Introduction

Defining sexual coercion is challenging because women’s perceptions of sexuality are heavily influenced by social and cultural expectations within their respective social environments (Population Council 2004). Despite this challenge, past research has defined sexual coercion as any forceful sexual behaviour that involves violence, threat, verbal insistence, and/or deception (Heise et al. 1999). It is important to note that this can include intercourse that is not consensual or is consensual but unwanted (Moore et al. 2007a).

Forced sexual debut is increasing globally, particularly among women (Jejeebhoy et al. 2005; Population Council 2004). Geary et al. (2008) noted, for instance, that first sexual intercourse through physical force, verbal pressure, or as an obligation for marriage was not uncommon among young women in Uganda. Similarly, 20–30 % of females in South Africa mentioned that they were verbally persuaded and physically forced at their first sexual experience (Buga et al. 1996; Jewkes et al. 2001). These findings are not very different from the Philippines where nearly 30 % of young women described their first sexual encounter as consensual but unwanted; 2 % reported their sexual debut as physically and verbally forced, commonly by their intimate partners (Ramiro 2005; Raymundo and Cruz 2004). While intimate partners have been identified as the most common perpetrators of sexual violence, Raymundo and Cruz (2004) also found that unwanted sex in the Philippines was predominantly perpetrated by strangers and immediate family members. The foregoing evidence sharply contrasts the Philippines’ Anti-Rape Law passed in 1997 that prohibits any sexual engagement through force, threat, intimidation, humiliation, and sex trafficking (Ramiro 2005).

Forced first sexual intercourse has deleterious consequences and can lead to serious complications for victims (Caceres et al. 2000; Khawaja and Hammoury 2008). For instance, in the Central African Republic, victims of forced first sexual initiation were significantly less likely to use condoms in subsequent sexual experiences (Somse et al. 1993). Furthermore, Molla et al. (2002) found in Ethiopia that female victims of forced first sexual initiation were more likely to experience forced sexual intercourse in the life course. Experience of forced first sexual intercourse also resulted in decreased sexual desire, fear of intimacy, lack of sexual pleasure, depression, feelings of worthlessness, and thoughts of suicide (Heise et al. 1999; Khawaja and Hammoury 2008; Koenig et al. 2004). In addition, women who experienced forced first sexual intercourse were more likely to be exposed to sexually transmitted diseases, including genital tract infections (Jejeebhoy 2011).

Meanwhile, the literature points to both structural- and individual-level factors contributing to women’s exposure to forced sex (Adhikari and Tamang 2010; Tenkorang and Owusu 2013). At the structural level, ethnocultural norms are considered influential to sexual violence. This may be due to the fact that some ethnocultural norms can be linked to the expression of masculinity, which is considered a root cause of men’s sexual coerciveness (Price 2009; Vandello and Cohen 2003). For example, Nwabunike and Tenkorang (2015) found significant ethnic variations in the incidence of sexual violence in Nigeria, and attributed this to differences in ethnocultural norms that create an enabling environment for sexual violence against women.

In the Filipino context, expression of masculine ideals is highly prized among some ethnic groups. For example, masculinity is considered traditionally expressed through sexual dominance over women in the Cebuano culture (Aguilar 2008), while interpersonal relations in traditional Ilonggo society often involve an emphasis on male power in many aspects of life (Camposano 2014). Besides ethnicity, religion is an important structural factor that influences sexual violence. Major religious groups in the Philippines include the Roman Catholic Church, Protestantism, and Islam. Religious norms can lead to unequal gender relations, given that they emphasize patriarchal norms that often provide an acceptable basis for women’s passivity and unchallenged availability to male partners (Minkah-Premo 2001; Heise et al. 1999). Contrarily, religious teachings can provide unacceptable grounds for engaging in early sexual intercourse whether or not it is consensual (Ellison et al. 1999).

As is the case in other less developed societies, socioeconomic disadvantage combines intricately with cultural norms and gender inequality to render women vulnerable to the risks of sexual violence in the Philippines. Aguilar (2008) noted that societal entrenchment of gender inequality in the Philippines was partially created through the expectation that men fulfill their educational needs to be future breadwinners for their families. It is widely acknowledged that such cultural norms not only widen the socioeconomic gaps between men and women, but also increase women’s dependence on men, and provide platforms for sexual dominance and coercion on the part of men (Jewkes 2002). It is thus not surprising that previous studies, using women’s educational backgrounds and wealth status as proxies for socioeconomic status (SES), found significant differences in the sexual experiences of women with higher SES compared to those with low SES (Koenig et al. 2004; Moore et al. 2007b; Santhya et al. 2007).

