Southern Online Journal of Nursing Research

www.snrs.org

Issue 1, Vol. 4
February 2003

Gender Differences in Goal Setting for HIV Prevention among College Students

Johanna E. Soet, M.A¹; Colleen DiIorio, Ph.D²; William N. Dudley, Ph.D³; Tammy M. Woodring, M.P.H.⁴

¹Senior Associate Faculty, Rollins School of Public Health, Emory University, Atlanta, GA; ²Professor, Rollins School of Public Health, Emory University; ³Professor, College of Nursing, University of Utah, Salt Lake City; ⁴Project Director, Rollins School of Public Health, Emory University

Interactive comments section 

ABSTRACT

The purpose of this study was to identify HIV prevention goals of college students, to determine if there are differences in goal setting between males and females, and to determine if an association exists between goal setting and behavior. The data are from a study designed to identify HIV prevention practices of college students. The results of the study showed that 71.4% of the respondents indicated that they had a goal to reduce their risk of contracting HIV. The primary goals identified were condom use, limiting number of partners, abstinence, and monogamy. Females were more likely to select abstinence as their first goal, and men, condom use. Females were more likely than males to write high specificity and definitely effective goals. Significant associations were also found between HIV prevention goals and sexual behaviors. When males and females stated abstinence as their goal, there was a significant association with reports of never having sex. This association was significant for both sexually experienced males and females when the goal of abstinence was compared with the occasions of sex in the last three months. For males, having a condom use goal was significantly associated with consistent condom use. However, no significant association was found between females’ condom use goals and reported consistency of condom use.

Key words: HIV/AIDS, college students, goal-setting, sexual behavior

Introduction

Estimates based on the age distribution of AIDS cases suggest that about half of new HIV infections are among people under the age of 25,¹ and the majority of these infections are acquired through sexual behaviors.² These statistics hold particular relevance for undergraduate college students most of whom are under 25 years of age and many of whom practice sexual behaviors that place them at risk for contracting HIV. In order to avoid HIV infection, college students, like others, must adopt behaviors to protect themselves. To date, the study of prevention behaviors among college students has focused on identifying antecedents to risk reduction behaviors including knowledge, attitudes toward condom use, and confidence in using a condom and discussing condom use with a sexual partner.^3-5This research has yielded important results that have been incorporated into risk reduction education programs. A much less studied area, but one that is important to self-regulatory behavior, is that of goal setting.

According to Bandura, a personal goal is something a person wants to accomplish.⁶ He notes that goals are important in the self-regulation of behavior because they help focus attention on the desired behavior, increase efforts toward the attainment of the desired behavior, and enhance persistence in the face of difficulties. Moreover, goal statements work to create internal standards against which current behavior can be compared. When behavior deviates from these predetermined standards, internal incentives can be created to modify behavior to meet desired performance goals.

The nature of goals and the association between goal setting and task performance has been examined in a number of studies.^7-11 Investigators have found that goal statements can vary in their level of specificity, the level of difficulty, and the proximity to desired outcomes.⁶ Overall people who set goals for the purpose of meeting some performance standard are more likely to be successful than those who do not set goals, but have the same desired outcome.⁶ Moreover, successful outcomes are more likely for people who set more specific or challenging goals.¹⁰ Although research addressing health related goals is minimal, investigators have shown the success of goal setting within the health domains of weight training,⁸ smoking cessation,⁷ and endurance performance.⁹ For example, Boyce and Wayda⁸ found among female university students engaged in a weight training experiment, the performance of those who had set their own goals (self‑set goals) or who were assigned goals was significantly better than that of women in the control group who had no goals.

Although there is little empirical data supporting the relationship between goal setting and HIV risk reduction behaviors, interventionists often incorporate goal setting into prevention education.^12,13 For example, a successful cognitive-behavioral risk-reduction intervention among adult men and women included goal setting as a technique to enhance perceived self-efficacy.¹³ However, in this study, the role of goal setting in changing behavior was not disentangled from that of the other mediators of change including self-efficacy and outcome expectancies. Because research in health behavior suggests that setting goals acts as motivation for behavioral change^7-9and goal setting is already included in many HIV prevention programs, ^12,13 the study of self-set goals and their relationship to the adoption of risk reduction behaviors is timely. To expand the understanding of risk reduction goals, the first aim of this paper was to determine the types of HIV prevention goals set by college students. Because men and women report differences in the adoption of HIV prevention behaviors, the second aim was to determine if gender differences exist in HIV prevention goals, and the third aim was to explore the association between goal setting and behavior.

