Longitudinal Effects of Perceived Maternal Approval on Sexual Behaviors of Asian and Pacific Islander (API) Young Adults
- First Online:
- Cite this article as:
- Hahm, H., Lee, J., Zerden, L. et al. J Youth Adolescence (2008) 37: 74. doi:10.1007/s10964-007-9234-y
Data were obtained from the National Longitudinal Study of Adolescent Health to examine the longitudinal association between Asian and Pacific Islander (API) adolescents’ perceptions of maternal approval of their sexual activity and contraception use, and four sexual outcomes during young adulthood. The study includes a nationally representative sample of 1,195 API adolescents. Logistic regression analyses were used to estimate the associations between predictors (Wave I) and outcomes (Wave III), controlling for covariates. API adolescents who perceived that their mothers approved of their sexual activities were more likely to have engaged in sex before age 15, contracted HIV/Sexually Transmitted Diseases, had multiple sex partners, and paid money for sex during young adulthood. Findings highlight the need for parent–adolescent communication and parental involvement in preventing negative sexual health outcomes among API adolescents transitioning to young adulthood.
KeywordsMaternal approvalAdolescent sexual risk behaviorsHIV/STDsAPIAsian AmericansAsian cultureAdolescents’ perceptionContraceptionAsian American sexualityHealth risk behaviors
The population of Asians and Pacific Islanders (APIs) in the United States (U.S.) is one of the fastest growing racial/ethnic minority groups (Barnes and Bennett 2000). Sexually Transmitted Diseases (STDs) are among the most common infectious diseases in the U.S. (Department of Health and Human Services 2002). The latest epidemiologic data show that APIs are not immune to HIV/AIDS risks. From 1999 to 2003, the number of AIDS cases increased at a higher rate (35%) for APIs than for other minority groups (African Americans had an increase of 8%, Latinos 7%, and Native Americans 21%) (Centers for Disease Control and Prevention 2005a; Choi et al. 2005). Heterosexual contact accounts for 80% of HIV transmissions among API women, the highest number compared to other racial or other ethnic groups (White 65%, Black 74%, and Latina 69%) (Centers for Disease Control and Prevention 2005a). API adolescents have shown more inconsistent patterns of condom use than other ethnic adolescents (Schuster et al. 1998). Furthermore, API communities often fail to utilize HIV/STD prevention programs resulting in low levels of awareness about HIV/STD prevention (Strunin 1991).
Despite recent advances in prevention, diagnosis, and treatment of certain STDs, it is estimated that 19 million new infections occur each year, almost half of them among adolescents and young adults ages 15–24 (Centers for Disease Control and Prevention 2002; Weinstock et al. 2004). Adolescents and young adults are at high risk for exposure to STDs as a consequence of engaging in sexual risk behaviors; recent findings indicate that both sexual risk behaviors and HIV/STD have increased among young adults. Specifically, the cumulative number of AIDS cases documented between 2001 and 2005 was highest among individuals between 20 and 24 years of age when compared to all other age categories. In 2005, both syphilis rates in women and gonorrhea rates in men were highest in the young adult group compared to other age groups (Centers for Disease Control and Prevention 2005b). Without timely and appropriate treatment, many STDs can cause serious complications such as pelvic inflammatory disease, infertility, complications of pregnancy, fetal loss, and cervical cancer (Hou and Basen-Engquist 1997).
Sexual risk behaviors, such as early age of sexual debut, paying money for sex, and having multiple sex partners, place individuals at increased risk of contracting and transmitting HIV/AIDS as well as other STDs (Santelli et al. 2000). Among urban minority adolescents, early age of sexual debut occurring in the 7th and 8th grades was found to increase the likelihood that these adolescents would be involved in coercive sexual intercourse, sex while using drugs or alcohol, sex with multiple sexual partners, and a greater frequency of sexual intercourse without condoms (O’Donnell et al. 2001). Other research findings have shown that White and African-American young adults with multiple sex partners were more likely to use drugs and have a history of gonorrhea (Johnson et al. 1994). Further, men who paid for sex were more likely to have multiple concurrent sex partners and engage in sex without condoms (Busch et al. 2002).
