Understanding Heterosexual Chinese College Students' Intention to Adopt Safer Sex Behaviors.

Researchers have argued that college students are at a particularly high risk for AIDS/HIV infection. College students are usually between 18 and 25 years old, ages when many people feel relatively immortal and immune to disease or accident. For most college students, they are away from home and parental control for the first time. Thus, the first few years of college are often a time for sexual experimentation and sexual irresponsibility (Baldwin & Baldwin, 1988; Centers for Disease Control and Prevention [CDC], 1991; DiClemente, Forest, & Mickler, 1990; Reinisch, Hill, Sanders, & Ziemba-Davis, 1995; Sheer & Cline, 1994). For example, 70% to 80% of the college students in the United States reported being sexually active, with one third of them never using condoms and only 20% or less using condoms regularly (Ratliff-Crain, Donald, & Dalton, 1999; Sheer & Cline, 1994). According to a newly released report by the Joint United Nations Program on AIDS/HIV and the World Health Organization (1999), one third of the 33 million people living with AIDS/HIV in the world are young people aged 15 to 24. Researchers have also suggested that many of those who are currently diagnosed with AIDS/HIV were probably infected during their college years (Schenken, 1989; Yzer, Fisher, Bakker, Siero, & Misovich (1998).

There is a paucity of studies that examine safer sex behaviors among Chinese college students in Hong Kong. Of the few studies in this area, results indicate that safer sex behaviors are largely uncommon among Chinese college students. For example, Ling (1993) found that about one quarter of the sexually active college students in Hong Kong did not use any condoms or other contraceptive methods during sexual activities. This researcher also found that even among those college students using contraception, a great proportion reported using unsafe methods in terms of protection against STDs and AIDS/HIV, such as coitus interruption, rhythm, pills, and post-coital douche. More recent data on condom use of Chinese college students in Hong Kong are not available.

The high-risk status of Chinese youths is also reflected by the official statistics. According to the Joint United Nation Program on AIDS/HIV and the World Health Organization's report (1999), the estimated HIV prevalence rates in Hong Kong for young people aged 15 to 24 range from 0.02% to 0.03% for females and 0.07% to 0.18% for males. These rates are higher than the rates estimated for Chinese youths in mainland China, which are 0.02% to 0.05% for females and 0.05% to 0.14% for males. In other words, about 1 in every 3,214 to 5,000 college-aged females and 1 in every 555 to 1,400 college-aged males in Hong Kong may be currently infected with HIV. About 80% of the people infected with AIDS/HIV in Hong Kong in 1999 were age 19 or younger at the time of reporting (Chin, 1999).

This study aimed to explore safer sex behaviors of Chinese college students in Hong Kong and to understand factors that are related to their intention to adopt AIDS/HIV-prevention behaviors. Results of the present study provide pertinent information for designing and implementation of AIDS/HIV prevention programs in Hong Kong.

Understanding Chinese College Students' Intention to Adopt Safer Sex Behaviors

Over the years, different models have been drawn upon to predict safer sex behaviors. Among them, the Theory of Reasoned Action (TRA: Ajzen & Madden, 1986; Fishbein & Ajzen, 1975) has shown the greatest power in predicting safer sex behaviors such as condom use (Basen-Engquist & Parcel, 1992; Bryan, Aiken, & West, 1996; Fisher, Fisher, & Rye, 1995; Jemmott, Jemmott, & Hacker, 1992; Winslow, Franzini, & Hwang, 1992). The TRA assumes that AIDS/ HIV-prevention behaviors such as condom use are a function of behavioral intentions which are, in mm, a function of attitudes, subjective norms, and perceived behavioral control concerning those behaviors. Attitudes are a function of people's beliefs about the consequences of performing those behaviors, multiplied by their evaluations of the consequences. Subjective norms are people's perceptions regarding the extent that specific others would approve of and their motivation to comply. Perceived behavioral control refers to people's belief about their own capability to perform the desired behaviors. The predictive value of the attitude, subjective norm, and perceived behavioral control components of this model in predicting AIDS/HIV prevention behaviors has been documented in recent studies (e.g., Fisher et al., 1995; White, Terry, & Hogg, 1994).

Despite support found in applying the TRA to predict condom use, some researchers have been critical about the definition of subjective norms and the sequence of various components of the model (Abraham & Sheeran, 1994; Ratliff-Crain et al., 1999). Others have also argued that actual behaviors of safer sex may not be the direct result of behavioral intentions, but are mediated by some other factors such as perceived vulnerability, the availability of and access to condoms, skills and sense of self-efficacy in condom use, safer sex negotiation skills, as well as subtle social influences such as people's family system, immediate environment, and broader media system (Guizzardi, Stella, & Remy, 1997; Harden & Ogden, 1999; Jemmott et al., 1992). However, these constructs are either omitted or not explicitly specified in the TRA.

Using the TRA as a framework, this study aimed to explore whether or not AIDS/HIV-related knowledge, attitudes, and vulnerability; normative beliefs such as attitudes toward safer sex behaviors; and perceived behavioral control in condom use would predict Chinese college students' intention to adopt safer sex behaviors.

AIDS/HIV-related knowledge, attitudes, and vulnerability. Both Western and Chinese studies have shown that college students are generally knowledgeable about AIDS/HIV facts and transmission routes, despite some misconceptions and myths (Davis, Tang, Chan, & Noel, 1999; DiClemente et al., 1990). However, having information about AIDS/HIV without attending to other components of the TRA is generally not predictive of safer sex behaviors (Bellingham & Gillies, 1993; Rimberg & Lewis, 1994; Serovich & Greene, 1997).