Methods

Data for this study were collected as part of a larger study on HIV risk-reduction practices of college students. Participants were selected from students attending six colleges and universities, both public and private, in a large southeastern metropolitan area. Once approval had been obtained from the institutional review board at each school, a request for a random sample of students currently enrolled in a degree-seeking program and under age 25 was made to each registrar. The address lists were checked for completeness; those students without a complete address were deleted from the sample. Survey packets that included the study questionnaire, a cover letter containing the elements of informed consent, a self-addressed, stamped envelope, and a five-dollar bill as an incentive to complete the survey were sent to students. Students were asked to complete the questionnaire and return it. Survey packets were sent by first class mail; a reminder postcard was sent one week after the first mailing, and a second survey packet was sent to the non-responders three weeks after the first mailing. Of 5,893 survey packets mailed, 2,468 were returned representing a 42.9% response rate.

Sample

The sample was limited to respondents who were unmarried, between 18 and 25 years of age who had written at least one HIV prevention goal (N = 1,525). The average age of the sample was 20.2 years (SD = 1.73). Fifty-four percent of the total sample was female, 31.7% was white, 58.8% African American, 4.9% Asian, 3.6% Hispanic, and 1% Native American or other. Twenty-nine percent of the participants were freshman, 22.2% sophomores, 24% juniors, and 24.6% seniors. Ninety-six percent of the sample identified themselves as heterosexual, 1.8% identified as bisexual, and 1.3% as homosexual, with .5% not responding. Eighty-six percent of the males and 87% of the females were sexually active. Sexually active was defined as ever having had vaginal, oral or anal sex. For the males in the sample, 18.2% reported no occasions of vaginal, oral or anal sex in the past 3 months; 16.2% of the females reported the same.

Measures

To measure HIV prevention goals, participants indicated whether they had personal goals by responding to the question, “Do you have any personal goals at this time about reducing your risk of being infected with HIV? By personal goals, we mean have you made up your mind to make some change, or to maintain some change that you have previously made in your sexual relationships, your use of drugs, or any other aspect of your life that might place you at risk for HIV?” They were then asked, “If yes, what is your personal goal or goals (i.e., what have you made up your mind to do)? Please be as specific as possible. List each goal separately if you have more than one goal.” Participants were not asked to rank their goals in order of importance.

Each goal was evaluated and coded on four dimensions—content, specificity, effectiveness, and control. These four dimensions were agreed upon and a goal coding manual was developed to guide the coding of goals. Goal content refers to the subject matter of the goal. Based on goal statements, 14 content categories were identified. The categories were the following:

1. condom use
2. protection/safer sex
3. discussion/communication
4. education
5. limit number of partners
6. monogamy
7. marriage
8. abstinence
9. no drugs/alcohol
10. no IV drugs
11. no anal sex
12. testing
13. avoiding “tempting” situations and being prepared
14. other

If a goal did not fit into any of the first 13 categories, it was coded as “other.” Specificity refers to the extent to which specific actions and/or timeframes are included in the goal statement. The specificity dimension was rated as one of three categories: high (e.g., always use a condom), medium (e.g., use condoms), or low (e.g., use condoms more). Effectiveness refers to the probable efficacy of the goal. The effectiveness dimension was rated as one of four possible categories: definitely effective (e.g., use a condom each and every time), possibly effective (e.g., use condoms), indirectly effective (e.g., getting tested for HIV), and ineffective (e.g., use the pill).

Control refers to the locus of control for the successful completion of the goal. The control dimension was divided into three categories: self-control (e.g., abstinence), mutual control (e.g., condom use), or other control behavior (e.g., have my partner tested). Before coding the entire data set, raters were trained on using the coding manual. The percent agreement between the trainer and the raters was assessed for each dimension and was required to be .90 or above before coding began.

Sexual activity status was determined by asking 3 questions: “How old were you when you first willingly had vaginal intercourse/oral (oral-genital contact) intercourse/anal intercourse?” For each question the respondent could provide an age or check “never had.” For this study, a dichotomous measure was used where 0 represented those who had never engaged in any sexual activity including vaginal, oral or anal sex, and 1 represented those who had engaged in at least 1 of the 3 behaviors.

Sexual activity in the past three months was assessed by asking, “ With how many different partners have you had sexual intercourse in the past 3 months?” For this analysis, the results were dichotomized into those who had no partners in the past three months and those who had one or more.

