Old Ideals and New Realities: The Changing Context of Adolescent Partnerships in Cebu, Philippines

Abstract

The Philippines has experienced rapid sociodemographic changes in recent years, with implications for adolescents. This study combines quantitative and qualitative data from Metro Cebu to assess the timing and predictors of adolescent partnerships, as well as the context in which these partnerships are occurring. The majority of adolescents (54%) had premarital sex, though this pattern varied by gender. Wealthier, urban men, and women with less education and lower reported religiosity, were more likely to have premarital sex. Engagement in risk behaviors was predictive of premarital sex for both males and females. The qualitative data contextualize the circumstances under which adolescents engage in sex and form partnerships, and illustrate how sociocultural norms contribute to gender differences in partnership patterns. Given the ‘new’ realities of young Filipinos’ lives, targeted efforts to support adolescents’ transition to adulthood are needed to avert potentially adverse lifeevents.

Introduction

The global population is comprised of the largest cohort of adolescents ever witnessed. Of the 6.9 billion people in the world, 1.8 billion are currently between the ages of 10 and 24 (United Nations 2011). In addition to immense growth in the adolescent population, globalization and urbanization are profoundly altering the social context in which today’s adolescents live and interact. As a result of the new realities imposed by these rapid demographic and social changes, today’s adolescents face unique challenges in successfully navigating transitions to adulthood. Many of these challenges are particularly pronounced in more resource-constrained settings due to pervasive poverty and income inequalities, inadequate education, and lack of viable, long-term employment opportunities (World Health Organization et al. 2006)and may differentially impact specific domains of adolescent lives’ in which the greatest transition is occurring, such as that within their sexual and reproductive lives (Bearinger et al. 2007).

The Philippines is one such country that has witnessed rapid demographic and social changes. The realities of today’s Filipino youth are substantially different from those of their parents, just a generation ago. The social context has changed in many ways: rapid urbanization, globalization, and the proliferation of mobile and internet technology (Medina 2001; Ogena 1999; 2004). Perhaps the most profound changes among Filipino youth, however, have been those associated with dating, sex, and marriage.

Using an ecological approach, this study combines quantitative and qualitative data to assess the factors associated with sexual and partnership patterns and to better understand the social context of adolescents living in a metropolitan area of the Philippines.

Study Setting

The Philippines is an archipelago of over seven thousand islands located between the Philippine Sea and the South China Sea, east of Vietnam. A predominantly Roman Catholic country of approximately 88 million people, the Philippines ranks 112^thof 187 countries in the Human Development Index compiled by the United Nations Development Program (UNDP)(National Statistics Office: Philippines 2007a; United Nations Development Program 2011).

The setting for this study is Metro Cebu, located on the island of Cebu in the Central Visayas region of the Philippines. Cebu has experienced rapid socioeconomic changes, as well as substantial population growth in the past several decades. Metro Cebu is the second largest metropolitan area in the Philippines, with a population of 1.4 million in 2007. The 2007 census estimated that there were 17.1 million 15–24 year olds in the Philippines, just over 19% of the population (National Statistics Office: Philippines 2007b).

Partnership Patterns and Sexual Behavior among Filipino Adolescents

Similar to other global settings, Filipino adolescents are more likely to delay marriage, to choose cohabitation over formal marriage, and to engage in premarital sex as compared to their parent’s generation. The mean age at marriage for men is 26.5 years (up from 24.8 in 1980) and is 23.8 for women (up from 22.4 in 1980)(National Statistics Office: Philippines 2009). These shifts in age at marriage have been accompanied by an increase in premarital cohabitation or cohabitation without marriage (Kabamalan 2004; Xenos and Kabamalan 2007).

Premarital sex is also more common. National survey data from 2002 indicate that 31% of males and 16% of females between the ages of 15–24 reported having premarital sex, an increase from 1994 levels(Natividad and Marquez 2004). A more recent analysis from Metro Cebu found that 67% of males and 47% of females had sex before age21, of whom 98% of males and 91% of females had sex before marriage (Upadhyay and Hindin 2007). Of those who had premarital sex, only 15% of females and 28% of males reported that they used contraception during their first premarital sexual encounter, and nearly 40% who reported using contraception used withdrawal (Natividad and Marquez 2004).

