SEX DIFFERENCES IN THE ASSOCIATION BETWEEN RISK AND PROTECTIVE FACTORS AND SELF REPORTED SERIOUS DELINQUENCY. Abigail Fagan1, M. Van Horn2, J. David Hawkins3, Michael Arthur3, 1Social Development Research Group, Seattle, WA United States; 2University of South Carolina, Columbia, SC United States; 3University of Washington, Seattle, WA United States
There is some debate regarding whether risk and protective factors are related to serious delinquency in the same way for females and males. While prevention science generally indicates that the same risk and protective factors are salient for both sexes (Hawkins, Herrenkohl et al. 1998), many feminist criminologists argue that different factors may explain male and female offending (Daly and Chesney-Lind 1988; Chesney-Lind 1997; Silverthorne and Frick 1999; Steffensmeier and Broidy 2001). It is important to resolve this issue not only to help explain the gender gap in offending (i.e., that males are much more likely than females to be involved in serious crime), but also to ensure that prevention services are aimed at those most likely to be influenced by risk and protective factors. This investigation uses data from the Communities that Care youth survey (Arthur et al., 2002; Glaser et al, in press), completed by nearly 8,000 10th grade students across the United States, to examine if males and females differentially experience and are differentially affected by risk and protective factors for serious delinquency.
The results clearly demonstrate that family, school, peer group, and individual risk and protective factors are significantly related to serious delinquency for girls as well as boys. Some sex differences are evidenced. For most outcomes, boys report higher levels of risk and lower levels of protection. Further, the influence of about half of the factors on serious offending is greater for young men than young women. These findings suggest that the gender gap in offending is due to a combination of males facing higher levels of risk and lower levels of protection, and their experiencing a greater association between risk and protective factors and serious delinquency.
Nonetheless, most of the sex differences are relatively modest in size, suggesting more similarities than differences regarding the relationship between sex, risk and protective factors, and serious delinquency. Thus, prevention programs that reduce risk factors and enhance protective factors should target both sexes for services, in order to lessen the likelihood that girls and boys will become involved in serious offending.
INTIMATE PARTNER VIOLENCE IS RELATED TO HIV RISK BEHAVIORS. Elizabeth Harris1, Bellamy Nikki2, 1Evaluation, Management & Training Associates, Inc., Encino, CA United States; 2Substance Abuse & Mental Health Services Administration, Rockville, MD United States
Background: This paper is based on data collected as part of a national study of a federally-funded HIV and substance abuse prevention initiative that targets minority populations. The data were analyzed to investigate relationships between high-risk sexual behaviors and intimate partner violence (IPV). The data are for 461 adults (76% female) at 12 study sites. Anecdotal reports from the community-based prevention programs suggest that many individuals who are at risk for HIV infection are also in abusive relationships. The analysis tested the hypothesis that recent IPV is a predictor of engaging in risky sexual behaviors.
Methods: Data were collected locally using a common instrument when the subjects entered the study. The independent variables are emotional, physical, and sexual abuse within the past 3 months. The dependent variables are risky sexual behaviors during the past 3 months. Logistic regression analysis is used.
Results: Emotional abuse was reported by 40% of the respondents, physical abuse by 17%, and sexual abuse by 8%. Reports of risky sexual behaviors include, unprotected oral sex (27%), unprotected vaginal sex (40%), unprotected anal sex (23%), multiple sexual partners (21%), and sex while under the influence of drugs or alcohol (25%). The analysis indicates that individuals who have been abused recently also engage in risky sexual behaviors. For example, the odds ratio for having multiple sex partners is 2.10 (P<.018) if a person has been physically abused, and 2.28 (P<.047) if a person has been sexually abused. The odds ratio for having sex while under the influence of drugs or alcohol is 3.06 (P<.000) if a person has been physically abused. Emotional abuse does not predict risky sexual behaviors.
