, being born in the United States, and living in the United States for morethan 10 years were considered highly acculturated. All others were considered lessacculturated. Parent education was assessed separately for both mother and father.Students responded to the question, "What level of education does your [mother/father](or most important [woman/man] who is responsible for you) have?" Responses werecategorized as don't know, less than high school, high school graduate, and post-highschool education. The higher of the two parent reports was used to classify parenteducation. School performance was assessed by asking students to circle their usualschool grades along the following dimensions: mostly As, mostly As and Bs, mostly Bs,mostly Bs and Cs, mostly Cs, mostly Cs and Ds, mostly Ds, mostly Ds and Fs, or mostly Fs.

The data in Table 1 illustrate that the sample was diverse with respect to gender (49%female), ethnicity (about one-third Latino, one-third Asian, and one-third other), parenteducation (about one-third high school graduate or less and about two-thirds post-highschool education), level of acculturation (41% low acculturation), and academic achieve-ment (41% reported usually getting As or Bs).

Measures

Formative Evaluation and Reliability Testing. The pool of items used for studymeasures was adapted from instruments used in previous studies either by the Stanfordteam (e.g., social cognitive items) or from instruments developed by others (e.g.,perceived policy control, sense of community). The instruments were then pilot-tested in




Table 2. Descriptive Data of Independent and Dependent Variables

Spearman Cronbach'sConstruct M SD Correlation Alpha

Community participation (1 = low, 5 = high) 1.67a 0.56 .68 .88Sense of community (1 = low, 5 = high) 3.45 0.61 .79 .80Self-efficacy (1 = low, 5 = high) 2.91 0.96 .58 .77Outcome expectancies (1 = low, 5 = high) 2.74 0.82 .63 .73Perceived incentive value (1 = low, 5 = high) 2.83 0.88 .70 .82Perceived policy control (1 = low, 5 = high) 2.86 0.78 .60 .70

a. Only 8.2% of the students reported no participation at all. The majority of students whoparticipated in community activities reported three or fewer activities.

several focus groups of local high school students. Individual items were modified ordropped based on feedback from students in these focus groups. Once data were collected,items with low reliability were dropped from further analysis.

Test-retest reliability on the primary variables (over a 24-hour period) were calculatedas kappa coefficients for categorical variables and Spearman correlation coefficients forscaled variables. Reliability was assessed by giving students randomly selected instru-ments that were completed either a day before or a day after the questions were answeredon the main survey. Instruments completed first were used in all subsequent analyses.Table 2 presents descriptive data on these variables. Test-retest (Spearman correlation)and internal consistency reliability (Cronbach's alpha) suggest that the constructedindexes in this sample were reliable and had approximately normal distributions.

The following section provides some background information on each measure.

Perceived Policy Control. This construct measures perceptions of individual controlor influence on policy makers. The perceived policy control instrument was adapted foruse with teenagers from a subscale of Zimmerman and Zahniser's Sociopolitical controlscale for adults.29 The instrument contained four items, and responses were made on a5-point Likert-type scale (e.g., "so many other people are active in local teen issues thatit doesn't matter much whether I participate or not"; "sometimes government seems socomplicated that I can't really understand what's going on").

Perceived Self-efficacy. An instrument was developed to measure perceived self-effi-cacy, the perception that one can successfully execute a specific behavior required toachieve an outcome.23 30 This four-item instrument asked students to record responses ona 5-point scale with responses ranging from I cannot do it to I definitely can do it andmeasured participants' perceived self-efficacy to perform behaviors to modify the envi-ronmental influences related to tobacco use, physical activity, and nutrition (e.g., "Howsure are you that you can write a letter to the executives of a TV network requesting thatthey advertise healthier foods when young people are watching TV [for example,Saturday morning]?" "How sure are you that you can circulate a petition and get 500signatures stating that you want a particular exercise class offered through your city'srecreation department?").