Contextual Influences on Filipino Adolescents’ Partnership Patterns and Sexual Behavior

Despite the rise in cohabitation and premarital sex, most Filipino youth are not in favor of these practices for men or for women(Kabamalan 2004; Ventura and Cabigon 2004). Only 31 % of males and 15% of females approve of women having premarital sex while 46% of males and 24% of females approve of men having premarital sex (Ventura and Cabigon 2004). Pregnancy outside of unions is not condoned in the Philippines -only 18% of young men and 12% of young women approve of a woman getting pregnant if she is not married to the father (Ventura and Cabigon 2004).

Stark differences in the occurrence and acceptance of these behaviors for young men versus young women attest to the persistence of traditional gender norms in the areas of sexual behaviors and partnership patterns in the Philippines. Young men have greater freedom to express their sexuality and are more likely to engage in dating, precoital, and sexual behaviors earlier than young women (Upadhyay et al. 2006; Ventura and Cabigon 2004). Social norms for young women are more conservative and dictate that women are expected to be modest and chaste, and should refrain from expressing interest in or knowledge about sex or contraception, especially before marriage(Medina 2001).

The persistence of traditional norms surrounding sexuality and disapproval of premarital sex, particularly for young women, are likely influenced by the strong presence of the Catholic Church in the Philippines. Catholicism is the dominant religion in the Philippines, representing 80% of the national population (National Statistics Office: Philippines and ICF Macro 2009). The Catholic Church is extremely active in the Philippines in promoting pronatalist teachings and beliefs of the Church, as well as in influencing reproductive health policy at the local and national levels(Linanganng Kababaihan et al.2007; Mello et al. 2006; Ruiz Austria 2004). Currently, the Church is actively engaged in a national campaign to prevent the passage of the Reproductive Health and Population Development bill, a bill which would ensure governmental support for and universal access to contraceptive methods(Senate Economic Planning Office 2009).

Urbanization and industrialization are often discussed as major factors affecting adolescents’ sexual and partnership behavior, specifically as they affect Filipino family structures and individuals’ roles within the family. Medina (2001) notes that these forces are likely to have mixed influences on Filipino adolescents’ partnership and sexual behavior. For example, a shift to more permissive norms and less reliance on traditional norms in which sexuality is regarded as a “shameful and private” topic may result in greater discussion and planning of sex and childbearing, yet rapid social changes and contrasting norms between older and younger generations may create conflicts in the family and may lead to “individualized or deviant behavior”.

Limited studies exist from the Philippines that assess the influence of peers and family context on partnership and sexual behaviors. A recent study from Cebu indicates that, similar to other settings, perceptions of peers’ engagement in sexual behaviors were predictive of adolescent respondents’ sexual activity (Upadhyay and Hindin 2006). Another study by Upadhyay and Hindin (2007) found that parents’ joint household decision-making was associated with a lower likelihood of having sex by age 22 among their adolescent sons, but no effect was found for daughters (Upadhyay and Hindin 2007). Daughters’ sexual debut was influenced, however, by other measures of the mother’s status within the household and whether the father physically beat the mother (Upadhyay and Hindin 2006; 2007). In one study that assessed parent-adolescent communication, mothers’ and teens’ reports of communication about sex were not associated with sexual debut.

Numerous studies point to the influence of individual-level sociodemographic and attitudinal factors on sexual and partnership behaviors among Filipino adolescents. In the Philippines, although 86% of the population is functionally literate, current school enrollment among 12–15 year olds in 2003 was 87%, reflecting the reality that adolescents, particularly males from low-income families, are compelled to work to contribute to the household income(Commission on Population Republic of the Philippines; Philippines National Statistics Office 2003). Recent analyses from Cebu found mixed effects of education on sexual debut (Upadhyay and Hindin 2006; 2007). One study found that educational attainment was associated with a lower likelihood of having sex by age 19 (Upadhyay and Hindin 2006); however, two subsequent studies found that educational attainment was associated with delayed sexual initiation for female participants only and had the opposite or null effect for young men (Upadhyay and Hindin 2007).