Promoting responsible sexual behavior among adolescents who are transitioning to young adulthood is a significant health imperative, as noted by the U.S. Surgeon General’s Public Health Priorities (Department of Health and Human Services 2002). Given this emphasis, identifying specific factors associated with STDs and sexual risk behaviors is a critical step in designing effective prevention programs for adolescents and young adults. Given the current trend among APIs, understanding risk factors associated with sexual behaviors, specifically for the API population, is quite important. This study focuses on the impact of API adolescents’ perceptions of maternal approval of early sexual activity and maternal approval of contraceptive use on sexual risk behaviors in young adulthood.
Sexual Risk Behaviors Among API Adolescents and Young Adults
There is limited knowledge about predictors of sexual outcomes among API adolescents who are transitioning to young adulthood. Researchers may have neglected this area of study, subscribing to the notion that Asians are the model minority as evidenced by low rates of adolescent sexual intercourse (Hou and Basen-Engquist 1997) and fewer lifetime partners for vaginal intercourse (Schuster et al. 1998). Contrary to the notion that APIs have a lower risk of HIV and STDs, studies have consistently found that after having sexual intercourse for the first time, their risks are as great as or greater than those of other ethnic groups. Sexually active API adolescents were more likely than their White counterparts to have had multiple sex partners in the past three months (Hou and Basen-Engquist 1997). Further, Chinese and Filipino students had lower HIV prevention scores than White students as well as reduced abilities to discuss HIV disease and prevention (Horan and DiClemente 1993).
Among API young adults, evidence documenting the sexual behaviors of API men shows an intriguing pattern. Approximately 90% of API men over the age of 18 in Houston, Texas, reported the highest rates of sexual activity in the past three months compared to African-Americans (78.2%), Hispanics (79.5%), and Whites (78.1%) (Ross et al. 2003). Approximately four out of ten API college students had a lifetime prevalence of unprotected sex and a majority of those who ever had anal sex did not use condoms (So et al. 2005). Moreover, a cross-sectional study of predictors of STDs among a nationally representative sample of API young adults found that 4% of the sample had chlamydia, a prevalence rate similar to that of the non-API young adult population in the U.S. (Hahm et al. (forthcoming)). Another survey of API college students in Southern California found that approximately one in five women reported that they had been possibly exposed to STDs and one in ten women reported that they had been treated for an STD (Cochran et al. 1991). Therefore, it is important to dispel the myth that STDs and sexual risk behaviors are not a public health concern in API communities.
Parental Influences on Sexual Risk Behaviors
For decades, scholarly attention has been paid to the issue of adolescents’ perceptions of parental attitudes toward sex, exemplified by Jessor and Jessor (1975) and Newcomer and Udry (1984). Recent studies have begun to document a strong relationship between adolescents’ perceptions of parental attitudes and adolescents’ sexual risk behaviors. For example, mothers’ permissiveness toward an early age of sexual debut has been found to increase the likelihood of sexual risk behaviors among adolescents (Dittus and Jaccard 2000; Hogan and Kitagawa 1985). However, previous studies are limited in their explanations of ways parental factors influence the occurrence of sexual risk behaviors during young adulthood.
It is commonly thought that most parents are opposed to their children having sexual intercourse during adolescence and would choose to prevent that behavior among their adolescents if they knew how (Newcomer and Udry 1984). However, some parents do not disapprove of adolescents’ sexual activities; they believe that experiencing sexuality is a “normal” adolescent developmental task (Hofferth and Hayes 1987). Findings from a survey of 751 Black adolescents documented that, although a majority of mothers disapproved of their adolescents having sexually intimate relationships, one in four mothers either approved or had ambivalent feelings about this (Jaccard et al. 1996).
Ecodevelopmental theory (Perrino et al. 2000) has contributed to our understanding of the roles of parents in the context of adolescents’ development of sexual risk behaviors. Ecodevelopmental theory points out that the parents are the most proximal and fundamental system that can influence adolescents’ behavior. Particularly, parents’ value-centered discussions with adolescents are critical in shaping adolescents’ values, attitudes, and intentions, thereby impacting risky behaviors. Thus, ecodevelopmental theory emphasizes that interventions should be targeted to interactions within the family.