Perceived Incentive Value. The perceived incentive value subscale was developed tomeasure the extent to which participants valued a heart-healthy environment (e.g., "How




Table 3. Correlation Matrix Among Independent and Dependent Variables

Construct CP SOC SE OE PIV PPC

Community participation (CP) .12 .27 .24 .38 .12Sense of community (SOC) .07 .08 .14 .00Self-efficacy (SE) - .41 .42 .32Outcome expectancies (OE) .35 .23Perceived incentive value (PIV) .26Perceived policy control (PPC)

important is it to you that your community has a variety of physical or recreationalactivities available to people under 18?" "How important is it to you that there are nationalpolicies [rules] against advertising and promoting tobacco products to people under18?"). This subscale contained six items on a 5-point scale from not at all important tome to extremely important to me.

Outcome Expectancies. Outcome expectancies are perceptions of the likely outcomesof specific behaviors.23'30 This four-item instrument was developed to measure beliefsthat participating in particular health-related advocacy activities would result in changesin the heart healthfulness of the surrounding environment (e.g., "You circulate a petitionand get 500 signatures stating that you want a particular exercise class offered throughyour city's recreation department. How likely is it that the petition will result in therecreational department offering an exercise class?" "You tell the people in your homewho do the food shopping that you would like them to switch from buying whole milkto buying low-fat milk. How likely is it that your request will result in the purchase oflow-fat milk?"). Responses were made on a 5-point scale ranging from not at all likelyto extremely likely.

Sense of Community. Sense of community has been considered part of the affectivedomain of neighboring.31 Sense of community is related to perceptions of group belong-ing, commitment, and loyalty.32 Based on this theoretical perspective, we developed anine-item instrument that tapped four key elements: membership, influence, sharedvalues, and shared emotional connection. Questions were asked on a 5-point Likert-typescale and included the following: "I think my neighborhood is a good place to live"; "Ifthere is a problem in my neighborhood, people who live there can get it solved"; and"Most people who live in my neighborhood would be able to tell ifsomeone was a strangeror if he or she lived in the neighborhood."

Community Participation. The community participation instrument was developed tomeasure students' involvement in health-related advocacy over the past year. We definedcommunity advocacy or participation as activities that were directed at building heart-healthy environments. The instrument included 15 items, each measured on a 5-pointscale. The questions asked students to report how many times in the past year of studythey had engaged in an activity related to tobacco, nutrition, and physical activity (e.g.,"How many times in the past year have you signed a petition on an issue related tosmoking?"). The response choices ranged from never to six or more times.




Table 4. Stepwise Multiple Regression With Community Participation as the Dependent Variablea

Independent Variable Partial R2 Total R2 p Value

Perceived incentive values .119 .119 < .0001Outcome expectancies .009 .129 <.0001African American versus white ethnicity .003 .132 <.005Perceived self-efficacy .003 .135 <.006Sense of community .002 .138 < .02Acculturation (high) .003 .140 < .02School grades (higher) .003 .143 < .02

a. We also ran a logistic regression analysis using a median split on participation (lower participa-tion, higher participation). The logistic regression analysis yielded findings similar to the stepwisemultiple regression analysis reported above except that Asian ethnicity was significant, andacculturation and school grades were not.

StatisticalAnalysis. Both univariate (Spearman correlations) and multivariate analyses(stepwise multiple linear regression) were completed.

RESULTS

Univariate Analysis

To examine the associations between the independent variables and communityparticipation, Spearman correlations were computed. Moderate and statistically signifi-cant correlations (.12 to .38) were found between the various independent variables andcommunity participation (see Table 3). The independent variables derived from socialcognitive theory (e.g., self-efficacy, outcome expectancies, perceived incentive value)were also moderately and significantly correlated with each other (.35 to .42).

Multivariate Analysis

Table 4 presents data from a stepwise multiple linear regression using communityparticipation as the dependent variable andp < .05 as the maximum significance level forentry into the model. Overall, the entire model accounted for 14.3% of the variance incommunity participation. Seven independent variables entered the regression equation.Perceived incentive value was the first independent variable to enter the model, account-

ing for 11.9% of the total variance (more perceived incentive value, more communityparticipation).