Studies from both developed and developing country contexts indicate that adolescents who engage in risk behaviors (e.g., smoking cigarettes, drinking alcohol, and taking drugs) are more likely to initiate sex at an earlier age and to be involved in risky sexual behavior. These associations were also found for Filipino adolescents, as measured in the 1994 and 2002 Young Adult Fertility and Sexuality Studies (YAFS). Change across the study periods indicated that there was an overall increase in smoking, alcohol, and drug use among young men and women, and that these behaviors were also associated with premarital sex (Commission on Population Republic of the Philippines).

Using an ecological framework approach, this study incorporates quantitative and qualitative data to explore the social context of adolescents in Cebu that may increase their vulnerability to poor sexual and reproductive health outcomes. Longitudinal, quantitative life event data from the Cebu Longitudinal Health and Nutrition Surveys (CLHNS) were analyzed to describe the timing and order of partnerships and sexual initiation among young men and women. Qualitative data further contextualize these life events, providing insight in to how and why partnership patterns appear to be changing and highlighting the unique challenges that adolescents face in navigating sexual and reproductive health decision in a dynamic, social environment.

Methods

Quantitative data and analyses

Quantitative data are derived from the Cebu Longitudinal Health and Nutrition Surveys (CLHNS). The CLHNS is an ongoing longitudinal study of a cohort of mothers and their index children (ICs) born in 1983–84. The CLHNS beganin 33 barangays(communities) from the Metro Cebu area. To date, cohort members participate in periodic surveys on health issues and basic demographic information, with the most recent full surveys fielded in 2002 and 2005. Further information on the cohort and data collection methods are detailed elsewhere.

The analyses for this paper focus on the data from the 2002 and 2005 CLHNS surveys. Sociodemographic data from the 2002 survey were used to predict life events reported in the 2005 survey. The 2005 survey included a life history matrix to collect information on the occurrence and timing of adolescent life events , including sexual activity, contraceptive use, and union status. In total, 2,051 and 1,935 adolescents participated in the 2002 and 2005 surveys, respectively (94% retention).

The dependent variable in this analysis is the occurrence and pattern of adolescents’ partnership and sexual activities. Using data from the 2005 life history matrix, we determined if each adolescent had initiated sex by the time of the survey (when they were approximately 21 years old), and if sex had occurred prior to or after entering into a formal partnership. Given increased rates of cohabitation and non-formal marriage in the Philippines and in this sample (only 35% of partnered adolescents were legally married in 2005), in this analysis a formal partnership is defined as either cohabitation, civil union, or legal marriage. The outcome variable is comprised of three, mutually exclusive categories—no sexual activity or partnership by 2005 (subsequently referred to as abstinence), partnership before sexual intercourse, or partnership following sexual intercourse.

Independent predictor variables from the individual and household levels were selected from the 2002 CLHNS survey. Individual level characteristics and behaviors included: religiosity, number of years of school completed, and self-report of risk behaviors: ever using drugs, alcohol, or smoking (summative score). Age was omitted from the multivariate analyses since all of the adolescents were enrolled in the study upon birth in 1983–84. Religion was also omitted from the multivariate analyses due to lack of variability (95% of participants are Catholic) and small cell sizes within subgroups (e.g., non-Catholic males in ‘partnership before sex’ group). Household predictor variables included household residence (urban or rural) and a measure of household wealth (weighted average of household assets, including items such as a television, bicycle, and air conditioner). Bivariate analyses and multinomial multivariate analyses were conducted. All analyses were disaggregated by gender given pervasive gender norms and disparate patterns of sexual activity and partnership patterns among young men and women (Upadhyay et al. 2006).