Adolescents’ Perceptions of Mothers’ Attitudes Toward Adolescents’ Sexual Activities
The impact of maternal attitudes on adolescent sexual behaviors has been consistently reported by Jaccard et al. (1996) and Dittus and Jaccard (2000). In a study using a nationally representative sample of adolescents, adolescents’ perceptions of maternal disapproval regarding sexual activities were found to be a protective factor that predicted adolescents’ sexual activity and pregnancy in a subsequent year (Dittus and Jaccard 2000). Similarly, Jaccard et al. (1996) reported that adolescents’ perceptions of their mothers’ disapproval of sex was related to a decrease in the adolescents’ sexual activities.
Because the majority of attention to these issues has focused on White and African-American adolescent populations, discretion is needed when applying findings to API families due to the differences in cultural values and norms. Traditional API parents dissuade children’s independence and instead encourage parent–child interdependence, perceiving adolescents’ premature sexual behaviors to be a “loss of face” that reduces family status and brings shame to family honor (Kim and Wong 2002). Within these traditional cultural values, sexuality is viewed as a private matter, and sexual restraint and strict moral and social codes of behavior are emphasized (Abrahams 1999; Kim and Wong 2002). However, given the minimal focus on APIs in this research area, the connection has not been established between two key factors: (a) adolescents’ perceptions of parental approval or disapproval of sexual activities and contraceptive use, and (b) the impact of these perceptions on adolescents’ future sexual behavior.
Adolescents’ Perceptions of Mothers’ Attitudes Toward Contraception
Related to adolescents’ sexual activities but separate from them is the question of parental approval of contraception (Jaccard et al. 1996). Should parents even discuss the use of contraception with their adolescents? Opponents of such parental discussions fear that it sends a message condoning sexual behavior or at least opens the door to sexual behavior, thus sending a mixed message. Proponents argue that adolescents should be equipped with comprehensive information as a means to make informed health decisions. Proponents of such discussions also worry that parents’ sole emphasis on abstinence may result in increased sexual activity due to adolescents’ lack of understanding of sexual consequences and lack of useful information on how to use contraceptives effectively (Jaccard et al. 1996).
Two empirical studies provide a rather complex picture of the relationship between perceived parental attitudes about contraception and adolescent sexual behavior. One cross-sectional study using 751 African American inner city adolescents reported that adolescents’ birth control discussions with parents increased the likelihood of first intercourse among these adolescents. However, among adolescents who were already sexually experienced, birth control discussions were not associated with the frequency of sexual intercourse (Jaccard et al. 1996). A prospective study using Adolescent Health (Add Health) data that included a range of racial groups found that the odds of sexual intercourse were twice as high among adolescents who perceived high parental approval of contraception than among adolescents who perceived low parental approval (Dittus and Jaccard 2000).
Although prior studies suggest that API parents hold conservative attitudes toward their adolescents’ sexual expression (Abrahams 1999; Kim and Wong 2002), empirical research is still needed to document to what extent API adolescents perceive their mothers’ approval or disapproval of sexual activities and contraception, and the relationship of these perceptions to sexual outcomes. A few studies have examined parent–child communication about sexuality among APIs (Chung et al. 2005; Kim and Ward 2007); however, the longitudinal association between parent–adolescent communication and adolescent sexual outcomes has rarely been assessed.
Other Variables Influencing Adolescent Sexual Outcomes
Several additional factors have been found to influence sexual risk behaviors. These include gender, age, adolescents’ school performance as measured by grade point average (G.P.A.), place of birth, depressive symptoms, self-esteem, parental education, and adolescents’ attachment to parents. Concerning gender, men are more likely to report more sex partners (Santelli et al. 2000). Age is a contributor in that older adolescents are more likely to be sexually active, and they are less likely to be controlled by their parents (Newcomer and Udry 1984). A previous study (Choi 2007) of G.P.A. showed that lower G.P.A. scores were associated with a higher likelihood of sexual outcomes including sexual intercourse, pregnancy, and an STD diagnosis among adolescents across the races. Concerning place of birth, a recent study found that U.S.-born adolescents had twice the risk of engaging in sexual activities than adolescents who were born in foreign countries (Hussey et al. 2007). Further, increased identification with American culture, has been found to be a risk factor for earlier sexual activity among API adolescents (Hahm et al. 2006).
