New model for predicting adolescents' sexual intentions.

Abstract: Using a sample of 411 urban 7th and 10th graders, the study developed and tested a model of sexual intentions for adolescents where beliefs, alcohol use, and past sexual behaviors were predictors of teenage' sexual intentions. Results of the path analysis revealed a negative direct relationship between beliefs and sexual intentions. Alcohol use demonstrated an indirect effect on sexual intentions mediated by past sexual behavior while having a positive direct effect on past sexual behavior. This relationship suggests that adolescents' beliefs may frequently be in opposition to the sexual behaviors in which they have engaged.

**********

There are many factors that have been found to be related to teens' sexual behaviors. Understanding teen sexual behaviors requires knowledge of teens' alcohol use, sexual beliefs and sexual intentions (Nahom, et al., 2001; Watts Sr. & Nagy, 2000; Schwarzer & Fuchs, 1996). Various teen risk taking behaviors (e.g., alcohol use) are important predictors of sexual behaviors. In fact, alcohol use is a strong predictor of sexual behaviors (Christopher, Johnson, & Roosa, 1993; Dryfoos, 1990; Flick, 1986; Klitsch, 1994; Ketterlinus, Lamb, & Nitz, 1994; Luster & Small, 1994; Small & Luster, 1994; Whitbeck, Yoder, Hoyt, & Conger, 1999). Perkins, Luster and Villarruel (1998) examined the relationship between sexual behaviors and the risk-factors among diverse ethnic teens. They found that teens who engaged in alcohol use had higher rates of sexual behavior than those who did not engage in alcohol use. In addition, Small and Bogenschneider (1994) reviewed the factors related to sexual behaviors among teens. They found that, as the frequency of alcohol consumption increased, teens were likely to be sexually active.

The Health Belief Model-Social Learning Theory approach (Eisen, Zellman, & McAlister, 1992) links behaviors with knowledge, attitudes, beliefs, and skills. The social learning theory claims that behavioral changes result from increasing the strength of the perceived relationships of a behavior and its consequences, in addition to how these consequences are evaluated. There are a variety of measures assessing sexual belief which depend on the research interests. Several studies which examine sexual belief in the relation to sexual behavior indicated different scope of sexual belief questions (e.g., Kinsman, Romer, Furstenberg, Schwartz, 1998). In Rosenthal, Moore, and Buzwell's study (1994), they found that homeless adolescents' sexual beliefs impacted their sexual behavior. Their instrument of sexual beliefs included self-perceptions of sexual anxiety, arousal, exploration, and commitment. However, Bettinghaus (1986), DiClemente, Zorn and Temoshok (1986), Kirby et al. (1994) found that teen sexual behaviors were not strongly linked to the changes in attitudes.

Nahom et al. (2001) studied 1,173 students in grades 8th through 10th to examine gender and differences between sexually experienced and inexperienced youth with regard to intentions to engage in sexual activity and use condoms, perceptions of peers' engagement in sexual activity, and pressure felt to engage in sexual activity. They reported that teens' intentions to engage in sexual behavior differed by gender and prior sexual experience. Sexually experienced teens had higher intentions to have intercourse in the next year than non-sexually experienced teens. Males were significantly more likely to have intercourse in the next year than were females.

Although several studies have examined factors related to teens' sexual activities, few studies have focused on the relationship among those factors. Based on these multiple factors associated with teen sexual behaviors, we have determined measured variables for a hypothesized model, which is an attempt to connect teen sexual behaviors with beliefs, intentions, and alcohol use (Figure 1). In the model, one-way arrows represent conceptual relationships and indicate the impact of one factor on another. Two-way arrows represent conceptually related factors. We hypothesized that sexual belief and alcohol use predict teens' sexual intentions mediated by teens' past sexual behaviors.

[FIGURE 1 OMITTED]

METHODS

PARTICIPANTS

Passive parental consent procedures were used. Confidentiality was assured before distributing the questionnaires. The students included in this study were a sample of 411 seventh and tenth graders from urban areas in Alabama. Forty-two percent were females, and the remaining fifty-eight percent were males. The participants were ethnically diverse with 52% African-American and 48% Caucasians.