Demographic and life course factors, including union formation and age at first sexual intercourse, have been documented as important correlates of forced sexual intercourse (Erulkar 2004; Peterman et al. 2011). In the Philippines, it is not uncommon for married women to underreport sexual violence, as reporting such cases can have far-reaching implications, including possible marital dissolution (US Department of State 2007). In other jurisdictions, research indicated that sexual violence was not only common among never-married women, but also among the married (Glover et al. 2003; Khawaja and Hammoury 2008; Moore et al. 2007a). With regard to age, younger women are often more likely to experience forced first sexual intercourse through threats and deception than their older counterparts (Population Council 2004). Similarly, early sexual debut is often forced, given that the self-efficacy and decision-making skills of young people are usually not fully developed at this stage of the life course, and that youth may also lack the skills to negotiate for safer sex (Koenig et al. 2004; Peterman et al. 2011). In summary, the literature shows that women may experience forced first sexual intercourse based on their individual-level socioeconomic characteristics, normative beliefs, and characteristics at different stages of the life course. Based on the extant literature, we examine the effects of socioeconomic, cultural and life course factors on forced sexual intercourse among Filipino women, especially as previous work has remained largely descriptive and failed to explore the complex nature of coercion at sexual debut.

Data and Methods

Data for this study were drawn from the 2008 Philippines Demographic and Health Survey (PDHS). The PDHS is a nationally representative dataset of Filipino women aged 15–49 conducted by the National Statistics Office (NSO) of the Philippines between August and September 2008, under the auspices of the Demographic and Health Surveys (DHS) program (National Statistics Office and ICF Macro 2009). The PDHS provides high-quality and reliable information on basic demographic indices and sexual violence, including the circumstances surrounding first sexual experience. The 2008 PDHS is the most recent data with information on forced first sexual intercourse. The PDHS employed a three-stage sampling framework in which systematic sampling with probability proportional to size was applied. Face-to-face interviews were conducted with 13,594 women, with a response rate of 98 %. The question on forced first sexual intercourse was asked only as a part of the domestic violence module. In this module, one eligible woman was randomly selected from each household (N = 9316). Finally, we restricted our sample to those who have already experienced first sexual intercourse (N = 7293).

Dependent Variable

The dependent variable used here was dichotomous and asked women of their experience at sexual debut. Specifically, respondents were asked, “The first time you had sexual intercourse, would you say you had it because you wanted, or because you were forced to have it against your will?” Response categories for this variable were “wanted” and “forced” (0 = wanted; 1 = forced).

Independent Variables

Three sets of independent variables (cultural, demographic and life course, and socioeconomic factors) were included in the analysis. Cultural variables included ethnicity (0 = Tagalog; 1 = Cebuano; 2 = Ilocano; 3 = Ilonggo; 4 = other) and religious affiliation (0 = Roman Catholic; 1 = Protestant; 2 = other Christian; 3 = Muslim; 4 = no religion). Three demographic and one life course variable were further included. These were age at first sexual intercourse (0 = 15 years or below; 1 = 16 years or above; 2 = at union), age of respondents (measured in completed years), marital status (0 = never married; 1 = currently married; 2 = formerly married), and urban–rural residence (0 = urban; 1 = rural). For socioeconomic predictors, we included household wealth quintiles (0 = poorest; 1 = poorer; 2 = middle; 3 = richer; 4 = richest) and education (0 = higher education; 1 = secondary education; 3 = primary education; 4 = no education). It is noteworthy that household wealth quintiles were constructed from a composite index based on the household ownership of consumer items, such as drinking water, car, and toilet facilities, among others.

Statistical Analysis

While the outcome variable was dichotomous, the cases were unequally distributed as evidenced in Table 1. Thus, using a simple logit link function which assumes symmetry could potentially produce biased parameter estimates (Tenkorang and Owusu 2010). To address this concern, we used the negative log–log model, which is considered more suitable when the lower category of the dependent variable is more probable. Results were reported in odds ratios. Odds ratios larger than one indicate higher odds of reporting forced first sexual intercourse, whereas those less than one indicate lower odds of reporting forced first sexual intercourse. Also, due to the multistage sample design, complex population-based surveys such as the PDHS usually have some degree of dependence among the observations. This is problematic, especially as standard regression models such as ours often assume independence among subjects. We addressed this problem by imposing a “cluster” variable (ID numbers) on our models. This potentially adjusts the standard errors, producing statistically robust parameter estimates (Bryk and Raudenbush 2002).