Conclusion: The findings support the anecdotal field reports that IPV is a risk factor for HIV-related sexual behaviors. HIV prevention and treatment providers should consider offering, as part of their service mix, services for IPV.
DEVELOPING A COMPREHENSIVE ETIOLOGICAL FRAMEWORK FOR ALCOHOL, TOBACCO, AND OTHER DRUG ATTITUDES AND USE IN RURAL TEENS. Larissa G. Duncan1, Mark E. Feinberg1, Mark T. Greenberg1, 1Pennsylvania State University, University Park, PA United States
Risk factors for substance use and externalizing problem behavior include a variety of individual, family, school, peer, and community factors. Research has demonstrated that the combined presence of multiple risk factors has a synergistic effect on increasing negative outcomes. However, prevention program developers require a more comprehensive and detailed etiological framework than presently available. For example, do certain domains of risk factors (e.g. individual) mediate the influence of other domains (e.g. school, community)? Do certain domains of risk factors account for a large proportion of unique variance in substance use? Answering such questions requires a large dataset with multi-method data at various levels. In addition, there is a particular need to ask these questions for rural youth, as their substance use rates are comparable to, if not higher than in some cases, urban and suburban youth in the United States.
This paper examines how cross-domain risk factors are related to each other in predicting longitudinal change in substance use attitudes and refusal cognitions in sixth grade students. The data are from the PROSPER project, a trial of dissemination of evidence-based prevention programs through community partnerships. PROSPER involves 28 rural and small town communities in two states. The data set includes 11,000 students' self-reports in the Fall and Spring of sixth grade, as well as school and community level data including principal and key informant reports and U.S. Census data.
Analyses utilized multilevel modeling techniques to account for clustered data at the school level. Findings indicate that at the individual level, risk factors associated with substance use attitudes include individual (school adjustment and bonding), parenting (general child management), and family (cohesion) factors. At the school and community levels, factors associated with substance use attitudes include the schools' provision of prevention services and community readiness. The paper further describes relative contributions of risk factor domains in accounting for unique variance in substance use attitudes, as well as the extent to which certain risk factors mediate the influence of other risk factors. Future work in this line of research will follow the students into high school and examine the array of risk factors in predicting substance use.
POSTERS: PROMOTING WELL-BEING
DIVERGING SES PROFILES FUEL THE HEALTH KNOWLEDGE/WELL-BEING GAP. Una Medina1, Everett M. Rogers1, Gill Woodall1, David Buller2, 1University of New Mexico, Albuquerque, NM United States; 2Cooper Institute, Denver, CO United States
Availability of free Internet telecenters at libraries and community centers is thought to be the answer to closing the knowledge gap and the digital divide. This is important for community well-being because one of the most accessed types of information on the Internet is health information. Although the presence of free public access Internet telecenters offers potential of bridging the divide, the gap continues to widen because the rate of increase of knowledge is faster for higher socioeconomic individuals and slower for lower socioeconomic persons. Pay-for-Internet cyber cafes versus free Internet telecenters have different accessibility thresholds, offer different environments and attract two different types of customers. The socioeconomic differences in the two types of patrons are quantified in this study via results of an electronic user survey and a telecenter survey. While a knowledge gap is evident between pay and free Internet access sites, both types of public Internet access, have severely limited access and a large knowledge gap compared to computer owners with Internet connections.