A mood disturbance, such as experiencing symptoms of depression, can affect sexual risk behaviors. One study (Shrier et al. 2002) found that higher levels of depressive symptoms among boys increased the odds of being diagnosed with STDs. Lehrer et al. (2006) also found that a high level of depressive symptoms predicted not using condoms and not using birth control during the last sexual encounter. Self-esteem too can influence sexual risk behaviors. Ethier et al. (2006) found that among sexually active adolescent girls, those who had lower self-esteem at baseline reported earlier sexual debut as well as a greater likelihood of having had risky sexual partners at a six-month follow-up.
Two parent-related variables are also important: parental education and adolescents’ attachment to parents. Newbern et al. (2004) found that lower maternal education was associated with increased rates of STDs among Black adolescent boys and girls. A parents’ low level of education increased the likelihood of an adolescent being sexually active. Only 28% of adolescents whose parents were college graduates were sexually active compared to 51% of adolescents whose parents had a high school education, GED, or less (Cubbin et al. 2005). Lower adolescent attachment to parents was found to be associated with higher rates of sexual intercourse among a nationally representative sample of adolescents in grades 7–11 in a subsequent year (Dittus and Jaccard 2000). Jaccard, Dittus, and Gordon also found that adolescents who reported a low level of satisfaction with their mothers were more than twice as likely to be sexually active compared to those who reported a high level of satisfaction (Jaccard et al. 1996). Thus, studies of predictors of sexual outcomes among API adolescents need to take such variables into consideration.
The present study has three aims. The first is to measure the extent to which API adolescents perceive that their mothers approve of their sexual activities and use of contraception. Based on the literature, we hypothesize that the majority of API adolescents will perceive their mothers to be disapproving of sexual activities and use of contraception.
The second aim is to examine the longitudinal link between API adolescents’ perceptions of maternal approval of their sexual and reproductive behaviors, and their later sexual risk behaviors. Perceived parental permissiveness has been consistently linked to sexual risk behaviors among adolescents from other ethnic groups, and this trend is likely to hold for API adolescents as well, particularly because API cultural values related to sexual involvement are so strong. Therefore, we hypothesize that API adolescents’ perceptions of maternal approval of sexual activities will be strongly associated with sexual risk behaviors.
The third aim is to explore the relationship between adolescents’ perceptions of maternal approval of contraception and sexual risk behaviors. No hypothesis has been developed for this relationship due to the mixed findings from prior studies. On the one hand, higher levels of perceived approval of birth control may reflect more permissive attitudes of mothers and thus be associated with more sexual risk behaviors. On the other hand, because adolescents receive safer sex messages from their own mothers, a higher level of perceived approval of birth control may be associated with fewer sexual risk behaviors. Data from our study will suggest a hypothesis that can be tested in future studies. To our knowledge, this study is the first examination of adolescents’ perceptions of maternal approval of sexual activities and use of contraception using a longitudinal representative sample of API adolescents.
Data are derived from in-home interviews conducted during Wave I (1995) and Wave III (2001) of the National Longitudinal Study of Add Health. Add Health was originally designed to provide information on the general health, mental health, and sexual and reproductive health status of a large nationally representative sample of adolescents in the U.S.. Detailed descriptions of the sampling design and Add Health procedures have been documented elsewhere (Bearman et al. 1997; Miller et al. 2004; Resnick et al. 1997). In order to maintain confidentiality, Add Health data were collected using a laptop computer. Audio Computer-Assisted Self-Interview (A-CASI) was used to collect information on sensitive topics, such as all questions about sexual behavior and history of STDs. A-CASI technology increases comprehensive reporting of sexual behaviors (Hosmer and Lemeshow 1989).
Add Health currently includes three sets of data: Wave I (1994–1995), Wave II (1995–1996), and Wave III (2001–2002); the same individuals were interviewed at three time points from adolescence to young adulthood. The total number of API adolescents who participated in Wave I was 1,584. Data collection for Wave III was conducted nationwide including Hawaii and Alaska. In Wave III, a nationally representative sample of APIs (n = 1,195) was collected. For this study, predictors are abstracted from Wave I, outcome variables are abstracted from Wave III, and the total sample (n = 1,195) is used.