Qualitative data and analyses

In 2007–08, qualitative data were collected to complement the quantitative data from the CLHNS surveys and to more holistically explore the social context in which adolescent partnerships and sexual encounters occur. The design of the qualitative data collection was informed by Ecological Systems Theory, in which individuals represent the core multiple systems, comprised of their immediate environment or microsystem (e.g., family and peers), intermediate levels (e.g., household and community) and the macrosystem, the larger cultural context in which individuals exist. Focus group discussions(FGDs) and in-depth interviews (IDIs) were used to explore the multiple systems that influence the lives of the adolescents in our study. Focus group discussions gathered information about the norms and practices that exist in the community and amongst the peers of our FGD participants while the in-depth interviews focused more on individuals and their proximate relationships with partners and family members.

A total of nine FGDs were conducted by trained moderators with a convenience sample of adolescents ages 21–29. All of the FGD participants were from CLHNS urban study barangays (communities) but were not members of the CLHNS cohort. In-depth interview participants were purposively sampled based on their report in the 2005 CLHNS survey of a negative reaction to a past pregnancy and their residence in an urban barangay. Each participant was interviewed separately and confidentially on two, separate occasions by a trained, Filipino interviewer in the local language, Cebuano. A total of 66 interviews were completed with 20 CLHNS adolescents and 13 of their partners (18 females; 15 males). Each interview lasted approximately 60–90 minutes and was audio-recorded upon obtaining the participant’s permission.

All of the FGD and IDI recordings were transcribed and translated in to English by the interviewers in the field. The transcripts were input in to word processing software and subsequently entered in to QSR NVivo 8.0 for analysis. First, a coding scheme was developed and refined by the research team according to main themes delineated in the study instruments and themes that emerged spontaneously during data collection. Second, using a constant comparative approach, the coding scheme was used to identify phrases and concepts that reappeared within the data and to combine these pieces of text from the IDIs and FGDs into larger, overarching domains (Glaser 1965). All of the research protocols and instruments were approved by the University of San Carlos Office of Population Studies and the ethical review boards at Johns Hopkins Bloomberg School of Public Health and the University of California, Los Angeles.

Results

Quantitative results

Fifty-four percent of females and 36% of males were neither sexually active nor in a partnership by age 21(Table 1). For males, the most common pattern was sexual intercourse before a partnership, while for females it was abstinence followed by sexual intercourse before a partnership. Among those adolescents who were sexually active by age 21, only 3% of females and less than 1% of males reported a partnership before sexual intercourse (Chi-squared test: p<0 .001).

Table 1.

Life Events by Age 21 Reported by CLHNS Participants

	Men (n = 1027)	Women (n = 908)	Total (men and women) (n = 1935)
Abstinent	36%	54%	44%
Partnership before sex	1%	3%	2%
Sex before partnership	64%	44%	54%

Table 2 displays sociodemographic characteristics and risk behaviors, measured in 2002, disaggregated by gender and life event patterns reported in the 2005 survey. For both males and females, adolescents in the abstinent group had lower reported levels of risk behaviors as compared to the sex before partnership group. Males who had sex before a partnership were more likely to live in urban areas and to have higher household wealth, as compared to the abstinent group. Females in the abstinent group had higher education and household wealth, as compared to the other two groups. Overall, young women are more educated, more likely to consider themselves religious, and less likely to engage in risk behaviors as compared to young men.

Table 2.

Sociodemographic Characteristics and Risk Behaviors (2002) by Life Event Patterns (2005) and Gender

	Abstinenta (n = 853)	Partnership before sex (n = 32)	Sex before Partnership (n = 1050)	Total (n = 1935)
Males (n = 1027)
Age	18.1	18.1	18.2†	18.2
Years of school completed	9.7	6.8***	9.7	9.7
Household assets (mean sum score)	4.7	3.8	5.2**	5.1
Urban	67%	63%	81%***	76%
Catholic religion	91%	88%	94%	93%
Consider themselves religious	48%	50%	48%	48%
Ever tried smoking	59%	100%*	79%***	72%
Ever tried alcohol	80%	88%	92%***	87%
Ever tried drugs	10%	38%	31%***	24%
Total risk behaviors (sum)	1.5	2.3*	2.0***	1.8
Females (n = 908)
Age	18.2	18.3	18.2	18.2
Years of school completed	11.5	8.3***	10.2***	10.9
Household assets (mean sum score)	5.4	3.6***	4.5***	5.0
Urban	75%	54%†	74%	74%
Catholic religion	94%	100%	96%	95%
Consider themselves religious	60%	67%	50%**	56%
Ever tried smoking	17%	33%	36%***	26%
Ever tried alcohol	65%	67%	77%***	70%
Ever tried drugs	1%	0%	7%***	3%
Total risk behaviors (sum)	0.8	1.0	1.2***	1.0