DEVELOPMENT OF A COMPREHENSIVE POLICY FOR PREVENTIVE PARENTING SUPPORT. Hilde Colpin1, Agnes De Munter2, Lieve Vandemeulebroecke2, 1Catholic University of Leuven, Tiensestraat 102, 3000 Leuven Belgium; 2Catholic University of Leuven, Vesaliusstraat 2, 3000 Leuven Belgium
Systematic analysis of prevention literature reviews points out that aiming to establish policies, institutional practices and environmental supports that nurture optimal development is one characteristic of coordinated prevention programming that works (Weissberg, Kumpfer & Seligman, 2003). The present study aimed at developing a comprehensive policy for preventive parenting support in the region of Flanders (one of the three regions in Belgium). More specifically, three research objectives were stated: (1) development of a comprehensive frame of reference for preventive parenting support, including quality standards, (2) analysis of strengths and weaknesses of the parenting support supply in the region and (3) proposition of future lines for a comprehensive regional policy on parenting support. In order to respond to these objectives, a Delphi study was conducted (e.g. Adler & Ziglio, 1996), with a panel of 28 experts representing different sectors (welfare, education, health and socio-cultural sector) relevant for parenting support in the region. In four consecutive Delphi rounds (resp. face-to-face interview, questionnaire and two round table workshops) participants were consulted about their definition and conceptualization of preventive parenting support and of quality standards, strengths and weaknesses of supply and suggestions for policy improvement.
The final model with quality standards was based on systems-ecological theory (e.g. Bronfenbrenner & Morris, 1998), including a macro- (society), meso (organization) and microlevel (actual supply of parenting support) and on a health-promotion and competence-enhancement framework (e.g. Dunst, Trivette & Deal, 1995). The overall quality standard is `empowerment´, serving a dual function: as a protective factor that decreases problem behaviour and as a foundation that supports healthy development and well-being (e.g. Weissberg et al., 2003). Empowerment is conceived as a multi-faceted construct with three aspects: a vision on man and society, an intervention process and an evaluation method. Further specification of these aspects led to operational quality standards at each of the three (macro-, meso- and micro) levels. The presentation will go further into these quality standards and briefly discuss the results of the strengths and weaknesses analysis.
THE ROLE OF ETHNIC IDENTITY AS A PROTECTIVE FACTOR AGAINST ALCOHOL AND MARIJUANA USE IN BLACK YOUNG ADULTS. Lisa A. Pugh1, Steven Allwood1, Dr. Brenna H. Bry1, 1Rutgers, The State University of New Jersey, Piscataway, NJ United States
Substance abuse remains a pervasive problem facing today´s young adults. The influence of personal characteristics on patterns and frequency of drug use must be considered in order to understand why some individuals are less vulnerable to drug use than others. One promising characteristic that recent research has uncovered is the degree to which individuals from America´s minority groups identify with their ethnic heritage (Marsiglia, Kulis, Hecht, & Sills, 2004). The majority of the research, however, has been conducted with Latino/a Americans. Thus, further consideration of this possible protective factor among Black Americans is warranted.To date, the few empirical studies conducted with Black young adults provide mixed results. In one study, ethnic identity was identified as a multi-dimensional construct, with a positive ethnic identity (stronger ethnic pride, sense of belonging, and affiliation with one´s ethnic group) serving to deter Black youth from drug use (Brook, Balka, Brook, Win, & Gursen, 1998). In contrast, another study found certain aspects of Black identity (i.e. ethnically consistent behaviors, speech, and looks) to be linked with higher use (Marsiglia, Kulis, & Hecht, 2001). Thus, more research is needed in order to explore these seemingly inconsistent findings.The current study will take place in October of 2004 and aims to isolate some key aspects of ethnic identity development that may function as protective factors for alcohol and marijuana use by assessing within group relationships between ethnic identity and drug use among Black college students.Approximately 150-200 college undergraduates from three introductory courses in the Africana Studies Department at a large northeastern university will be given the opportunity to complete a questionnaire packet about their everyday behaviors and beliefs in relation to ethnic identity and the extent of their alcohol and marijuana use and non-use. The survey packet will consist of a quantity/frequency measure for alcohol and marijuana use (Bates & Labouvie, 1997), The Multigroup Ethnic Identity Measure (Phinney, 1992), and a brief demographic questionnaire. Separate hierarchical regression analyses for alcohol and marijuana use will be conducted to explore the extent to which two aspects of ethnic identity help account for use among Black college students. The following variables will be included: a) gender, b) ethnic identity search (a developmental and cognitive component of ethnic identity), and c) affirmation, belonging, and commitment (an affective component of ethnic identity). By the time of the conference, we will be able to present results of these analyses.