Negative attitudes about AIDS/HIV and toward people with AIDS/HIV are probably related to the deadly nature of the disease and the link between AIDS/HIV and groups of people who are already stigmatized, such as homosexuals and drug addicts. DiClemente et al. (1990) found that despite having a high level of knowledge about AIDS/HIV transmission routes, college students generally held negative attitudes toward AIDS/HIV and were prejudiced against people with this disease. These negative attitudes have a significant impact on the control of the spread of AIDS/HIV as they can decrease the impact of AIDS/HIV preventive educational programs (Neilsen & Young, 1994). Westerman and Davidson (1993) have found that high school students with prejudice against gay men tended to engage in more indiscriminate sexual activities and more unprotected sexual intercourses as compared to students with less prejudice against gay men. People with negative attitudes toward gay men and lesbians were found to be less responsive to information about AIDS/HIV (Stripp & Kerr, 1989).

Perceived vulnerability is not included in the TRA; however, it is closely tied to the motivational component of the model. As college students often believe they have little risk of contracting AIDS/HIV (Baldwin & Baldwin, 1988; DiClemente et al., 1990), they may not be motivated to adopt safer sex behaviors. This "not me" phenomenon may be a remnant of college students' adolescent egocentrism of feeling invincible and believing natural laws relating to mortality and fertility do not apply to them (Greene, Rubin, & Hale, 1995). College students, believing themselves to be the future pillars of the society, may also find it difficult to identify themselves with the high-risk behaviors commonly misconceived to be only associated with marginal youths or commercial sex workers. Recent studies have shown that a low level of perceived vulnerability or personal concern about contracting AIDS/HIV is related to not using a condom and/or an increasing number of sexual partners for college students (Ratliff-Crain et al., 1999; Serovich & Greene, 1997; Yzer et al., 1998).

Self-efficacy toward condom use. Self-efficacy refers to people's beliefs in their own ability to carry out a behavior. It affects people's choice, effort, perseverance in the face of difficulties, engagement in self-debilitating or self-encouraging thoughts, and the amount of stress and depression they experience. Self-efficacy is very similar to the perceived behavioral control component of the TRA. Bandura (1994) has suggested that low self-efficacy among college students may be related to their inconsistent condom use. Brafford and Beck (1991) have identified four dimensions of self-efficacy toward condom use: Mechanics of condom use, partner's disapproval of condom use, assertiveness about condom use, and condom use under conditions of intoxication and arousal. Researchers have found that college students with high self-efficacy toward condom use were more likely to use condoms and/or other safer sex behaviors (O'Leary, Goodhart, Jemmott, & Boccher-Lattimore, 1992; Wulfert & Wan, 1993; Yzer et al., 1998). Enhancing college students' self-efficacy beliefs about condom use was also related to an increase in their intention to use condoms, as well as an increase in their actual condom use 2 months after the AIDS/HIV prevention program (Basen-Engquist, 1994).

Attitudes toward safer sex behaviors. The behavioral options that protect against the sexual transmission of AIDS/HIV are abstinence and consistent use of condoms. It is important, therefore, to understand college students' normative beliefs about these behaviors to target intervention. To most college students, the benefits of safer sex behaviors are remote but the hindrances are immediate. Given college students' increasing interest in sexual relationships, adopting safer sex behaviors may mean a hindrance to the release of their sexual desires and role experimentation which is often coupled with other, more immediate benefits such as excitement and emotional support. In addition, adopting safer sex behaviors may make them subject to negative peer pressure, especially when the peer group the students identify with adopts an opposing view to safer sex behaviors (Serovich & Greene, 1997). The widely accepted belief that condom use would affect the spontaneity of passion and cause inconvenience (Lear, 1997) is found to relate to college students' prevalent negative attitudes toward using condoms during sexual encounters (Ratliff-Crain et al., 1999; The Family Planning Association of Hong Kong, 1994).

There is a dysfunctional norm that the discussion of safer sex behaviors is only raised out of mistrust, and actual safer sex behaviors are only for those who are promiscuous (Wight, 1992). Talking directly about sex, especially for women, may imply being sexually experienced or prepared, which may be a severe blow to their reputation (Lear, 1997). As female condoms are still comparatively expensive and uncommon, male condoms are heavily relied on as the major safer sex resource. For female college students, safer sex not only means condom use, but also includes the courage to address their partners and the skills to persuade their sexual partners to use condoms every time they have sexual intercourse (Chan & Fishbein, 1993; Harden & Ogden, 1999).

Normative beliefs relating to safer sex behaviors are of particular relevance to Chinese. First, sex is still a taboo topic, and sex-related matters are always communicated in signals and codes in Chinese culture (Zhang, Li, & Beck, 1999). Thus, open and direct discussion about safer sex behaviors is rare among Chinese. For example, sexual desire of men is symbolized as fire and that of women as water. For thousands of years, "cloud and rain" has been used to represent sexual intercourse, as "cloud" is interpreted as the "ovum" of the earth and rain as the "sperm" of the sky. Second, cultural emphasis on premarital chastity and marital fidelity also discourages the Chinese from openly discussing safer sex behaviors throughout their sexual life. The traditional inferior position of Chinese women relative to Chinese men in social relationships also reduces the former's ability to negotiate and persuade their male partners to engage in safer sex behaviors.

Research Objectives

Over the last two decades, there has been an accumulation of studies about AIDS/HIV-prevention behaviors of college students and various theoretical models that predict the adoption of safer sex behaviors. However, a vast majority of these studies were conducted in Western societies. Very few studies have examined Chinese college students' safer sex behaviors, and no study has applied the TRA to predict condom use among Chinese college students. The present study represents the first attempt to fill in these knowledge gaps, and aimed to explore safer sex behaviors and their correlates in a sample of Chinese college students in Hong Kong. Specific objectives of the studies included: (a) to examine current sexual behaviors of Chinese college students, (b) to explore their current safer sex behaviors and intention to adopt safer sex behaviors in the future, and (c) to investigate whether or not the various components of the TRA would predict safer sex behaviors among Chinese college students. Regarding the last objective, it was hypothesized that condom use would be related to accurate information about AIDS/HIV and less prejudice toward persons with AIDS/HIV. Perceived vulnerability to AIDS/HIV infection, which might motivate college students to adopt safer sex behaviors, would be related to condom use. Furthermore, normative beliefs about safer sex behaviors and perceived behavioral control as reflected by condom use self-efficacy would also be associated with condom use.

