Join us for YTH Live Global! Learn More

ETR Logo ETR Logo ReCAPP logo

etr presents YTH LIVE Global 2021: October 4-6: Virtual. Register now at

Theories & Approaches

All Theories & Approaches

Research Study Summaries

Example #1: Predicting Condom Use in Seattle, WA4

Between 1990 and 1994, researchers in Seattle, WA conducted a study based on several behavioral theories to try to prospectively predict condom use among four groups at risk for HIV infection: 1) injection drug users, 2) men who have sex with men, 3) female commercial sex workers, and 4) multi-partnered heterosexuals.

The researchers developed a questionnaire to measure behaviors, intentions, attitudes, subjective norms (all according to the TRA), and perceived behavioral control (an element of an expanded TRA model called the Theory of Planned Behavior or Problem Behavior Theory). Consistent with TRA, participants were asked about the factors that might affect their intent to use condoms with different types of partners (e.g., regular, casual, customers) or different types of sex (vaginal, anal, or oral). These included questions about their beliefs regarding the outcomes of using condoms, who they listened to regarding condom use, and what factors made it easier to use or not use condoms.

From the answers to these questions, the researchers developed specific survey questions that measure for attitude, subjective norm, and perceived control for different combinations of partners and types of sex, since the respondents had indicated that different factors affected each behavior. For example, intention to use condoms was measured on a 7-point scale ranging from "extremely unlikely" to "extremely likely." Attitudes toward condom use (for each type of sex and partner) covered 32 different outcomes and attributes because of all the different combinations (e.g., "using a condom implies you don't trust your partner").

Outcomes were rated on a 7-point scale that ranged from "extremely bad" to "extremely good." Likewise, participants rated their beliefs about whether or not 15 different influential people in their lives thought they should use condoms, and whether or not they (the participants) were motivated to comply with these beliefs or norms.

In their analysis of the results (which was done using the statistical tools of correlation and multiple regression analyses), the researchers found that of the 31 behavioral beliefs they measured to compute "attitude," 28 were correlated for condom use with a regular partner and 19 with condom use with a casual partner. A similar analysis about normative support showed that family support was equally important for both types of partners but support from friends was more influential with casual partners.

By differentiating the types of situations and factors, the researchers were able to identify the specific contributions of attitudes and subjective norms with different types of partners (and among the different participant subgroups), indicating that interventions to influence condom use with regular partners should emphasize different factors than those trying to influence condom use with casual partners.


Example #2: Comparing the Theory of Reasoned Action and Problem Behavior Theory in Quebec, Canada5

Researchers in Quebec, Canada tested the predictive power of the TRA by comparing it to an expanded version of the TRA (Problem Behavior Theory, or PBT) in explaining two behaviors related to teen pregnancy and reproductive health: age at first intercourse and consistency of condom use. (The PBT is an expanded version of the TRA that adds the concept of perceived control over the behavior to the attitudes and norms described by the TRA.)

Each theory was combined with the Theory of Reproductive Development, which suggests that stressful family living conditions accelerate the onset of puberty and thus lower the age of first intercourse. The study explored which theoretical model (TRA plus the Theory of Reproductive Development, or PBT plus the Theory of Reproductive Development) best predicted the two sexual risk behaviors.

To test the two, the researchers distributed questionnaires to parents, teachers, and children drawn from a sample of 3,018 French Canadian families tracked while the children were 6 to 18 years old (from 1986 to 1998). Participants completed questionnaires in each of these years, with the youth completing them themselves as they moved into secondary (high) school.

To gauge both attitudes and norms, the questionnaires asked about:

  • delinquency at age 13
  • deviant behavior among peers (asked of parents)
  • sensitivity to peers' influence (questions about peer pressure, asked of adolescents)
  • attitudes toward sexuality (e.g., agreeing or disagreeing with the statement, "It is all right to have a one-night stand . . . Or sex just for fun . . . Or if you're not really in love")
  • attitudes about contraception (e.g., true or false, "If I use contraception, I won't get pregnant," . . . "Contraceptive methods are complicated and make sex unpleasant" . . . "I don't know where and how to get condoms" . . . "Condoms are boys' responsibilities only."
  • Subjective norms — peers' norms toward condom use (e.g., "Most boys do not want to use a condom."
  • Knowledge about sex and contraception ("Only gay people can get AIDS" . . . "A condom is not an efficient contraceptive method". . . )

When the results were analyzed, 5 of the 14 variables in the TRA combined model significantly predicted age at first intercourse. (None of the combined PBT variables did so.) Interestingly, different sets of variables predicted age at first sexual intercourse and the consistency of condom use, suggesting to the researchers that these two behaviors had different meanings to adolescents (and thus should be addressed in different ways).


Next: How Can I Use the TRA in my Setting?