INFLUENCE OF KINSHIP TIES ON THE DEVELOPMENT OF RESILIENCE. Jean Hall1, 1University of Tennessee, Knoxville, TN United States
Two different research questions were posited for this mixed method study. In Phase I the author examined levels of self-esteem, kinship social support and coping responses among adult children of alcoholics (ACOAs) and adult children of non-alcoholics (non-ACOAs). African American undergraduate students in philosophy, criminal justice and sociology classes, ages 20-45 (n=150) completed four self-report measures and a demographic questionnaire in a group setting. The first hypothesis to be tested in this study examined whether there were statistically significant differences in the level of self-esteem among the sampled African American adult children of alcoholics and adult children of non-alcoholics. Using the composite scores from the Rosenberg Self-Esteem scale, the results indicate no statistical difference between ACOAs and non-ACOAs self-esteem (t (53.02) =. 089, two tailed, p=. 929). With regard to closeness to kinship networks respondents completed a Kinship Social Support measure. The results were not statistically significant (t (38.09) = -.761, two tailed, p = .452). A t-test was also used to determine whether mean scores on the coping responses inventory differed for ACOAs and non-ACOAs. The mean composite score for ACOAs group (80.21) was higher than that for the non-ACOA group (72.2). T-tests analysis did not show statistically significant differences between the two groups (t (57.50) = 1.84 two tailed, p= .070). T-tests were computed to compare the groups on each of the eight subscales within the CRI. The results indicate no statistically significant differences in logical analysis, seeking guidance and support or problem solving coping responses between the two groups. However, ACOAs exhibit more positive reappraisal coping responses (t (44.93) =2.16, two tailed, p= .036) than adult children of non-alcoholics. In Phase II of the study the questions were: How do African American adult children of alcoholics perceive their relationships with kin or fictive kin (i.e. grandparent, aunt, uncle, mentor, schoolteacher, coach, etc.) influenced their experience of living with an alcoholic parent and impacted the way they felt about themselves and their ability to solve problems? African American adult children of alcoholics, ages 20-45 (n=10) were interviewed for the study. Utilizing grounded theory, qualitative data were examined by the author who developed a “Theoretical Model of Kinship Social Support”. The findings suggest ACOAs who had kin-fictive kin attachments are resilient, have healthy self-esteem, and cope with living with an alcoholic parent.
EVALUATION OF A SCHOOL-BASED PEER EDUCATION AND SEXUAL HEALTH PROMOTION PROGRAM. Sherry Barr1, Bonnie Parker2, Sharon Rose Powell1, Alfred Vasapolli3, 1Princeton Center for Leadership Training, Princeton, NJ United States; 2HiTOPS, Inc., Princeton, NJ United States; 3New Jersey Department of Health and Senior Services, Trenton, NJ United States
Although teen pregnancy and teen birth rates have declined in the US in recent years, these rates are still much too high. The US continues to have the highest teen birth rate and one of the highest rates of sexually transmitted infections (STIs) among all industrialized nations. The New Jersey Teen Prevention Education Program (Teen PEP) began in 1999 as a statewide response to empower students across the state to take an active role in educating their peers about sexual health. Teen PEP is a school-based peer education and sexual health promotion program to increase adolescents´ knowledge, skills, and behaviors related to sexual health. Teen PEP is currently active in 45 urban, suburban and rural public high schools across NJ.
The core of Teen PEP is a curriculum-based, year long course for credit in sexual health and peer education which is team taught daily to selected junior and/or senior students. Students are trained to facilitate workshops that address postponing sexual involvement, unintended pregnancy, HIV/AIDS and other STIs, and a range of other teen health concerns. Workshops help students build critical skills, including communication with peers and parents, problem solving, decision making, negotiation and refusal skills. In each school, peer educators conduct at least five workshops with a target population of approximately 30 freshmen and/or sophomore students.