METHOD

Participants

A total of 455 Chinese college students from a major university in Hong Kong participated in the present study. Twenty-seven of them (5.9%) were eliminated from data analysis by the criteria of insincere test-taking, obvious response sets, or leaving a great number of items unanswered. Four hundred and twenty-eight valid cases with 161 males and 267 females constituted the present sample. Participants' ages ranged from 18 to 37, with the mean age being 21.05 years old. There was no significant age difference between males and females (Mean = 20.93, 21.12, respectively; t = -0.792, p [is greater than] .05). Participants came from different departments and years of study. Both undergraduate (92.8%) and postgraduate students (7.2%) were included in the sample.

Procedures

The researchers first translated the English measurement scales into Chinese, with the meaning and wording of the items as close to the original scales as possible. The Chinese questionnaire was then examined by an expert in AIDS/HIV research and by two service workers experienced in providing AIDS/HIV services to college students to ensure that the questionnaire was comprehensible and of relevance to the experience of the college students. The questionnaire was then given to five Chinese college students in Hong Kong who were asked to rate the items in terms of comprehension and ambiguity. The questionnaire was further refined on the basis of their feedback.

Participants were recruited in the libraries and dormitories of a major university in Hong Kong in 1999. A self-explanatory note stating the procedures and purposes of the present study was given to the participants before obtaining their consent to participate in the study. The questionnaires had no personal identifiable information of the participants to ensure confidentiality. The participants were also asked to put the completed questionnaires into an included self-addressed envelope, seal them, and mail them back to the researchers. Four hundred and fifty-five out of 500 questionnaire were received, yielding a response rate of 91%.

Instruments

The International AIDS Questionnaire - Chinese Version (IAQ-C). The 18-item IAQ-C was developed by Davis, Tang, Chan, and Noel (1999) to assess knowledge and attitudes toward AIDS/HIV among Chinese adolescents and college students. This scale has satisfactory internal reliability with a Cronbach alpha value of 0.76 (Davis et al., 1999). It has four subscales: seven items on transmission myths ("HIV can be spread through coughing and sneezing."), five items on attitudes ("People with AIDS/HIV should stay home or in a hospital."), three items on personal risks ("Asians are less susceptible of contacting AIDS/HIV than are Westerners."), and three items on facts ("Condoms will decrease the risk of AIDS/HIV transmission."). Davis et al. (1999) also reported the internal consistency of these four subscales, with alpha values being .71 and .68 for the subscales of transmission myths and attitudes and .45 for the personal risk and facts subscales. The IAQ-C is scored on a 5-point Likert scale ranging from 1 as strongly disagree to 5 as strongly agree, with higher relative mean scores indicating less endorsement of transmission myths, less prejudice, and greater perceived personal risks and factual knowledge about AIDS/HIV. For the present study, the personal risk and facts subscales yielded low internal reliability. As these two subscales were strongly correlated with each other (r = .70, p [is less than] .001), they were thus combined to form the vulnerability subscale. The internal reliability coefficients for the present sample were .84 for the transmission myths subscale, .71 for the attitudes subscale, .63 for the vulnerability subscale, and .85 for the total scale.

Self-efficacy Toward Condom Use. A 12-item scale developed by Schwarzer, Hahn, and Wegner (1993) was translated into Chinese to assess Chinese college students' perceived abilities and skills in the use of condoms when under sexually aroused or excited situations and in influencing or negotiating with their sexual partners on condom use. Sample items for this scale are "Even if I am extremely aroused sexually, I can still manage to use a condom" and "If my partner declines to use a condom, I don't insist upon it." Participants were asked to answer the items assuming that they had (a) sex partner(s) on a 4-point Likert scale ranging from 1 as very untrue to 4 as very true. Items were then recoded such that higher relative mean scores indicate higher levels of perceived self-efficacy toward condom use. For the present study, the internal consistency alpha for the scale was .79.

Attitudes Toward Safer Sex Behaviors. The scale on attitudes toward safer sex behaviors (Barker, Battle, Cummings, & Bancroft, 1998; Van, Marin, Tschann, Gomez, & Gregorich, 1998) was translated to Chinese. This scale consists of 17 items, with sample items being "Practicing safer sex is an act of respect and care for my partner" and "If my friends know that I use condoms, they would joke about me." The items are scored on a 4-point Likert scale ranging from 1 as strongly agree to 4 as strongly disagree. Higher relative mean scores indicate more positive attitudes toward safer sex behaviors. For the present study, the internal consistency coefficient for the scale was .76.

Current Sexual Behaviors and Safer Sex Behaviors. Following St. Lawrence, Eldridge, Reitman, and Little's (1998) format in assessing self-reported sexual behaviors, participants were asked to describe the frequency of their sexual activities in the past 12 months and their use of condoms in the past 12 months. These two items are scored on a 4-point format (1 = none at all, 2 = rarely, 3 = sometimes, and 4 = quite a lot). Higher scores indicate more frequent sexual activities and condom use.

Future Safer Sex Behaviors. Following St. Lawrence et al.'s (1998) format in assessing intentions for future safer sex behaviors, participants were asked the following questions: "When having sex activities in the future, how often will you use condoms?"; "How often do you think you would have a 'one-night stand' in the future?"; and "Assuming that you would have 'one-night stand', how often would you use condoms?" The items are also scored in a 4-point format similar to those for current sexual behaviors and safer sex behaviors. Higher scores indicate greater intentions to engage in casual sex and condom use in the future.