INSTRUMENTATION

Based on a review of the literature, a 41-item scale was developed to measure beliefs toward sex, alcohol use, behavioral intention, self-report sexual activities and demographic information. Eight items were used for the present analyses. All of the variables were examined for content validity and indices were tested for internal consistency using Cronbach's alpha. Most of the variables have moderate or strong alpha coefficients.

Alcohol use. Alcohol use was assessed in the student survey by asking about a series of experiences in a dichotomous scoring scheme (e.g., Have you ever...). Those experiences included drinking alcoholic beverages and getting drunk. Cronbach's alpha for the alcohol use was .6743.

Behavioral intentions. Behavioral intentions were measured by two questions and with 5 choices of responses format (e.g., asked about intentions to refuse or abstain from having sex and one question asked the reason for staying abstinent). Cronbach's alpha for the behavioral intentions was .8152.

Sexual beliefs. Sexual beliefs were measured by three questions which dealt with the appropriateness of being sexually active as a teen, and the importance of relationships. A 5-point Likert-type scale ranging from 1 = strongly disagree to 5 = strongly agree was used. Cronbach's alpha for sexual beliefs was .6159.

Sexual behavior. Teen sexual behavior was assessed by a single question that asked "Have you ever had sexual intercourse?" All responses were coded into a dichotomous variable, sexually experienced and not sexually experienced. This self-reported sexual behavior provided useful indicators of teens' actual sexual behavior (Kirby et al., 1994).

Demographic information. Questions concerning gender, age, parents' education and maternal marital status were included. Students reported their grade point average from a prior semester.

VARIABLES IN THE STUDY

Sexual beliefs, alcohol use, behavioral intention, and sexual behaviors were included in this study. The structural model contained two exogenous variables, Beliefs and Alcohol Use, and two endogenous variables, Behavioral Intentions and Sexual Behavior.

DATA ANALYSIS

AMOS 4.0 (Arbuckle, 1999) was used to assess the validity of the hypothesized model. This proposed model was evaluated three ways. First, departure of the data from the specified model was tested for significance by using a chi-square test (Joreskog & Sorbom, 1989). Second, goodness-of-fit between the data and the specified model was estimated by employing the Comparative Fit Index (CFI; Bentler, 1990), the Tucker-Lewis Index (TLI; Bentler & Bonett, 1980), and the Root Mean Square Error of Approximation (RMSEA; Browne & Cudeck, 1993). Third, all path coefficients were assessed for statistical significance at p < .05.

MISSING DATA

Because missing data appeared to be randomly scattered among the variables, a full information maximum likelihood (FIML) imputation was performed to estimate missing data.

RESULTS

INITIAL MODEL

The results of the initial path analysis failed to support the proposed model (Figure 2). The chi-square test was significant, [chi square] (1) = 193.37.93, p < .001 indicating the hypothesized model did not fit the observed data. The CFI and the TLI yielded non-impressive indices of .877 and .878 respectively while the RMSEA achieved a value of .993 indicating a poor fit of the model (Browne & Cudeck, 1993). Two path coefficients failed to achieve statistical significance: (1) Past Sexual Behavior to Values and Beliefs, and (2) Intentions to Alcohol Use.

[FIGURE 2 OMITTED]

RE-SPECIFIED MODEL

The initial model was re-specified by eliminating the two paths that failed to achieve statistical significance (Past Sexual Behavior to Values and Beliefs, and Intentions to Alcohol Use; Figure 3).

[FIGURE 3 OMITTED]

The results of the re-specified model indicated an acceptable fit (Figure 4). Although the chi-square test was significant, [chi square] (2) = 11.96, p = .003, the CFI and the TLI yielded impressive indices of .994 and .968 respectively with the RMSEA of .160, indicating a value slightly exceeding the maximum value of .1 (Browne & Cudeck, 1993). In addition, all path coefficients achieved statistical significance at a prior alpha level of .05.

[FIGURE 4 OMITTED]

The direct, indirect, and total effects, plus the variances explained in the endogenous variables are presented in Table 1.