DISCUSSION

This study is perhaps the first to examine factors that are associated with youthparticipation in community CVD prevention activities. The findings have implications




for the design of school- and community-based CVD interventions. We found that amongthe independent variables, perceived incentive value, defined as the importance or valuethat students placed on having a heart healthy environment, was most strongly associatedwith community participation. That is, there was a positive association between students'perceptions of the importance of community change outcomes and actual participation.This finding is consistent with the emerging societal trends of melding values intoeducation and a renewed emphasis on citizenship skills. Conveying a strong rationale forcommunity participation to students may tap into students' wishes to live in a healthiercommunity and may be a prerequisite to actual community participation. Although themean for perceived incentive value (M = 2.8, SD = .88) indicates that a ceiling effect wasnot observed, it suggests that there are some students, preintervention, who value theimportance of heart-health policies. This may be a result of aggressive health educationefforts in California schools and in the larger media environment or a reflection of thenorms around health among California residents.

Perceived policy control and sense of community have also been found to influenceadult involvement with community change efforts. That is, when adults believe that theyinfluence a policy and when they perceive a connection to their neighborhood or

26-28community, they are more likely to expend energy and time on community activities.In the multiple regression analysis in the current study, sense of community, but notperceived policy control, was significantly associated with community participation,although the magnitude of the effect was small. Two other constructs derived from socialcognitive theory, self-efficacy for participation and outcome expectancies regarding theeffects of community participation, were also associated with reported communityparticipation. These constructs, along with perceived incentive value, form a coherent setof variables that could be melded into youth education programs.

Implications for Practice

It is not difficult to imagine how designers of school curricula, after-school programs,or community programs for young people could capitalize on these data. The findingsfrom this study could be incorporated into the classroom by building into existing lessonplans lectures on the historical and present-day importance of citizen participation andthe role that environmental influences play in health, the importance of participating inthe community, and skills development in such participation (e.g., this could be done inhealth, civics, or history classes). Likewise, teachers could facilitate opportunities foractual participation through homework assignments and extra credit projects. At thepolicy level, school policy could reinforce student participation in the community bylinking students to community organizations; establishing after-school community ser-vice clubs; and highlighting in school publications, community lectures, and the mediastudents who participate in the larger community. Unfortunately, more often than not,students taking health-related classes have been passive recipients of programs designedby adults, rather than active participants of programs that involve them directly in thecommunity. Opportunities for incorporating our findings into school programs could wellresult in more favorable student evaluations of curricula, improvements in the communityenvironment, and perhaps even the practice of more healthy behaviors by studentsreceiving interventions. Our findings suggest that these areas should be a priority forfuture research and practice.




We found that community participation will not be as likely to occur unless studentssee the value of attempting to shape the larger environment in which they live. Thus, itseems important for school and community interventionists to consider the school wallsas extending beyond the physical structure of the school into the community at large.Furthermore, in light of increasing concern about the "character" of young people, theremay well be widespread community support for programs that provide students withopportunities to help build better places to live, through facilitating their active involve-ment in community projects and activities.

To attempt to increase community participation around CVD prevention, school-basedCVD prevention programs could specifically increase the value that students place onbuilding heart-healthy environments (incentive value), increase student perceptions thattheir involvement in the community will have desirable effects (outcome expectancies),and provide skills training and mastery opportunities to increase students' confidence thatthey can successfully participate in community activities (self-efficacy). We also foundthat ethnicity (being African American), acculturation (high), and school grades (bettergrades) were independently associated with community participation. Although statisti-cally significant, the variance accounted for by these demographic factors (less than 1%)was substantially smaller than the variance accounted for by the social cognitive variables(13%).

Although we are optimistic that youth participatory interventions can be developedand implemented successfully, we recognize that these interventions may present uniquechallenges to teachers. Based on research with adults, we would expect that, like students,teachers who see value in student participation in community activities are most suitedfor this type of intervention. It is reasonable to hypothesize that teachers who havepersonally engaged in community participation or who are confident in their ability to doso will likely be more effective teachers than teachers who have not had these experiencesor hold these attitudes. This suggestion, however, has not been tested.