Exposure to AIDS/HIV information. Participants indicated whether or not they had read materials or attend any workshops or programs about AIDS/HIV and its prevention in the past 12 months.

Demographics. Participants were asked about their sex, age, and year of study.

RESULTS

Sexual Behaviors and Safer Sex Behaviors

Participants were asked to report their current sexual behaviors. Results showed that 76.1% of the participants reported abstinence in the past 12 months. Within the sexually active group (N = 102), 37.3% reported infrequent sexual activities, half of them engaged in sexual activities occasionally, and only 12.7% had frequent sexual intercourse. Among the sexually active participants, 13.8% never used condoms in the past 12 months, 23.5% rarely used condoms, 24.5% sometimes used condoms, and 38.2% had regular condom use. There was no gender difference regarding the mean scores for current sexual behaviors and condom use (p [is greater than] .05).

Only 23.4% of the participants intended to use condoms regularly when having regular sexual activities in the future, 45.2% would use condoms some of the time, and about 31.4% would rarely or never use condoms. With regard to intention to engage in casual sex encounters in the future, 3.7% of the participants reported that it was either "sometimes" or "very often" a possibility, and 26.7% indicated "rarely." Only 64.3% of the participants admitted their intention to use condoms regularly in their sexual experimentation.

Effects of Gender and Sexual Status on Future Safer Sex Behaviors

Table 1 summarizes the means and standard deviations for a variety of variables for total participants and broken down by gender. A 2 (Gender: Male, Female) X 2 (Sexual Status: Active, Inactive) analysis of variance was performed to determine the effects of gender and sexual status on participants' future safer sex behaviors. Results are summarized in Table 2. There was no significant Gender X Sexual Status 2-way interaction effect (p [is greater than] .05). A main effect for gender was found on intention for future casual sex encounters, with males being more likely than females to engage in a one-night stand (Mean = 1.52, 1.25, respectively; [F.sub.(1,414)] = 17.41, p [is less than] .005; d = .47). Results on sexual status main effects indicated that, as compared to those who were sexually inactive, participants who were currently sexually active were more likely to use condoms when having sexual activities in the future (Mean = 3.29, 2.57, respectively; [F.sub.(1,414)] = 39.49, p [is less than] .005; d = .71), to experiment with casual sex encounters (Mean = 1.48, 1.32, respectively; [F.sub.(1,414)] = 6.30, p [is less than] .01; d = .28), and to use condoms in future causal sex activities (Mean = 3.55, 3.32, respectively; [F.sub.(1,414)] = 5.23, p [is less than] .05; d = .23).

Table 1. Means and Standard Deviations of Major Variables and Intention
to Use Safer Sex Behaviors

                                             Male (N = 161)
                                        Relative Mean      SD

IAQ-C
  Transmission myths                         4.42          .55
  Attitudes                                  3.84          .68
  Vulnerability                              4.32          .54
  Total                                      4.22          .46
Self-efficacy toward condom use              2.83          .34
Attitude toward safer sex                    2.83          .27
Current sexual activity                      1.41          .81
Current condom use for currently             2.68         1.09
 sexually active(a)
Intention for future condom use              2.70          .94
Intention for future one-night stand         1.52          .65
Intention for condom use in future           3.22         1.09
 one-night stand

                                           Female (N = 267)
                                        Relative Mean      SD

IAQ-C
  Transmission myths                         4.23          .57
  Attitudes                                  3.78          .59
  Vulnerability                              4.17          .56
  Total                                      4.08          .45
Self-efficacy toward condom use              3.01          .35
Attitude toward safer sex                    2.92          .26
Current sexual activity                      1.42          .82
Current condom use for currently             2.98         1.06
 sexually active(a)
Intention for future condom use              2.77         1.05
Intention for future one-night stand         1.25          .50
Intention for condom use in future           3.46          .95
 one-night stand

                                       Total (N = 428)
                                        Relative Mean      SD

IAQ-C
  Transmission myths                         4.30          .57
  Attitudes                                  3.80          .63
  Vulnerability                              4.22          .56
  Total                                      4.13          .46
Self-efficacy toward condom use              2.94          .35
Attitude toward safer sex                    2.88          .27
Current sexual activity                      1.42          .82
Current condom use for currently             2.87         1.08
 sexually active(a)
Intention for future condom use              2.74         1.01
Intention for future one-night stand         1.35          .58
Intention for condom use in future           3.37         1.01
 one-night stand

(a) Total number of currently sexually active participants was 102,
with 38 males and 64 women.

Effects of Gender and Sexual Status on Components of TRA

Similar 2 (Gender) x 2 (Sexual Status) analyses of variances were performed to determine the effects of gender and sexual status on various components of the TRA as in AIDS/HIV-related knowledge and attitudes, normative beliefs about safer sex behaviors, and perceived behavioral control in condom use self-efficacy. No significant Gender X Sexual Status 2-way interaction effects were found (see Table 2). Significant gender main effects showed that, compared to females, males were more sensitive to their own vulnerability to AIDS/HIV and had less endorsement of transmission myths about AIDS/HIV, lower self-efficacy toward condom use, and more positive attitudes toward safer sex behaviors [(F.sub.(1,414)] = 8.05, 5.42, 20.33, 9.73, respectively; p [is less than] .05; d = .33, .28, -.51, -.33, respectively). On the other hand, significant sexual status main effects were not found on any components of the TRA.