Condom use was measured using responses to the item, "How often do you use a condom?" The item was rated on a 5‑point scale ranging from never to every time. For this analysis, this measure was dichotomized into consistent condom users (every time or almost every time) and inconsistent users (sometimes to never).

Results

Data were analyzed using SPSS 9.0. Descriptive statistics were used to identify the types of HIV prevention goals for males and females, and chi square analyses were used to assess differences in the dimensions of goal statements for males and females and to examine the association between goal setting and behavior. The large sample size (n = 1, 525) provides considerable statistical power to detect small differences as statistically significant. To aid in the interpretation of the chi square statistics, we have included Cohen’s (1988) measure of effect size, w, which in all tables reported here is equal to (also equal to phi and Cramer’s V in the tables). We have only included w for chi squared values associated with p < .001. Cohen¹⁴ has proposed small, moderate, and large effect size of w = .1, .3, and .5 respectively.

For the analyses, only the first goal written by each respondent was evaluated because by using only the first goal all participants who had goals were included. The top four responses written for the HIV prevention goals were: 23% condom use, 22% limiting number of partners, 19% abstinence, and 17% monogamy. Forty percent of participants identified a high specificity goal, 35% a medium specificity goal, and 25% a low specificity HIV prevention goal. Forty-five percent of all participants identified a possibly effective HIV prevention goal, 33% a definitely effective HIV prevention goal, 22% an indirectly effective HIV prevention goal, and less than one percent an ineffective HIV prevention goal. Fifty-seven percent of participants identified a self-control goal, 42% a mutual control goal, and less than one percent another control goal.

With respect to the types of goals males and females wrote, a significant association was found between gender and the content dimension of participants’ HIV prevention goals (χ² (13, 1,525) = 34.63, p < .001) (Table 1). Males were significantly more likely to state a condom use goal (χ² (1, 1,525) = 4.27, p < .05) and a no IV drug use goal (χ² (1, 1,525) = 5.62, p < .01). Females were more likely to state an abstinence goal (χ² (1, 1,525) = 8.67, p < .01).

Table 1. Content dimension of HIV prevention goals by gender

With regard to the other three dimensions (specificity, effectiveness, control), a significant association was found between gender and the specificity dimension of participants’ HIV prevention goals (χ² (2, 1,525) = 19.04, p < .0001). Females were significantly more likely than males to write a goal that was coded as high specificity, and males were significantly more likely than females to write a goal that was coded as low specificity. A significant difference was also found between males and females in the effectiveness dimension for the goals (χ² (3, 1,525) = 9.25, p < .026). Females were found to be more likely to write a definitely effective goal (Table 2). No significant gender differences were found for the control dimension.

Table 2 . Specificity and effectiveness dimension of HIV prevention goals by gender

In the last stage of analysis, the association between the participants’ goals and their self-reported behavior was explored. The association between abstinence and condom use goals and self-reported measures of sexual activity and condom use were assessed. In the overall sample, those who defined abstinence as their goal were significantly more likely to never have engaged in sexual intercourse (vaginal, oral or anal) (χ² (1, 1,525) = 290.79, p < .001) (Table 3). Additionally, those who were sexually experienced and who defined abstinence as their goal were more likely to have not engaged in sex in the past 3 months (χ² (1, 1,328) = 80.581, p <.001). Participants who had condom use as their goal were more likely to report consistent condom use (χ² (1, 1,266) = 6.32, p <.05).

Table 3. Association between HIV prevention goals and behavior

To examine gender differences in the association between goals and behavior, separate chi square statistics were run for males and females. When males (χ² (1, 699) = 133.01, p <.001) and females (χ² (1, 862) = 162.277, p <.001) stated abstinence as their goal, there was a significant association with reports of never having sex (Table 4). In addition, the association was significant for both sexually experienced males (χ² (1, 605) = 32.55, p < .001) and females (χ² (1, 723) = 50.85, p <.001) when the goal of abstinence is compared with the occasions of sex in the last three months (no sex v. had sex). For males, having a condom use goal was significantly associated with consistent condom use (χ² (1, 582) = 8.43, p <.01). However, no significant association was found between females’ condom use goals and reported consistency of condom use (χ² (1, 684) = .36, p = .5393) (Table 5).

Table 4. Association between HIV prevention goal and behavior for students who define abstinence as their first goal