Table 1 Univariate analysis of selected dependent and independent variables
Full size table

Results

Univariate Analysis

Table 1 shows univariate results for selected dependent and independent variables used for the analysis. The majority of respondents (95.4 %) described their sexual debut as wanted, although 4.63 % (N = 338) described theirs as forced. Tagalog (22.8 %) and Cebuano (24.2 %) were the two largest ethnic groups in the sample, followed by Ilocano (10 %) and Ilonggo (9.31 %). In addition, more than three-quarters of respondents (76.9 %) reported that they belonged to the Roman Catholic Church, whereas 4.7 and 6.4 % identified themselves as Protestants and Muslims, respectively. A small portion (0.19 %) of women reported that they did not belong to any religion. The average age of respondents was 33.6 years old. For age at first sexual intercourse, 77.2 % of respondents indicated experiencing their sexual debut at the age of 16 or later, while 6.53 and 16.3 % reported sexual debut at the age of 15 or earlier and within marriage, respectively. It is noteworthy that a majority (91.4 %) reported they were currently married, while only 3.19 % were never married. Also, more women lived in rural areas (55.3 %) than urban areas (44.7 %). Regarding household wealth quintiles, 24.6 and 22.2 % of women were in the poorest and poorer quintiles, respectively, and only 15 % of the respondents were in the richest quintile. For education, a substantial portion of women indicated they had secondary education (44.3 %) and higher education (28.4 %), while only 2.06 % of the respondents did not have any formal education.

Bivariate Analysis

Results from bivariate analysis are shown in Table 2. The bivariate findings indicated that cultural, life course, demographic, and socioeconomic factors were significantly associated with women’s experience of forced first sexual intercourse in the Philippines. For ethnicity, Cebuano (OR = 1.46; p < 0.001) and Ilonggo women (OR = 1.19; p < 0.05) and women from the other ethnic groups (OR = 1.30; p < 0.001), compared to Tagalog women, were all more likely to report their first sexual intercourse as forced. Compared to Roman Catholics, women who reported that they did not belong to any religious denomination were significantly more likely to experience forced first sexual intercourse (OR = 2.04; p < 0.05). Regarding age at first sexual intercourse, we found that compared to women who had sexual debut at the age of 15 or earlier, those who did so at the age of 16 or older (OR = 0.63; p < 0.001) and at union (OR = 0.47; p < 0.001) were both less likely to report being forced at their sexual debut. Also, a year’s increase in the age of women decreased the odds of reporting first sexual intercourse as forced (OR = 0.99; p < 0.001). Moreover, currently married (OR = 0.66; p < 0.001) and formerly married women (OR = 0.80; p < 0.05) were less likely to report their first sexual encounter as forced than the never married. Also, we found that women in rural areas were more likely to describe first sexual intercourse as forced than their counterparts in urban areas (OR = 1.16; p < 0.001). For socioeconomic variables, household wealth quintiles and education were significantly associated with forced first sexual intercourse. Compared to women from the poorest households, those from the middle (OR = 0.87; p < 0.01), richer (OR = 0.76; p < 0.005), and richest households (OR = 0.78; p < 0.001) were all less likely to report forced first sexual intercourse. Similarly, women with secondary (OR = 1.18; p < 0.001) and primary education (OR = 1.16; p < 0.001) were both more likely to describe their sexual debut as forced than those with higher education.

Table 2 Bivariate analysis of selected dependent and independent variables
Full size table

Multivariate Analysis

Bivariate results are useful but limited because they only represent the gross effects of the independent variables on the outcome variable. The multivariate models shown in Table 3 provide the net effects of the independent variables on the outcome variable. Model 1 included cultural variables, such as ethnicity and religion. Consistent with the bivariate results, Model 1 indicated that Cebuano (OR = 1.46; p < 0.001) and Ilonggo women (OR = 1.19; p < 0.01) and women belonging to other ethnic groups (OR = 1.28; p < 0.001) were all more likely to describe first sexual intercourse as forced than their Tagalog counterparts. Also, compared to Roman Catholics, women who did not belong to any religious denomination were more likely to report forced first sexual intercourse (OR = 1.97; p < 0.05).