Adolescents’ perceptions of mothers’ approval of sexual activity: The item measuring perception of mothers’ disapproval of sexual intercourse is, “How would your mother feel about your having sex at this time in your life?” Original Add Health responses range from 1 (strongly approve) to 5, (strongly disapprove). For this study, scores are reverse coded such that higher scores indicate a greater degree of approval.
Adolescents’ perceptions of mothers’ approval of contraception use: The item measuring perception of mothers’ approval of contraception use is scored from 1 (strongly approve) to 5 (strongly disapprove) and asked, “How would your mother feel about your using birth control at this time in your life?” For this study, this variable is also reverse coded with higher scores indicating greater degrees of approval.
Sexual risk behaviors: Three variables are assessed as sexual risk behaviors. Each question is dichotomized: (a) whether respondents had sex before the age of 15 was determined using, “How old were you the first time you had vaginal intercourse?”; (b) whether respondents paid money for sex was determined using, “Have you ever paid someone to have sex with you?”; and (c) whether respondents had multiple sex partners was determined by, “With how many different partners have you had vaginal intercourse in the past 12 months?”
Sexually Transmitted Diseases outcome: This variable is measured using the question, “Have you ever been told by a doctor or nurse that you had...?” for each of the following STDs: chlamydia, syphilis, gonorrhea, HIV/AIDS, genital herpes, genital warts, trichomoniasis, and human papilloma virus. Responses are added and dichotomized.
There are eight covariate variables. Gender is measured as a dichotomous variable. Age is measured in years.
Grade point average is measured by averaging respondents’ grades in English, mathematics, history or social science, and natural science. The mean of G.P.A. is 3.0 (SD = 0.2).
Place of birth: This study uses place of birth as a proxy for acculturation. This variable is dichotomized (U.S.-born vs. foreign-born).
Depressive symptoms: All measures of depressive symptoms are self-reported from Wave I using 19 items about being more easily bothered, loss of appetite, changes in energy level, trouble with improving bad moods, feelings of self-worth, hope about the future, ability to focus, changes in personal interactions, and feelings of depression. Each statement is scored from 0 to 3 such that higher scores indicate greater depression. Shrier et al. (2002) demonstrated that this scale has construct validity. Cronbach’s alpha coefficient for this study sample was 0.86. The mean of depressive symptoms was 13.0 (SD = 0.18, range = 0–51).
Self-esteem: Adolescents’ self-esteem is measured by ten items which ask about energy level, health and recovery, coordination, personal fitness, social and personal acceptance, and self-evaluation of problem solving. Responses are on an ordinal scale ranging from 1 (strongly agree) to 5 (strongly disagree). All scores are reverse coded and assigned such that higher scores indicate greater self-esteem (α = 0.88). This measure was successfully used by Resnick et al. (1997). The mean score for self-esteem was 23.1 (SD = 0.15, range = 11–46).
Parental education: Two parental education variables are assessed, years of education of each parent, and these are combined to determine parental educational level. This score is assigned according to the higher parent score or the single score for respondents reporting information on only one parent. This variable is also successfully used by Cuffee et al. (2007). The mean score for parental education was 6.8 (SD = 0.5, range = 0–9).
Adolescents’ attachment to parents: This is measured by 12 questions concerning warmth, feeling loved and encouraged by mother and father, satisfaction with parental relationships, and satisfaction with communication style and closeness and intimacy with parents. Responses ranged from 1 to 5, with higher scores indicating greater parental attachment (α = 0.86). The mean of parental attachment is 50.6 (SD = 0.20, range = 18–60). This scale was adopted by Resnick et al. (1997) in prior research and was also successfully used in a previous study (Hahm et al. 2003).