Participatory interventions require teachers to combine didactic and participatoryeducational approaches. They also require contact with community organizations so thatstudent participation in community activities can be arranged. We recognize that teachersare increasingly being asked to add innovations to already packed curricula and that theapproach proposed here may be asking too much of some teachers or underresourcedschool systems. However, if participatory interventions result in higher academicachievement, or at least greater motivation among students to attend class and to

participate in school activities, the added burden of incorporating new innovations intothe classroom may be offset by a more engaged and participatory student body. Futureresearch needs to examine the costs and benefits for students, teachers, school systems,and the community at large of incorporating participatory programs into the classroom.

Limitations

There are a few caveats to the interpretation of the findings that deserve mentioning.First, since the data were cross sectional, it is conceivable that the order of the independentand dependent variables should be reversed. That is, community participation could beantecedent to perceived incentive value, outcome expectancies, and self-efficacy. If theorder were reversed, one could conclude that students who participated in the communitywere more likely than those who did not to perceive value in participating, believe that




participating would make a difference, and be confident in their participatory skills. Thisissue can only be unraveled in subsequent longitudinal research. Second, while there wereseven independent variables that were statistically significant in the multiple regressionanalysis, one (perceived incentive value) accounted for the bulk of the explained variancein community participation. This finding may be due in part to the fact that incentivevalue, relative to the other independent variables in the regression, had higher test-retestand internal consistency reliability. Future development of similar instruments shouldstrive for higher test-retest reliabilities. In addition, other significant variables in theregression that are correlated with perceived incentive value may represent constructsthat could have been equally or more highly associated with community participation hadthey been measured better. This finding suggests that decision makers should considerspending more intervention time and resources on perceived incentive value versus othermodifiable influences. Also, as noted in Table 4, we analyzed the data with both linearand logistic regression. For all of the key social cognitive variables, the results weresimilar. We reported the linear regression results because the data were not highly skewedand it allowed us to define participation along a continuum, the latter of which isconsistent with our conceptual framework.

In summary, this study illuminates a set of theoretically derived variables that wereassociated with student participation in community-based CVD activities. This studycontributes important data to the measurement of these constructs among youths. Spe-cifically, we found that the key variables in our theoretical framework could be measuredwith acceptable internal consistency and test-retest reliability among a sociodemographi-cally diverse population of high school students. These variables, derived from socialcognitive, community development, and empowerment theories, can be incorporated intothe design of school and community interventions that strive to increase youth participa-tion in community health-related activities. This study is consistent with the growingadult literature on citizen participation, which has found that adults are more likely topursue community change when they believe that change is worthwhile and achievableand that they have the skills to achieve desired changes.'O 21,24,25,33 Data reported here helpidentify a set of variables that are associated with community participation among youngpeople. To further support the rationale for community participation interventions foryouths, we suggest that subsequent work assess links longitudinally between thesevariables, community participation, and actual health behaviors.34

References

1. Allen R: Is coronary heart disease a lifestyle disorder? Areview ofpsychological and behavioralfactors, I. Cardiovasc Rev Rep 13:13-53 (nonsequential), 1992.

2. Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas AM, Pajak A: Mycardialinfarction and coronary deaths in the World Health Organization MONICAProject. Circulation90: (1) 583-612, 1994.

3. National Heart, Lung, and Blood Institute: Report ofthe Task Force on Research in Epidemiol-ogy and Prevention ofCardiovascular Diseases. Washington, DC, U.S. Department of Healthand Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute,1994.

4. Resnicow K, Robinson T, Frank E: Advances and future directions for school-based healthpromotion research: Commentary on the CATCH intervention trial. Prev Med 25:378-383,1996.




5. Stone EJ, Perry CL, Luepker RV: Synthesis of cardiovascular behavior research for youth healthpromotion. Health Educ Q 16:155-169, 1989.

6. Rundall T, Bruvold W: A meta-analysis of school-based smoking and alcohol use preventionprograms. Health Educ Q 15:317-334, 1988.

7. Luepker RV, Perry CL, McKinlay SM, et al: Outcomes of a field trial to improve children'sdietary patterns and physical activity. JAMA 275:768-775, 1996.

8. Schmid TL, Pratt M, Howze E: Policy as intervention: Environmental and policy approachesto the prevention of cardiovascular disease. Am J Public Health 85:1207-1211, 1995.