DISCUSSION

Results of the path analysis revealed a negative direct relationship between beliefs and sexual intentions. This relationship indicated that adolescents' beliefs might frequently be in opposition to their intentions to engage in sexual behaviors. Such a finding raises questions about the strength of adolescents' beliefs in comparison to the strength of social settings and norms that may foster sexual intentions. Careful attention should be given to the social contexts and culture factors (e.g., peer pressure, school environment, and family factors). By examining the influential factors between sexual beliefs and sexual intention, we may be able to identify the specific aspects of teen sexual behaviors.

Alcohol use demonstrated an indirect effect related to behavioral intentions mediated by past sexual behavior while alcohol use revealed a positive direct effect on past sexual behavior. Teens who engage in alcohol use have higher degree of intention to engage in sexual activity than those who do not engage in alcohol use. The predictive variable (risk-alcohol use) shows a relationship with teen sexual behavior, which is consistent with several prior studies (Christopher, Johnson, & Roosa, 1993; Dryfoos, 1990; Flick, 1986; Klitsch, 1994; Ketterlinus, Lamb, & Nitz, 1994; Luster & Small, 1994; Small & Luster, 1994; Whitbeck, Yoder, Hoyt, & Conger, 1999) that found that alcohol use was one of the predictors of sexual behavior.

There was a positive direct relationship between teens' sexual behaviors and their sexual intentions. Sexually experienced teens were more likely to express an intention to have sexual intercourse in the near future. This finding is similar to that of Nahom et al., (2001) study which found intentions to engage in sexual behaviors are related to actually engaging in the sexual behaviors. Also there was an inverse relationship between values and beliefs and alcohol use. Teens that drank were more likely to agree with statements about the appropriateness of being sexually active as a teen.

This study's findings indicate that teens' alcohol use and past sexual behaviors are significant predicators of their sexual intentions. The significance of this finding has implications not only for prevention program planning but also for understanding of the effectiveness of abstinence programs. Because educating teens about abstinence until marriage has not always translated into behavioral change, efforts to improve abstinence programs should focus on the influences of adolescence sexual intentions in the cultural and social context of teens. Further understanding of how past sexual behaviors and teens' alcohol use relate to different perceptions/values/ beliefs among teens, particularly in the application of the model presented herein, would help the abstinence education program developer to take these factors into account in order to reduce the negative consequences of teen sexuality.

ACKNOWLEDGEMENT

This study was supported by a grant from the Department of Health and Human Service: Office of Adolescent Pregnancy Program (CFDA 84.349A NO, 1APHPA002058-01-0).

REFERENCES

Arbuckle, J. L. (1999). Amos 4.0 User's Guide. Chicago: SmallWaters Corporation.

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238-246.

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606.

Bettinghaus, K. P. (1986). Health promotion and knowledge-attitude-behavior continuum. Preventive Medicine, 15, 475-491.

Browne, M. W. & Cudeck, R. (1993). Alternative ways of assessing model fit. In Bollen, K.A. & Long, J.S. (Eds.) Testing structural equation models (pp. 136-162). Newbury Park, CA: Sage Publications.

Christopher, F. S., Johnson, D., & Roosa, M. (1993). Family, individual, and social correlates of early Hispanic adolescent sexual expression. Journal of Sex Research, 30, 54-61.

DiClemente, R. J., Zorn, J., & Temoshok, L. (1986). Adolescents and AIDS: A survey of knowledge , attitudes, and beliefs about AIDS in San Francisco. American Journal of Public Health, 76, 1143-1145.

Dryfoos, J. G. (1990). Community schools: New institutional arrangements for preventing high-risk behavior. Family Life Educator, Summer, 4-9.

Eisen, M., Zellman, G. L., & McAlister, A. L. (1992). A health belief model-social learning theory approach to adolescents' fertility control: Findings from a controlled field trial. Health Education Quarterly, 19(2), 249-262.

Flick, L. H. (1986). Paths to adolescent parenthood: Implications for prevention. Public Health Reports, 101(2), 132-147.

Joreskog, K. J. & Sorbom, D. (1989). LISREL 7: A guide to the program and applications (2nd ed.). Chicago: SPSS.