A program evaluation was conducted during three consecutive academic years (2001-2004). Knowledge, attitudes and behaviors related to sexual health were measured using a self-report survey instrument. Peer educators and the target population were assessed prior to participation in the program and after participation for one year. Changes in peer educators´ knowledge, attitudes and behaviors were measured against pre-program survey results and against an incoming group of comparable peer educators prior to participation in the program. Pre to post-test changes in knowledge, attitudes and behaviors of the target population were also measured.
Evaluation results consistently demonstrate that after participation in Teen PEP, students are more knowledgeable about sexual health issues, more likely to report attitudes that support safer sex practices and responsible decision-making, more likely to discuss issues related to sexual health with peers and partners, better able to identify resources for information on sexual health, more likely to report using contraception most or every time they were sexually active and more likely to visit a health care professional or clinic for issues related to reproductive health care.
Teen PEP is a collaboration among the NJ Department of Health and Senior Services, the Princeton Center for Leadership Training and HiTOPS, Inc.
SCHOOL-BASED MENTAL HEALTH EDUCATION: ITS ROLE IN PREVENTION. Heather Parris1, Catherine Dulmus1, Karen Sowers1, William Rowe2, James Blackburn3, 1University of Tennessee, Knoxville, TN United States; 2University of South Florida, Tampa, FL United States; 3Hunter College, New York, NY United States
Mental health education delivered in the school setting is often used to increase knowledge on the signs and symptoms of mental illness. A local agency delivered a mental health educational curriculum to students in grades 7-12 (n=2165), in both rural and urban school settings, located in the southeast region of the United States. All students were pre-tested and post-tested on curriculum information. Findings indicate that regardless of grade (middle school or high school) or school location (urban or rural) that students were knowledgeable of mental illness prior to receiving the curriculum. An overview of the study will be presented, as well as recommendations on how limited available monies might be better spent on prevention of suicide through depression screening rather than on basic mental health education in the school setting.
SENSE OF SCHOOL BELONGING AND CLASSMATE ACCEPTANCE: FACTORS THAT CONTRIBUTE TO ACADEMIC ACHIEVEMENT AND PROTECT AGAINST ADOLESCENT SUBSTANCE ABUSE?
. Phuong-Anh Urga1, Brenna Bry2, Valerie Johnson1, 1Rutgers, The State University of New Jersey, Piscataway, NJ United States; 2Rutgers, The State University of New Jersey, New Brunswick, NJ United States
Academic achievement has been consistently identified as a protective factor against substance abuse for early adolescent youth. Several studies have found that sense of school belonging, the degree of psychological and personal connection students feel with the school social environment, is positively correlated with academic achievement. An important question, then, is what variables contribute to sense of school belonging, and do they, directly or indirectly, also contribute to academic achievement? A recent study addressed this question and suggests that a student´s peers are important to both sense of school belonging and academic achievement. Urga and Bry (2003) found that classmate acceptance correlated positively with both sense of school belonging and academic achievement in an ethnically diverse, cross-sectional sample of early adolescents. Furthermore, findings from this study support a model in which sense of school belonging mediates the relationship between classmate acceptance and academic achievement. Thus, it appears that classmate acceptance and sense of school belonging are important variables that may strengthen academic achievement, a protective factor against early adolescent substance abuse. These findings, however, require replication. Furthermore, although it is speculated that these variables are negatively correlated with substance use, no study has examined, simultaneously, the relationships among classmate acceptance, sense of school belonging, academic achievement and adolescent substance use. We propose to test these relationships in a cross-sectional sample of diverse 9th graders from an urban high school using validated self-report measures. In addition to replicating the findings from Urga and Bry (2003), we hypothesize that classmate acceptance, sense of school belonging, and academic achievement will be negatively correlated with adolescent alcohol and substance use.
This research is supported by NIDA Grant 5P20DA017552-02