^aReference group in Bonferroni tests of differences between groups.

^***p ≤ 0.001

^**p ≤ 0.01

^*p ≤ 0.05

^†p ≤ 0.10

The differences in sociodemographic and risk behaviors between males and females illustrated in Table 2 prompted the development of the gender-disaggregated multivariate multinomial logistic regressions depicted in Table 3.¹ In these analyses, the abstinent group is the comparison category. For each additional year of schooling, women were 40% less likely to be in a partnership before sex as compared with women who neither had sex nor a partnership (RRR=0.60). In comparing adolescents who had sex before a partnership with those in the abstinent group, males were wealthier and more likely to live in urban areas, whereas females were less likely to be religious, and had less education and household wealth. Risk behaviors were associated with having sex before a partnership for both males and females (RRR=2.23 males; 1.93 females, for each risk behavior), as compared to adolescents in the abstinent group.

Table 3.

Multivariate Relative Risk Ratios of Adolescent Life Event Patterns in 2005, By Gender

	Males (n = 1027)	Females (n = 908)
Partnership before sex
Number of years of school	0.86	0.60***
Household assets	0.93	0.99
Urban residence	1.09	0.50
Religious	1.03	1.05
Risk behaviors	2.56†	1.57
Sex before partnership
Number of years of school	1.02	0.74***
Household assets	1.09**	0.94†
Urban residence	2.16***	0.97
Religious	1.09	0.67**
Risk behaviors	2.23***	1.93***

Reference group: Adolescents who did not have a partnership or sex by 2005.

^***p ≤ 0.001

^**p ≤ 0.01

^*p ≤ 0.05

^†p ≤ 0.10

Qualitative results

Contextual Influences on Partnerships and Sex

Mirroring trends from national data on Filipino adolescents’ sexual behaviors, both male and female study participants said they felt there is greater temptation and peer pressure now to initiate sex earlier and before a partnership.

Most of the participants reported meeting their boy/girlfriends through mutual friends or neighbors, or while out with their group of friends (barkada). In the FGDs, discos were cited as one of the examples of the changing norms in youth today, and were associated with ‘early marriage’, uyab-uyab (boy-girl relationships), and creating an environment where young people could hang out, unsupervised, late at night. Although FGD participants were, overall, more disparaging in their comments regarding discos, they were frequently mentioned in the IDIs as places in which the participants met past or current partners.

Participants indicated that adolescents are curious about sex and engage in premarital sex , often within weeks of knowing one another, due to lust (biga) or being hot (kumag). Motivated by these curiosities, adolescents are having sex, often without consideration of the possibility of pregnancy, sexually transmitted infection, or long-term implications for their lives. According to one participant, “They [adolescents] only think about sexual desires and not the consequences or their future.” Several IDI participants mentioned the short-term use of motels to facilitate sexual encounters and were often combined with drinking alcohol. Participants indicated that these hotels, which can be rented on a per-hour basis, offer a private, discreet place that they could go with their partner or group of friends without their parents knowing.

One of our study participants became pregnant after visiting one of these motels with her boyfriend of one month and a group of friends:

R: We saw a movie and after that we had a drinking spree with our barkada. I can’t go home at dawn because my mother would get angry. It was about 3 AM. We checked-in [to a motel]. There were 3 pairs of us who checked-in. We were all drunk… It all happened there. Really, we made our child only once, only that time.