Table 2. Summary of ANOVA Results for all Participants

                                  Gender X Sexual status     Gender
                                    2-way interaction      main effect
                                         F value             F value

IAQ-C
Transmission myths                        0.29                8.05(**)
Attitudes                                 0.71                1.83
Vulnerability                             0.08                5.42(*)
Total                                     0.12                8.75(**)
Self-efficacy toward condom use           0.16               20.33(***)
Attitudes toward safer sex                0.33                9.73(**)
Intention for future condom use           0.01                0.41
Intention for future one-night            0.06               17.41(***)
 stand
Intention to use condom in                1.35                2.40
 future one-night stand

                                      Sexual Status
                                       main effect          Observed
                                         F value            power(a)

IAQ-C
Transmission myths                        0.79                0.78
Attitudes                                 1.87                0.40
Vulnerability                             0.07                0.69
Total                                     0.56                0.82
Self-efficacy toward condom use           0.08                1.00
Attitudes toward safer sex                0.74                0.87
Intention for future condom use          39.49(***)           1.00
Intention for future one-night            6.30(**)            1.00
 stand
Intention to use condom in                5.23(*)             0.84
 future one-night stand

(a) Computing using alpha = .05

(*) p < .05. (**) p < .01. (***) p < .005.

Associations Between Components of TRA and Current Safer Sex Behaviors

Pearson correlation analyses were conducted to determine associations between current safer sex behaviors and TRA components. Bonferroni adjustments were made to minimize Type I errors, and only results with a p value less than .001 were considered significant (Table 3). Regarding the associations among various TRA components, participants with less endorsement of transmission myths and prejudice, higher sensitivity to their own vulnerability about AIDS/HIV, and higher self-efficacy toward condom use were more likely to demonstrate more positive attitudes toward safer sex behaviors. However, associations between specific AIDS/HIV-related variables with condom use efficacy and attitudes toward safer sex behaviors were more evident for females than for males.

Table 3. Correlation Matrix Among Variables for Total Participants
(N = 428)

                                    (1)       (2)       (3)      (4)

IAQ-C
  (1) Transmission myths            -
  (2) Attitudes                    .38(*)     -
  (3) Vulnerability                .51(*)    .35(*)     -
  (4) Total Scale                  .85(*)    .72(*)    .76(*)     -
(5) Self-efficacy toward           .18(*)    .17(*)    .27(*)    .25(*)
     condom use
(6) Attitude toward safer sex      .25(*)    .21(*)    .32(*)    .33(*)
(7) Current sexual activities     -.01       .01       .04       .01
(8) Intention for future           .07       .04       .13       .10
     condom use
(9) Intention for future          -.01       .01      -.02      -.01
     one-night stand
(10) Intention to use condom       .15(*)    .20(*)    .17(*)    .22(*)
      in future one-night stand

                                    (5)       (6)       (7)      (8)

IAQ-C
  (1) Transmission myths
  (2) Attitudes
  (3) Vulnerability
  (4) Total Scale
(5) Self-efficacy toward            -
     condom use
(6) Attitude toward safer sex      .61(*)     -
(7) Current sexual activities      .06       .03        -
(8) Intention for future           .23(*)    .18(*)    .32(*)     -
     condom use
(9) Intention for future          -.17(*)   -.18(*)    .11       .17(*)
     one-night stand
(10) Intention to use condom       .27(*)    .19(*)    .14       .35(*)
      in future one-night stand

                                    (9)       (10)

IAQ-C
  (1) Transmission myths
  (2) Attitudes
  (3) Vulnerability
  (4) Total Scale
(5) Self-efficacy toward
     condom use
(6) Attitude toward safer sex
(7) Current sexual activities
(8) Intention for future
     condom use
(9) Intention for future            -
     one-night stand
(10) Intention to use condom       .02        -
      in future one-night stand

(*) p < 0.001 (two-tailed) with bonferroni adjustment for Type I errors

Results also showed that participants' current sexual behaviors were unrelated to their self-efficacy toward condom use and their knowledge, attitudes, and perceived vulnerability about AIDS/HIV. Attitudes toward safer sex behaviors were also unrelated to participants' current sexual behaviors. Compared to those who were less sexually active or sexually inactive, participants who were currently more sexually active were more likely to use condoms in future regular sexual activities. Correlation results between current sexual behaviors and other variables were generally similar across genders.

The associations between current safer sex behaviors and components of the TRA were also calculated for participants who were sexually active and used condoms (N = 88). Results showed that current safer sex behaviors of sexually active participants were related to their level of self-efficacy toward condom use, positive attitudes toward safer sex, and likelihood to use condoms in future regular sexual activities (r = .46, .34, .58, respectively).

Associations Between TRA Components and Future Safer Sex Behaviors

Results in Table 3 also revealed that participants' intention to use condoms in future regular sexual activities was related to their level of self-efficacy toward condom use, attitudes toward safer sex behaviors, and likelihood to engage in and to use condoms in future casual sexual encounters. Participants' intention to engage in casual sex encounters was associated with a low level of self-efficacy toward condom use and negative attitudes toward safer sex behaviors. On the other hand, participants' intention to use condoms in future casual sexual encounters was related to the depicted components of the TRA in the expected directions.

Associations between future safer sex behaviors and components of the TRA broken down by gender were also examined. For intentions to use condoms in future regular sexual activities and to engage in future casual sex counters, correlation results were similar across genders with the exception of females' condom use in future regular sexual activities, which were unrelated to their intention to engage in a one-night stand. For males, future condom use in casual sex encounters was related to high scores on the IAQ-C total scale, self-efficacy toward condom use, and intention to use condoms in future regular sexual activities. For females, future condom use in casual sex was related to less prejudice toward AIDS/HIV, high scores on the IAQ-C scale, high self-efficacy toward condom use, and increased likelihood to use condoms in future regular sexual activities.