9. Green LW, Kreuter MW: Health Promotion Planning: An Educational and EnvironmentalApproach. Mountain View, CA, Mayfield, 1991.

10. Green LW: The theory of participation: A qualitative analysis of its expression in national andinternational health policies. Adv Health Educ Health Promot 1:211-236, 1986.

11. Kurth-Schai R: The roles of youth in society: A reconceptualization. Educ Forum 52:113-131,1988.

12. Millstein SG, Nightingale EO, Peterson AC, Mortimer AM, Hamberg DA: Promoting thehealthy development of adolescents. JAMA 269:1413-1415, 1993.

13. Pittman K: A framework for defining and promoting youth participation. Future Choices:Toward Natl Youth Policy 3:85-90, 1991.

14. Milton C: National service: Secret weapon in the fight against crime. Public Manage July, 1994,pp. 6-10.

15. Price RH: Wither participation and empowerment?AmJ Community Psychol 18:163-167, 1990.16. Zeldin S: Opportunities and supports for youth development: Lessons from research and

implications for community leaders and scholars. Washington, DC, Center for Youth Devel-opment and Policy Research, Academy for Educational Development, 1995.

17. Calabrese RL, Schumer H: The effects of service activities on adolescent alienation. Adoles-cence 21:675-687, 1986.

18. Batchelder TH, Root S: Effects of an undergraduate program to integrate academic learningand service: Cognitive, prosocial cognitive, and identity outcomes. JAdolesc 17:341-355, 1994.

19. Giles DE, Eyler J: The impact of college community service laboratory on students' personal,social, and cognitive outcomes. JAdolesc 17:327-339, 1994.

20. Israel BA, Checkoway B, Schulz A, Zimmerman M: Health education and communityempowerment: Conceptualizing and measuring perceptions of individual, organizational, andcommunity control. Health Educ Q 21:149-170, 1994.

21. Zimmerman MA: Psychological empowerment: Issues and illustrations. Am J CommunityPsychol 23:581-599, 1995.

22. Florin P, Wandersman A: Cognitive social learning and participation in community develop-ment. Am J Community Psychol 12:689-708, 1995.

23. Bandura A: Self-Efficacy: The Exercise of Control. New York, Freeman, 1997.24. Maddux J, Norton L, Stoltenberg C: Self-efficacy expectancy, outcome expectancy, and

outcome value: Relative effects on behavioral intentions. JPers Soc Psychol 51:783-789, 1986.25. Rappaport J: Terms ofempowerment/exemplars of prevention: Toward a theory for community

psychology. Am J Community Psychol 15:121-148, 1987.26. Chavis DM, Wandersman A: Sense of community in the urban environment: A catalyst for

participation and community development. Am J Community Psychol 18:55-81, 1990.27. Schulz AJ, Israel BA, Zimmerman MA, Checkoway BN: Empowerment as a multi-level

construct: Perceived control at the individual, organizational, and community levels. HealthEduc Res 10:309-327, 1995.

28. Zimmerman MA, Rappaport J: Citizen participation, perceived control, and psychologicalempowerment. Am J Community Psychol 16:725-749, 1988.

29. Zimmerman MA, Zahniser JH: Refinements of sphere-specific measures of perceived control:Development of a socio-political control scale. J Community Psychol 19: 189-204, 1991.

30. Bandura A: Social Foundations ofThought andAction: A Social Cognitive Theory. EnglewoodCliffs, NJ, Prentice Hall, 1986.




31. Unger DG, Wandersman A: The importance of neighbors: The social, cognitive and affectivecomponents of neighboring. Am J Community Psychol 13:139-169, 1985.

32. McMillian DW, Chavis DM: Sense of community: A definition and theory. J CommunityPsychol 14:6-23, 1986.

33. Florin P. Wandersman A: An introduction to citizen participation, voluntary organizations, andcommunity development: Insights for empowerment through research. Am J CommunityPsychol 18:41-54, 1990.

34. Zimmerman MA, Maton KI: Life-style and substance use among male African-American urbanadolescents: A cluster analytic approach. Am J Community Psychol 20:121-138, 1992.



Psychosocial Factors Associated With Youth Involvement in Community Activities Promoting Heart Health

Documents