(Female IDI)

Gender, partnerships, and sexual behavior

Similar to the quantitative results, the qualitative data indicate substantial differences in the social norms and sexual behavior for young men and young women. Young men are perceived to be more interested in sex and more aggressive (barako) than young women. This notion was raised more frequently in the female FGDs than in the male FGDs; however, young men often mentioned sexual desire and curiosity as reasons to engage in premarital sex. Several male IDI participants indicated they had girlfriends as young teenagers (e.g., ages 13–14) and some had sexual encounters with older women or “sugar mommies”.

As voiced by many of our study participants and as described by this 23-year old female IDI participant, sex was often initiated suddenly and without prior discussion:

P: [We did not use a method], because honestly we did not intend to do it. It was sudden. We did not plan to do it because it was all sudden.

I: What do you mean that you did not plan to have sex with him?

P: That’s it. It all happened suddenly, maybe because it was just us at that time. You know how teenagers get hot sometimes. We were in his room. We did not plan it. We did not talk about sex. It came very sudden. You just begin to feel that you want to do it.

Rape and forced sex were mentioned in several FGDs and IDIs. In the FGDs, forced sex was described as a reason for an unintended pregnancy (disgrasya), or as something that might occur between a husband and wife, particularly after the husband has been drinking. In the IDIs, however, two young men and three young women described forced sexual encounters. The following story by a female IDI participant illustrates the potential for blame and the fragility of women’s (and children’s) reputations given persistent and strong social sanctions of premarital sex and out-of-wedlock childbearing:

P: I told him that he was so strong, while I was so skinny. Then he would force me. And I didn’t want to…I was afraid that I might get pregnant. He told me “No, you won’t be.” And so I had sex with him. It sounds like rape because I was forced, right?

I: How did he force you?

P: He undressed me. I was angry. I cried, “Of all my boyfriends you are the only one who forced me! What if something happens to me?” By December I was pregnant. I was so mad at him because he would not own my pregnancy. He accused me of having a relationship with one of his friends. He also told his mother about it and his mother thought that I was not a proper woman. How could his mother think that? After I gave birth his grandmother called my child a” totoy angkon”(somebody who is not owned).

Similar to the results from the quantitative analyses, the qualitative data indicated that the likelihood of engaging in partnerships and sexual behavior were quite different for young women than for young men. In contrast to young men, young women seemed more concerned about maintaining their virginity until marriage or once in a stable partnership. For women, it was important to trust that a partner would not abandon them if they became pregnant. Study participants indicated that in the case of pregnancy, the consequences are more severe for a woman and her family. A young woman will often stop her studies and cease working once she discovers she is pregnant, when the pregnancy becomes evident, or when she is physically unable.

The process involved in deciding whether to cohabit or to marry also differed by gender. Young men were more concerned with employment, the increasing cost of living, and the desire to support a family. Men also mentioned the shame that many men feel when they are forced to live with their parents or, worse yet, with their in-laws, due to financial constraints. Recurring themes of responsibility, fulfilling an obligation, and pity for the woman and child emerged as reasons that men were prompted to make a long-term commitment to a woman, especially in the case that she was pregnant. For women, the decision to get married was dependent on her parents, but was even more dependent on her partner. Given that most partnerships were precipitated by a pregnancy, women and their families were portrayed as essentially powerless, left to the whims of the male partner to decide whether or not to accept responsibility for the woman and the pregnancy. Women also indicated that ‘forcing’ a man to take responsibility for a pregnancy could cause problems later in the relationship (e.g., infidelity or violence) if he really did not choose for himself.

I: And what was the reaction of your parents when your plans to marry were not realized?

P: Nothing. It seemed that my parents couldn’t demand anything more. It’s really different when the woman is not yet pregnant because the woman can still make demands. At that time my parents just went along with everything because I was already pregnant. My parents really wanted us to get married right away because it is far different than when you are just living together.

(Female IDI)

P: I do not think it is possible that she would not want to marry me especially that she is pregnant. For sure, she would marry me. Don’t tell me she would not want to marry me, what can she do especially since she is pregnant?

I: What do you mean by “what can she do”?