DISCUSSION

Safer Sex Behaviors of Chinese College Students

The investigators found that an increasing number of Chinese college students in Hong Kong engaged in sexual activities during the surveyed year. Results showed that 24% of the surveyed college students were sexually active in the past 12 months, reflecting a two-fold increase as compared to 11% reported by Tang, Lai, and Chung (1997), and a three-fold increase as compared to 8% reported by Chan and Cheung (1998). Despite these increases, sexual activities of Chinese college students are still much more infrequent than their Western counterparts. Typically, 70% to 80% of the college students in the United States are sexually active during their college years (DiClemente et al., 1990; Ratliff-Crain et al., 1999; Reinish et al., 1995; Sheer & Cline, 1994). The discrepancy in rates between the present study and previous work may be due to sampling and methodological differences. Emphasis on academic achievement, parental control, and lack of opportunity for sexual behaviors of Chinese students are also plausible reasons to account for the differences in sexual activities between Western and Chinese college students (Tang et al., 1997). Furthermore, this study has also shown that Chinese college students generally had accurate information and awareness of their vulnerability to AIDS/HIV, which may also have some impact on their sexual practices.

The present study also revealed slight increases in condom use among the sexually active Chinese college students, 86.2% as compared to 75% to 80% reported by Ling (1993). It was further noted that the present 86.2% condom use also included students who reported inconsistent use or rare use of condoms; thus, caution should be taken in interpreting this figure. In fact, only 38.2% of the present sample reported regular condom use, and this percentage was comparable to recent studies of Western college students (Ratliff-Crain et al., 1999). Furthermore, Chinese college students' tendency to engage in risky and unprotected sexual behaviors in the future is also an area of concern. About one third of the surveyed students would want to experiment with casual sex in the future, but only 64% of the students would use condoms regularly if they were to have future sexual encounters. With increases in their sexual experiences and experimentation but low rates of regular condom use, Chinese college students are similar to their Western counterparts as a group particularly vulnerable to contracting AIDS/HIV (Reinisch et al., 1995; Sheer & Cline, 1994).

The present investigators did not find gender differences with regard to Chinese college students' current sexual behaviors and current safer sex behaviors. However, males were more likely than females to report that they would engage in a one-night stand in the future. This gender difference may be due to the fact that females tend to focus on love and emotional factors when considering sexual relationships, while males may consider lust and physical pleasure as the more primary reasons to engage in a sexual relationship (Chan& Cheung, 1998; Whitley, 1988). These results may also reflect the subtle double standards put on the two genders by the society at large, in which women are honored for their asexuality while male sexuality is portrayed as a train that cannot be derailed once engaged (Lear, 1997).

Results of the present study indicate that compared to sexually active students, students who were sexually inactive at the time of survey had lower intentions to use condoms if they were to engage in regular or casual sexual activities in the future. Sexually inactive students may have generalized their current low risk for AIDS/HIV infection to future sexual encounters, and were not motivated to engage in protected sexual behaviors in the future (Ratfliff-Crain et al., 1999; Serovich & Greene, 1997). In addition, as they were currently sexually abstinent, they were probably not motivated to seek information about AIDS/HIV and safer sex behaviors to avoid being perceived by other students as sexually experienced or prepared (Lear, 1997). Thus, they might not have adequate knowledge or skills in protecting themselves against AIDS/HIV in future sexual encounters. Indeed, this second explanation was consistent with students' responses to questions regarding their participation in AIDS/HIV-prevention programs. Compared to those who were sexually active, sexually inactive students were less likely to read materials or participate in workshops or programs about AIDS/HIV prevention during the surveyed year. This result points to the importance of extending AIDS/HIV education and prevention programs to Chinese college students in Hong Kong.

Using the TRA to Predict Intentions to Adopt Safer Sex Behaviors

The present investigators found support for using the TRA to understand AIDS/HIV-prevention behaviors of Chinese students. In general, Chinese college students' knowledge and attitudes about AIDS/HIV, perceived behavioral control in condom use self-efficacy, and normative beliefs toward safer sex behaviors were related to each other and to safer sex behaviors in the expected directions. In particular, results showed that the TRA was most applicable in understanding Chinese college students' intention to use condoms in future casual sexual encounters, as all the selected components of the TRA were associated with this safer sex behavior in the expected directions. Specifically, students who would adopt protected sexual behaviors in future casual sex encounters tended to have accurate AIDS/HIV information, show little prejudice against people with AIDS/HIV, have greater awareness of their own vulnerability to contracting AIDS/HIV infection, feel confident in condom use, and have positive attitudes toward safer sex behaviors (e.g., Fisher et al. 1995; White et al., 1994).

Findings on using the TRA to predict Chinese college students' current condom use and intention to use condoms in future regular sexual activities were less encouraging. For the present study, AIDS/HIV-related variables such as knowledge, attitudes, and vulnerability were not linked to these two safer sex behaviors; instead, perceived behavioral control of condom use self-efficacy and normative attitudes toward safer sex behaviors were more predictive. Similar findings were also noted in previous studies with Western college samples (O'Leary et al., 1992; Ratliff-Crain et al., 1999).

Gender disparity was noted when associations among various components of the TRA and safer sex behaviors were examined separately for males and females. While results for females were similar to those for the total sample, males demonstrated a different pattern of associations among the correlates. For males, global scores on AIDS/HIV-related variables were related to condom-use efficacy and attitudes toward safer sex behaviors. Separating the global scores into distinct components of knowledge, attitudes, and vulnerability toward AIDS/HIV for males did not have any added value as these separate components were unrelated to the other components of the TRA. There were also slight variations across genders in the pattern of associations between TRA components and intention to adopt safer sex behaviors. In general, while females' safer sex behaviors were related to specific AIDS/HIV-related variables, similar behaviors in males were linked to global scores of variables on AIDS/HIV. The above findings were in line with previous studies on AIDS/HIV and its prevention, as gender differences have been consistently noted in the prevalence rates of the disease and in the predictors of various safer sex behaviors (CDC, 1996; Chan & Cheung, 1998; Morokof, Harlow, & Quina, 1995). Researchers have proposed that gender disparity regarding power in interpersonal relationship, sexuality, and responsibility to initiate sex and safer sex behaviors needs to be examined in more detail in order to design more effective AIDS/HIV prevention programs for both males and females (Doyal, 2000; Holland, Ramazanoglu, Scott, Sharpe, & Thompson, 1990).