Ketterlinus, R. D., Lamb, M. E., & Nitz, K. A. (1994). Adolescent nonsexual and sex-related problem behaviors: Their prevalence, consequences, and co-occurrence. In R. D. Ketterlinus & M. E. Lamb (Eds.), Adolescent problem behaviors: Issues and research (pp. 17-39). Hillsdale, NJ: Erlbaum.

Kinsman, S. B., Romer, D., Furstenberg, F. F., & Schwartz, D. F. (1998). Early sexual initiation: The role of peer norms. Pediatrics, 102(5), 1185-1192.

Kirby, D., Short, L., Collins, J., Rugg, D., Kolbe, L., Howard, M., et al. (1994). School-based programs to reduce sexual risk behaviors: A review of effectiveness. Public Health Reports, 109, 339-360.

Klitsch, M. (1994). Risk factors for teenage sex. Family Planning Perspectives, 26(4), 181-192. Luster, T., & Small, S. A. (1994). Factors associated with sexual risk-taking behaviors among adolescents. Journal of Marriage and the Family, 56, 622-632.

Nahom, D., Wells, E., Gillmore, M. R., Hoppe, M., Morrison, D. M., Archibald, M., et al. (2001). Differences by gender and sexual experience in adolescent sexual behavior: Implications for education and HIV prevention. Journal School Health, 71(4), 153-158.

Perkins, D. F., Luster, T., & Villarruel, F. A. (1998). An ecological, risk-factor examination of adolescents' sexual activity in three ethnic groups. Journal of Marriage & Family, 60(3), 660-673.

Rosenthal, D., Moore, S., & Buzwell, S. (1994). Homeless youths: Sexual and drug-related behavior, sexual beliefs and HIV/AIDS risk. AIDS Care, 6(1), 83-94.

Schwarzer, R., & Fuchs, R. (1996). Self-efficacy and health behaviors. In M. Conner & P. Norman (Eds.), Predicting health behaviour. Research and practice with social-cognitive models (pp. 163 - 196). Buckingham: Open University Press.

Small, S., & Bogenschneider, K. (1994, June 30). Youth at risk for early sexual activity and teenage parenthood. Wisconsin Youth Futures Technical Report #11. Retrieved May 13, 2003 from http://www.cyfernet.org/research/youthfut11.html

Small, S. A., & Luster, T. (1994). Adolescent sexual activity: An ecological, risk-factor approach. Journal of Marriage and the Family, 56, 181-192.

Watts Sr., G. F., & Nagy, S. (2000). Attitude toward sexual intercourse and relationship with peer and parental communication. American Journal of Health Studies, 16(3), 156-163.

Whitbeck, L. B., Yoder, K. A., Hoyt, D. R., & Conger, R. D. (1999). Early adolescent sexual activity: A developmental study. Journal of Marriage & Family, 61(4), 934-946.

HEALTH EDUCATION RESPONSIBILITY AND COMPETENCY ADDRESSED

Responsibility IV--Evaluating Effectiveness of Health Education Programs

Competency D: Infer implications from findings for future program planning

Sub-Competency 3: Apply findings to refine and maintain programs

Huey-Ling Lin, PhD, is affiliated with Alabama State University. A. J. Guarino, PhD, is affiliated with Auburn University. Jeffrey Gorrell, PhD, is affiliated with George Mason University. Tina Vazin, PhD, is affiliated with Alabama State University. Please address all correspondence to Huey-Ling Lin, PhD, Alabama State University, 217 Councill Hall, 915 S. Jackson St., Montgomery, AL 36104; PHONE: [phone]; FAX: [phone]; E-MAIL: [e-mail].

Table 1. Summary of Causal Effects for the Hypothesized Model

                                               Causal Effects
Outcome            Determinant          Direct     Indirect    Total

Intentions         Alcohol Use          --         .168        .168
(R2 = .35)         Values/Beliefs       -.351       --        -.351

Sexual Behavior    Sexual Behavior       .444       --         .444
(R2 = .14)         Alcohol Use           .379       --         .379
                   Values/Beliefs       --          --          --