P: What would she do if she is pregnant and nobody wants to stand up for her to become the father of the child? She doesn’t have the choice. How could she take care of herself, especially if she is pregnant?

(Male IDI)

Discussion

Amidst rapid global socioeconomic changes, there is growing recognition that adolescent reproductive health research and interventions need to consider not only individual-level and developmental characteristics of the adolescent, but also the broader, social systems in which adolescents exist and interact (Bayer et al. 2010; Bearinger et al. 2007; Fatusi and Hindin 2010; Mmari and Blum 2009). As noted by Bearinger, et al., “Questions about why and how to invest in turning the tide [of negative SRH outcomes for young people] can only be answered with an understanding of the uniqueness of this age group and the social contexts that increase adolescents’ vulnerability to poor sexual health outcomes (Bearingeret al. 2007, 1220).”Results from this mixed -method study highlight the respective influences and interaction of both contextual and individual-level factors on Filipino adolescents as they transition in to adult partnerships and sexual relationships.

A key finding from this study is that there is significant discordance between the predominant norms regarding adolescent lives and the reality. Despite persistent disapproval of premarital sex, even among younger generations, the majority of respondents (64% of males and 44% of females) in our sample reported that they had sex before entering in to a formal partnership, either cohabitation or marriage.

Moreover, the gender-disaggregated findings revealed stark differences in young men’s and young women’s partnership behaviors. Young men were more likely to engage in sex before partnerships, as compared to young women in this study. This finding is similar to that of an earlier study in this area (Upadhyay and Hindin 2006)and reflects the persistence of conservative social norms for young women as compared to young men, especially with respect to sexual behavior. The qualitative data indicated that young women were more concerned about protecting their reputation as a ‘proper’ or chaste woman by preservation of virginity and reluctance to exhibit knowledge of or to discuss contraception with a partner. Young women may also be differentially impacted by transgressions from traditional norms, either through the expectation that women should forego educational and occupational opportunities to take care of a child, or through the dependence of unmarried women on their partners or their parents to provide support for her and the child. Young men, however, also mentioned specific pressures – the need to find a stable job and to be able to provide for a family before entering in a formal relationship.

The predictors of sexual behaviors and partnership patterns also differed for young men and women. Similar to another study from the Philippines by Williams, et al., religiosity and educational attainment appeared to influence women’s life event patterns, yet this was not the case for young men(Williams et al. 2007). This is in contrast to other findings from Cebu, however, indicating no influence of religiosity for either young men or women and increased risk of sexual debut among young men with higher education (Hindin 2006). Although nearly all of the young men and young women identified as Catholic in the 2005 survey (95–96%), a greater proportion of young women consider themselves religious (Table 2) and attend church with greater frequency than young men (50% versus 33% attend church at least once per week; data not shown). It is likely that the intersection of gender norms and Church teachings promoting traditional images of women as ‘chaste’ and ‘pure’ may influence both the likelihood that women attend church, as well as the exposure to messages influencing partnership patterns and sexual behaviors among this subset of young women. Although no effect of education was found for young men, the protective effect of education on young women’s sexual initiation mirrors findings from other global settings (National Research Council and Institute of Medicine of the National Academies 2005).

As compared to other young men, urban men and men from wealthier households were more likely to engage in sex before a partnership. This effect remained after controlling for education, indicating that there may be factors associated with living in these environments that may make sex more attainable for these young men. For both young men and young women, there were strong effects of risk behaviors on partnership patterns. For young men, urban residence and engaging in risk behaviors had a synergistic, interactive effect, indicating that these two factors placed young men at an even greater risk of sex before partnership. In the two previous studies from Cebu, similar effects of urban residence were found for young males (Upadhyay and Hindin 2006; 2007). These findings point to the roles that societal expectations and gender norms may have on the upbringing and socialization of young girls and boys (Williams et al. 2007), as well as the role of context and situational factors that may make some adolescents more likely to adopt these non-traditional (and perhaps higher risk) patterns. Data from the developing world is quite limited in this arena; however, evidence suggests that the prevalence of risk behaviors such as tobacco use, alcohol and substance abuse, is increasing due to the pressures of increased urbanization, poverty, and reduced social support (Uchtenhagen 2004).