Stata 9.0 was used for all analyses. The research team used Stata survey procedures for all analyses to account for the clustered sampling design, regional stratification, and population weights. All reported estimates, including descriptive statistics, were weighted according to the population-based sampling weights. Logistic regression analyses were used to estimate associations between Wave I predictors and Wave III outcomes, while controlling for gender, age, G.P.A., place of birth, depressive symptoms, self-esteem, parental education, and adolescents’ attachment to parents. These variables were controlled as covariates in the model in order to avoid confounding relationships.
The characteristics of our sample in Wave III are well described in previous studies, see, for example, Hahm et al. (forthcoming). The sample comprises young adults from more than six Asian ethnic backgrounds (e.g., Asian Indian, Chinese, Filipino, Japanese, Korean, and Vietnamese), with Filipino being the plurality (37.8%). Gender was evenly distributed. The mean age of all API young adults was 22 years old, and approximately 66% of respondents were between the ages of 22 and 24. Approximately 90% of our sample reported speaking English at home.
Adolescents’ Perceptions of Maternal Attitudes Toward Sexual Activities and Contraception
Weighted percentage of distribution of adolescents’ perception on mothers’ approval of sexual activities and use of contraception, Add Health, Wave I
Adolescents’ perceptions of mothers’ approval of sexual activities (%) (n = 1,124)
Adolescents’ perceptions of mothers’ approval of using contraception (%) (n = 1,112)
Longitudinal Effects of Perception of Maternal Attitudes toward Adolescents’ Sexual Activities on STDs and Sexual Risk Behaviors
Odds ratios (and 95% confidence intervals) from multivariate analyses assessing longitudinal association between perception of mothers’ approval of sexual activities and contraception and sexual outcomes
First sex before age 15 (n = 657)
Paying money for sex (n = 330)
Multiple sex partners (n = 478)
STDs (n = 648)
Odds ratio (95% CI)
Odds ratio (95% CI)
Odds ratio (95% CI)
Odds ratio (95% CI)
US birth status
Adolescents’ perception of mothers’ approval of sexual activities
Adolescents’ perception of maternal approval of contraception use
This prospective study of APIs examined how adolescents’ perceptions of their mothers’ approval of their sexual activities and use of contraception affected their likelihood of having STDs and engaging in various sexual risk behaviors. This study yielded two principal findings regarding adolescents’ perceptions and risk behavior outcomes.
High Proportion of Adolescents Perceived Mothers’ Disapproval of Sexual Activities
The overwhelming majority of API adolescents (89%) perceived that their mothers would disapprove of their involvement in sexual activities. Only 1.4% of API adolescents perceived that their mothers would be likely to approve of their engagement in sexual activities while nearly 10% reported ambivalent attitudes. This perception is probably accurate because studies have shown that Asians compared to other racial groups espouse more conservative attitudes toward their adolescents’ sexual activities (Meston et al. 1998), and API mothers tend not to regard sexual behaviors as a “normal” part of healthy adolescent development (Kim and Ward 2007). This is true for Filipinos as well as for East Asians (Espiritu 2001). This finding, from a nationally representative sample of APIs, is very similar to results from a survey of a Los Angeles County school district which found that 88% of APIs reported that their mothers would disapprove if respondents had vaginal intercourse (Schuster et al. 1998).
In contrast to the adolescents’ perceptions of high maternal disapproval of sexual activities, 25% of the adolescents from our study perceived that their mothers would be likely to approve of their contraceptive use. The correlation between these two variables is moderate (r = 0.43, p = 0.001). Thus, API adolescents perceived that maternal attitudes toward sexual activities were quite different from those toward contraceptive use. Furthermore, perceptions of mothers’ approval of contraceptive use was not predictive of sexual risk behaviors, which suggests that adolescents do not equate approval for contraception with approval for sexual activities.
If adolescents correctly perceived their mothers’ view of contraceptive use, it is possible that the mothers’ approval came only after mothers discovered that their adolescents had already had sex. API mothers may disapprove of their adolescents’ sexual activities, but if the behavior has already occurred, they may be more likely to approve of contraceptive use as a precaution against negative health consequences. Thus, maternal attitudes could be consistent in disapproving of sexual activities and approving of contraceptive use.