Limitations and Implications for Future Studies

The present study attempted to minimize, but could not rule out, the shortcomings of the self-reported method, which may be subject to self-selection, social desirability, and recall biases. As participants were not asked about their marital status, some of them may have been married, especially among the graduate students. This may account for some participants reporting that they had frequent sexual activities but rarely used condoms. Furthermore, the scale options for the measures on sexual behaviors and condom use were too limited and lacked precision to differentiate among college students. For example, the scales did not include the options of responding always and never, and thus could not distinguish the fully at-risk students from the never at-risk students. The cross-sectional nature of the data also hindered causal-effect inferences. This study was conducted with a nonrandom sample of Chinese college students, and generalizability of the present findings to other samples was thus restricted. Caution should also be taken when extending the present results to Chinese college students residing in other Chinese societies such as urban and rural mainland China, Taiwan, and Singapore, as they may be subject to varying degrees of urbanization, socialization, and Western influences.

Future studies regarding safer sex behaviors among Chinese college students should include the examination of subtle social influences on their intention to adopt safer sex behaviors, especially whether or not there are different effects on the two genders. This would undoubtedly be difficult given the rather ambiguous operational definitions which have attempted to frame and measure the broad range of social influences over the years. Another line of research may focus on the possible causal relationships among AIDS/HIV knowledge, psychological factors, and safer sex behaviors. The present investigators have shown that the TRA provides a useful theoretical framework in mapping out the pathways of various psychological variables when predicting safer sex behaviors in Chinese samples.

CONCLUSION

AIDS/HIV prevention programs have been the focus for controlling the spread of AIDS/HIV since the epidemic was discovered. Effective interventions require the understanding of the target groups' characteristics, their particular stage of needs, and predicting factors specific to their behavior change. This exploratory study has shown that Chinese college students are a vulnerable group for AIDS/HIV infection, and a theoretical framework based on Western samples can be adopted to understand safer sex behaviors in Chinese. Future studies should work on the subtle influences of social factors as well as the causal link between psychological variables and safer sex behaviors.

REFERENCES

Abraham, C., & Sheeran, P. (1994). Modeling and modifying young heterosexuals' HIV-preventing behavior: A review of theories, findings, and educational implications. Patient Education and Counseling, 23, 173-186.

Ajzen, I., & Madden, T J. (1986). Prediction of goal-directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experimental Social Psychology, 22, 453-474.

Baldwin, J. D., & Baldwin, J. I. (1988). Factors affecting AIDS-related sexual risk-taking among college students. Journal of Sexual Relationships, 25, 181-196.

Bandura, A. (1994). Social cognitive theory and exercises of control over HIV infection. In R. J. DiClemente, J. L. Petersen, & J. Mann, (Eds.), Preventing AIDS: Theories and methods of behavioral interventions (pp. 25-59). New York: Plenum Press.

Barker, J. C., Battle, R. S., Cummings, G. L., & Bancroft, K. N. (1998). Condoms and consequences: HIV/AIDS education and African American women. Human Organization, 57, 273-283.

Basen-Engquist, K. (1994). Evaluation of theory-based HIV prevention intervention in college students. AIDS Education and Prevention, 6, 512-624.

Basen-Engquist, K., & Parcel, G. S. (1992). Attitudes, norms, and self-efficacy: A model of adolescents' HIV-related sexual risk behavior. Health Educational Quarterly, 19, 263-277.

Bellingham, K., & Gillies, P. (1993). Evaluation of an AIDS education program for young adults. Journal of Epidemiology and Community Health, 47, 134-138.

Brafford, L., & Beck, K. (1991). Development and validation of a condom self-efficacy scale for college students. Journal of American College Health, 39, 219-225.

Bryan, A. D., Aiken, L. S., & West, S. G. (1996). Increasing condom use: Evaluation of a theory-based intervention to prevent sexually transmitted diseases in young women. Health Psychology, 15, 371-382.

Centers for Disease Control and Prevention. (1991). Youth risk behavior surveillance study. Atlanta, GA: Author.

Centers for Disease Control and Prevention. (1996). HIV/AIDS surveillance report (through June 1996). Atlanta, GA: Author.

Chan, D. K., & Cheung, S. F. (1998). An examination of premarital sexual behavior among college students in Hong Kong. Psychology and Health, 13, 805-821.

Chan, D. K., & Fishbein, M. (1993). Determinants of college women's intentions to tell their partners to use condoms. Journal of Applied Social Psychology, 23, 1455-1470.

Chin, J. (1999). Estimation and projection of HIV infection and AIDS cases in Hong Kong. Hong Kong: Department of Health, Hong Kong SAR Government.

Davis, C., Tang, C. S., Chan, S. E, & Noel, B. (1999). The development and validation of the International AIDS Questionnaire-Chinese Version (IAQ-C). Educational and Psychological Measurement, 59, 481-491.

DiClemente, R. J., Forrest, K. A., & Mickler, S. (1990). College students' knowledge and attitudes about AIDS and changes in HIV-preventive. AIDS Education and Prevention, 2, 201-212.

Doyal, L. (2000). Gender equity in health: Debates and dilemma. Social Science & Medicine, 51, 931-939.

Fishbein, M., & Ajzen, I. (1975). Belief attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.

Fisher, W., Fisher, J., & Rye, B. (1995). Understanding and promoting AIDS-preventive behavior: Insights from the Theory of Reasoned Action. Health Psychology, 14, 255-264.

Greene, K., Rubin, D., & Hale, J. (1995). Egocentrism, message explicitness, and AIDS message directed toward adolescents: An application of the theory of reasoned action. Journal of Social Behavior and Personality, 10, 547-570.