In the Philippines, Medina (2001) has described the added strains and stresses that urban adolescents may be exposed to due to greater social structural complexity and faster rates of social change within the Filipino family (e.g., changing role structure of the family, decline in authority of the husband and father, decline of the family’s influence on the individual, more permissive norms and behavior, breakdown of the consanguineal family as a functional unit) (Medina 2001). As described in our qualitative findings and studies from this and other settings, dating behaviors of adolescents differ greatly from that of previous generations (Gastardo-Conaco et al.2003; Medina 2001; Remes et al. 2010). Formal courtship and the emphasis on meeting one another’s parents have shifted to more informal and casual encounters and through outings with friends. Although these practices are not risky in and of themselves, parents’ lack of knowledge or acceptance of these changing norms may create a situation in which adolescents surreptitiously frequent these locales to fit in with peers or may rebel against parents who are perceived to be too strict. Moreover, alcohol was mentioned in several contexts, among both married and unmarried adolescents, and often precipitated ‘sudden’ and unprotected sexual encounters, exposing adolescents to unintended pregnancy and sexually transmitted infections.

Lastly, the collection of in-depth data provided insight in to two, important and neglected areas of adolescent sexual and reproductive health research from the Philippines and other global settings: the occurrence of forced sex and the role of sexual desire. Forced sex was mentioned within several of the FGDs and IDIs; the occurrence of forced sex is corroborated by quantitative data from the CLHNS. Among adolescents who had sex before the 2005 survey, the majority (52%) said that their first sexual experience was something they either ‘did not want to happen’ (12%) or ‘did not plan to happen’ (40%); another 3% reported that their first sexual experience happened ‘against their will’. Similarly large proportions of coerced first sex are reported in national data from the Philippines, with the highest rates among younger females (Natividad and Marquez 2004). Globally, it is estimated that up to one-third of adolescent girls experience a forced first sexual experience (Krug et al. 2002). Given the large proportion of the global adolescent population that is exposed to forced sex and other forms of sexual violence, as well as the numerous, negative psychological and physical health consequences associated with its occurrence, it is of utmost importance to assess and address this important public health issue (Jejeebhoy and Bott 2005; Krug et al. 2002).

A second area of neglect within global adolescent sexual and reproductive health research and programs is the need to better understand sexual pleasure and desire and its influences on sexual and reproductive health outcomes such as engagement in unprotected sex, unintended pregnancy, and sexually transmitted infections (Dixon-Mueller 1993; Higgins and Hirsch 2007; Philpott et al. 2006). There is increasing acknowledgement of the importance of addressing this issue within adolescent populations, especially given well-documented restrictions on comprehensive sexuality education and reproductive health services for adolescents and young adults (Ott et al. 2006; Singh et al. 2006). In the Philippines, limitations in the provision of sexual and reproductive health education and services are well-documented (IRIN Humanitarian News and Analysis 2010). These restrictions may be placing adolescents in a particularly precarious position as they transition in to sexual partnerships and union formation.

Limitations

Two caveats should be mentioned in the interpretation of and inferences drawn from this analysis. First, due to small sample size in the male ‘partnership before sex’ group, there may have been insufficient power to detect significant effects of the covariates (i.e., apart from the marginal effect found for risk behaviors). Second, the quantitative data were derived from the entire youth sample of the CLHNS cohort and the focus group data were drawn from members of several CLHNS sample communities. The in-depth interview data, however, were collected from a sub sample of participants who reported a negative reaction to a past pregnancy. While the inferences and overall themes discussed in this paper were represented across all data collection methods, some of the specific experiences of interview participants may not be representative of the entire CLHNS cohort, nor of the entire Filipino young adult population.

Conclusions

Overall, this study incorporates unique, mixed-method data to better contextualize partnership decision-making in Metro Cebu, Philippines. Findings from the quantitative and qualitative data highlight areas for further research and programs, specifically the need to identify and address the generational and societal changes that may be putting Filipino adolescents at risk for potentially adverse life events.