Follow-up analyses were done to compare whether APIs and other ethnic groups differed in terms of the proportion of perceived mothers’ disapproval of sexual activity and contraceptive use. APIs’ proportion of perceived mothers’ disapproval of sexual activity (89%) was higher than that of Whites (83.4%), African Americans (71.8%), and Latinos (79.6%). Likewise, API’s proportion of perceived mothers’ disapproval of contraceptive use (57.3%) was higher than Whites (45.9%), African Americans (36.3%), and Latinos (50.0%). These results suggest that the traditional Asian value of sexual conservatism may influence communication about sex related topics between parents and adolescents in API families (Kim and Wong 2002). These cross-cultural differences in perceived maternal approval of adolescents’ sexual behaviors, and why they exist, constitute areas that should be explored in-depth in future research.
Adolescents’ Perceptions of Maternal Approval of Sexual Activities is a Risk Factor Predicting Sexual Risk Behaviors
Adolescents’ perceptions of their mothers’ permissive attitudes about sexual activities emerged as a predictor of STD diagnosis, paying money for sex, having multiple sex partners during young adulthood, and early sex debut. These results are congruent with the findings of other researchers (Donenberg et al. 2002; Jaccard et al. 1996). The multivariate model confirmed our hypothesis of the significant influence of mothers’ approval on sexual risk behaviors. This provides evidence of the critical role parents play in young people’s health risk behaviors.
However, a question remains: If adolescents’ perceptions are correct, why would maternal disapproval of sexual activities be so powerful in protecting adolescents from risk-taking behaviors over time? Ecodevelopmental theory suggests that such maternal disapproval is an extension of parental monitoring or behavioral control (Perrino et al. 2000). If adolescents perceive that their parents’ monitoring of their behavior is evidence of concern or caring, adolescents may internalize this orientation, choose to avoid engaging in sexual risk behaviors, or otherwise act in ways that ultimately minimize parental concern. API adolescents may equate some aspects of strictness or firm control and governance of children with parental caring and involvement. Chinese parenting has been found to be more “controlling” or “authoritarian” (reflecting the expectation of children’s unquestioning obedience) compared to European American parents’ more “authoritative” style (reflecting the expectation of children’s mature behavior and open, two-way parent–child communication) (Chao 2000). Values associated with parenting in Korean culture are similar. Korean adolescents’ perceptions of parental control were found to be positively correlated with perceived parental warmth and low neglect, as long as there was no “very strict parental control” (Rohner and Pettengill 1985); Korean children perceived that parental control was a legitimate right and responsibility of parents.
If adolescents lack a connection with parents, they may become over-reliant on peers. For API adolescents in the U.S., the normal psychosocial stresses of transitioning to adulthood can be greatly intensified by increased exposure to Western culture and conflicting expectations from their families and peer groups (Vega et al. 1993). In addition to acquiring language competence, children of immigrant parents must function in an individualistic culture that is quite different from the culture in which their parents were raised. Like most adolescents, API adolescents struggle daily with peer acceptance. Adolescents who feel less connected to their parents or do not perceive clear parental limits and expectations may rely even more on peers or romantic partners for interpersonal connections (Hahm et al. 2004). Future research should explore the mediating role of adolescents’ peers in the development of sexual risk behaviors.
Adolescents’ perception of mothers’ approval of contraception did not have any longitudinal effect on any sexual outcomes. Reasons for this lack of association are not immediately clear. One possibility is that mothers conveyed approval of contraceptive use, but provided insufficient information about types of contraceptives and ways to access and use them, resulting in no positive or negative impact. Thus, the level or type of maternal communication related to contraceptive use may have been inadequate.
It is important to note that our study focused on adolescents’ perceptions of their mothers’ attitudes about sexual behavior; it did not focus on their mothers’ actual attitudes. We focused on adolescents’ perceptions because it is these, rather than mothers’ actual attitudes, which were found by other researchers (Dittus and Jaccard 2000) to be more powerful in influencing adolescents’ sexual behaviors. This suggests that the sexual outcomes of youth are more determined by the youths’ own interpretations of parental communication than by the parental messages themselves. Given the aforementioned findings of other researchers, our decision to use adolescents’ perceptions of maternal attitudes is a methodologically sound approach.