Guizzardi, G., Stella, R., & Remy, J. (1997). Rationality and preventive measures: The ambivalence of the social discourse on AIDS. In L. V. Campendhoudt, M. Cohen, G. Guizzardi, & D. Hausser (Eds.), Sexual interactions and HIV risk: New conceptual perspectives in European research (pp. 159-180). London: Taylor & Francis.

Harden, A., & Ogden, J. (1999). Condom use and contraception non-use amongst 16-19 year olds: A within subjects comparison. Psychology and Health, 14, 697-709.

Holland, J., Ramazanoglu, C., Scott, S., Sharpe, S., & Thomson, R. (1990). Sex, gender, and power: Young women's sexuality in the shadow of AIDS. Sociology of Health and Illness, 12, 336-350.

Jemmott, J. B., Jemmott, L. S., & Hacker, C. I. (1992). Predicting intentions to use condoms among African-American adolescents: The risk-associated behavior. Ethnicity Discussions, 2, 371-380.

Joint United Nations Programme on HIV/AIDS & World Health Organization. (1999). AIDS epidemic update. Geneva, Switzerland: Author.

Lear, D. (1997). Sex and sexuality: Risk and relationships in the age of AIDS. Thousand Oaks, CA: SAGE Publications.

Ling, L. (1993). Sexual behavior of HKU students (1989-1992). Hong Kong, China: University Health Service, HKU.

Morokoff, P., Harlow, L., & Quina, K. (1995). Women and AIDS. In A. Stanton & S. Gallant (Eds.), The psychology of women's health: Progress and challenges in research and application (pp. 171-198). Washington, DC: American Psychological Society.

Neilsen, G., & Young, E (1994). HIV/AIDS advocacy and anti-discrimination legislation: The Australian response. International Journal of STD & AIDS, 5, 13-17.

O'Leary, A., Goodhart, F., Jemmott, L., & Boccher-Lattimore, D. (1992). Predictors of safer sex on the college campus: A social cognitive theory analysis. Journal of American College Health, 40, 253-263.

Ratliff-Crain, J., Donald, K. M., & Dalton, J. (1999). Knowledge, beliefs, peer norms, and past as correlates of risky sexual behaviors among college students. Psychology and Health, 14, 625-641.

Reinisch, J., Hill, C,, Sanders, S., & Ziemba-Davis, M. (1995). High-risk sexual behavior at a Midwestern university: A confirmatory survey. Family Planning Perspectives, 27, 79-82.

Rimberg, H., & Lewis, R. (1994). Older adolescents and AIDS: Correlates of self-reported safer sex behaviors. Journal of Research on Adolescence, 4, 453-464.

Schenker, I. (1989). Fighting AIDS -- an educational module for army units. Paper presented at the meeting of the International Conference on AIDS, Montreal, Canada.

Schwarzer, R., Hahn, A., & Wegner, M. (1993). Self-efficacy toward use of condoms. In R. Schwarzer (Ed.), Measurement of perceived self-efficacy: Psychometric scales for cross-cultural research. Berlin, Germany: Der Prasident der Freien Universitat Berlin.

Serovich, J. M., & Greene, K. (1997). Predictors of adolescent sexual risk taking behaviors which put them at risk for contracting HIV. Journal of Youth and Adolescence, 26, 429-444.

Sheer, V., & Cline, R. (1994). The development and validation of a model explaining sexual behavior among college students: Implications for AIDS communication campaigns. Human communication Research, 21, 280-304.

St. Lawrence, J. S., Eldridge, G. D., Reitman, D., & Little, C. E. (1998). Factors influencing condom use among African American women: Implications for risk reduction interventions. American Journal of Community Psychology, 26, 7-28.

Stripp, H., & Kerr, D. (1989). Determinants of public opinion about AIDS. Public Opinion Quarterly, 53, 98-106.

Tang, C. S., Lai, E D., & Chung, T. K. H. (1997). Assessment of sexual functioning for Chinese college students. Archives of Sexual Behaviour, 26, 79-90.

The Family Planning Association of Hong Kong. (1994). Youth sexuality study (out-of-school survey)-1991. Hong Kong, China: Author.

Van, B., Marin, O., Tschann, J. M., Gomez, C. A., & Gregorich, S. (1998). Self-efficacy to use condom in unmarried Latino adults. American Journal of Community Psychology, 26, 53-71.

Westerman, P., & Davidson, P. (1993). Homophobic attitudes and AIDS risk behavior of adolescents. Journal of Adolescent Health, 14, 208-213.

White, K., Terry, D., & Hogg, M. (1994). Safer sex behavior: The role of attitudes, norms, and control factors. Journal of Applied Social Psychology, 24, 2164-2192.

Whitley, B. E. (1988). The relation of gender-role orientation to sexual experience among college students. Sex Roles, 19, 619-638.

Wight, D. (1992). Impediments to safer heterosexual sex: A review of research with young people. AIDS Care, 4, 11-21.

Winslow, R., Franzini, L., & Hwang, J. (1992). Perceived peer norms, casual sex, and AIDS risk prevention. Journal of Applied Social Psychology, 22, 1809-1827.

Wulfert, E., & Wan, C. (1993). Condom use: A self-efficacy model. Health Psychology, 12, 346-353.

Yzer, M., Fisher, J., Bakker, A., Siero, W., & Misovich, S. (1998). The effects of information about AIDS risk and self-efficacy on women's intention to engage in AIDS preventive behavior. Journal of Applied Social Psychology, 28, 1837-1852.

Zhang, K., Li, D., Li, H., & Beck, J. (1999). Changing sexual attitudes and behaviors in China: Implications for the spread of HIV and other sexually transmitted diseases. Aids Care, 11, 581-589.

Manuscript accepted February 27, 2001

Address correspondence to Professor Catherine So-kum Tang, Department of Psychology, The Chinese University of Hong Kong, Shatin, N.T, Hong Kong, e-mail: [e-mail].