Further, existing evidence reveals discrepancies in mother and adolescent reports of their joint communication about adolescent sexual activities and contraception (Dittus and Jaccard 2000; Donenberg et al. 2002; Jaccard et al. 1998): mothers reported greater levels of actual disapproval than their teens perceived (Jaccard et al. 1998). Our findings are somewhat consistent with this. The correlation between mothers’ actual approval and adolescents’ perceptions in our sample was low (r = 0.19, p = 0.001), suggesting that these two variables are weakly associated and discordant.
Readers should consider several limitations in assessing findings from this study. First, our outcome variable of an STD diagnosis was based on self-reports. Using self-report data on STDs can be less than ideal due to possible measurement error stemming from participants’ false reports of their STD diagnoses or lack of awareness of their STD status, resulting in underreporting (Turner et al. 1998). Second, like most observational studies, there are potential biases from unmeasured confounders, such as parental rearing practices (Minuchin 1974), family values about sexuality which need to be further unpacked, and adolescent personality dimensions such as risk-proneness, which may be associated with sexual outcomes (Kalichman et al. 2003).
Despite the aforementioned limitations, this study contributes to our knowledge of adolescent and young adult sexual behavior. It adds to accumulating evidence on the relationship between adolescents’ perceptions of parental approval of adolescent sexual activity and contraception use gleaned from studies of other ethnic groups, suggesting that parental direction can serve as a protective factor. Our prospective design and analytic approach—which controlled for important confounders such as adolescents’ attachment to parents, depressive symptoms, and which used an adequate sample size with minimal loss to follow-up—permitted identification of predictors that have not been addressed in previous studies.
To further advance a precise understanding of maternal attitudes toward adolescents’ sexual behaviors, we offer the following recommendations. Future research should provide a more multidimensional picture of sexual communication between parents and adolescents. Besides adolescents’ perception of mothers’ approval of sexual behaviors, Add Health data also contains some information on parents’ perceptions of the frequency and content of communication with their adolescents about sexual issues. Using these variables, it may be possible to identify mediating and interactional processes that could be helpful in structuring an intervention program. Since parents’ and adolescents’ perceptions differ considerably, developing psychometrically sound measures of adolescents’ perceptions would be important. Parameters should include adolescents’ perceptions of the frequency, content, and timing of such communication, as well as adolescents’ satisfaction with communication from parents about sexual issues. To obtain a comprehensive and multidimensional picture of sexual discourse between parents and adolescents, researchers should also seek more detailed information about discussions of contraception, for example, whether adolescents received instructions about usage and access.
This study suggests that, in order to prevent their adolescents from engaging in sexual risk behaviors, API parents need to be firm in their beliefs as well as active and clear in communicating these beliefs and setting behavioral limits concerning sexual behaviors. Given the consistent and critical role that mothers play in influencing the development of risky sexual behaviors among APIs adolescents, innovative approaches are needed to target API parents as well as adolescents. Due to the bicultural experience of APIs in the U.S., intervention may require a two-pronged approach that delivers sex education in family settings with a combination of both direct and indirect communication. Strategies involving indirect methods of communication (e.g., leaving sex education pamphlets in restrooms) can help parents who may be uncomfortable engaging in open discourse about sexual activities (Lam et al. 2004).
Sex educators need to understand the prevailing levels of avoidance of any conversation related to sex among API families (Okazaki 2002). This study clarifies the parents' responsibility to educate their adolescents about their beliefs on the appropriate time for beginning sexual practices. It also highlights how powerful and simple an intervention can be between mothers and adolescents at home. For adolescents, the perception of their mothers’ attitudes about sexual activities has a continuous effect on their sexual behaviors during their next developmental stage. This fact should motivate API parents to communicate clearly given that their expectation about their adolescents’ sexual behaviors will contribute to protecting their adolescents’ safety and sexual health.
This study was supported by a Dean’s Award from the Boston University School of Social Work. This research uses data from Add Health, designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by grant P01-HD31921 from the National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgement is due to Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. To obtain data files from Add Health, go to http://www.cpc.unc.edu/addhealth/contract.html or contact Add Health, Carolina Population Center, 123 West Franklin Street, Chapel Hill, North Carolina 27516-2524.