Table 3 Multivariate analysis of “Forced first sexual intercourse” among women in the Philippines
Full size table

Model 2 further controls for demographic and life course variables. Consistent with the bivariate results, we found that women who reported sexual debut at age 16 or later (OR = 0.65; p < 0.001) and at union (OR = 0.51; p < 0.001) were less likely to report sexual debut as forced than those who reported sexual debut at the age 15 or earlier. Demographic variables also remained statistically robust. A year’s increase in the age of women decreased the odds of reporting first sexual intercourse as forced (OR = 0.99; p < 0.001). Also, compared to never-married women, their currently-married counterparts were less likely to report forced sexual debut (OR = 0.67; p < 0.001). Similar to the bivariate result, we found women in rural areas were more likely to report forced first sexual intercourse than those in urban areas (OR = 1.15; p < 0.001). It is noteworthy that the significant impact of having no religious affiliation on forced sexual debut vanished after controlling for demographic and life course variables (see Models 1 and 2 of Table 2). Specifically, this effect vanished when the variable for age at sexual debut was added to the model. A cross-classification analysis of religion and age at sexual debut (results not shown here) indicated that women who did not belong to any religious denomination were more likely to experience premarital sex than those identifying as Roman Catholics, Protestants, other Christians, and Muslims.

In Model 3, we included socioeconomic variables (household wealth quintiles and educational background) and found that the significant effects of socioeconomic variables were largely attenuated (see Table 1 and Model 3 of Table 2). However, household wealth quintiles remained statistically significant. Compared to those from the poorest households, women from richer households were less likely to describe their first sexual intercourse as forced (OR = 0.85; p < 0.05).

Discussion and Conclusions

Sexual violence is a global public health and human rights issue, due largely to its association with sexual, physical, and psychological health problems among victims (WHO 2002). While consensual first sexual intercourse has been studied extensively in academic research, there is a growing, yet scant, body of literature on women’s experiences of forced first sexual intercourse (Population Council 2004). A few studies have noted that some Filipino women have been vulnerable to forced sex (Ramiro 2005; Raymundo and Cruz 2004), but results from these studies were mostly descriptive. Using the 2008 Philippines Demographic and Health Survey (PDHS), we explored what influences women’s experiences of first forced sexual intercourse in the Philippines.

We found that ethnicity was significantly associated with forced first sexual intercourse in the Philippines. Specifically, Cebuano and Ilonggo women were more likely to report forced first sexual intercourse than their Tagalog and Ilocano counterparts. These ethnic variations in sexual coercion have been attributed to differences in societal expectations and ethnocultural norms that emphasize patriarchy among the two ethnic groups (Hunt and Gatbonton 2000; Hunnicutt 2009). Such expectations may create platforms that favour male sexual entitlement and offer women limited options to challenge sexual advances even at the least coercion from men. Aguilar (2008) asserted that male power among Cebuano men is strongly linked to the active expression of sexuality that may even encourage risky sexual behaviours. In this regard, men may easily get away with risky sexual acts, including forced sex, because society often overlooks these risky acts as they fit into the concept of maleness (Angeles 2001).

The finding that currently married women were less likely to report forced sexual debut than their never-married counterparts is also important. This finding is consistent with a report from the US Department of State (2007) showing that, although sexual violence may be common among married women in the Philippines, such cases are often not reported as they have far-reaching implications, including possible marital dissolution. This is particularly important in the Philippines, given that divorce is illegal and marital dysfunction is considered extremely deviant due to cultural ideologies derived from Catholicism, familism, and collectivism (Yodanis 2005). The “Anti-Violence Against Women” and “Anti-Rape” laws prohibits forced sex both outside and within marriage in the Philippines (United Nations 2008), yet women are constrained in reporting cases of sexual violence due to stigma and the implications this may have on their marital unions.

Our findings demonstrate that the age at which women experience sexual debut strongly influences whether their sexual experiences will be forced or not. For instance, women who experienced first sexual intercourse at the age of 16 or later, or at union, were significantly less likely to report their sexual debut as forced than those who experienced it at the age of 15 or younger. Similarly, younger women were more likely to experience forced first sexual intercourse than their older counterparts. These findings corroborate previous studies that argue that younger girls are exposed to sexual violence, largely because they lack sexual negotiation power and self-efficacy to reject forced sex, making them vulnerable to perpetrators who mostly use threats and deception (Koenig et al. 2004; Population Council 2004; Santhya et al. 2007). Furthermore, formal sex education rarely exists in the Philippines (de Irala et al. 2009), thus making teenage girls vulnerable to early sexual intercourse, which occurs forcefully than voluntarily.

Our analysis indicated that women who did not belong to any religious group were more likely to experience forced first sexual intercourse compared to Roman Catholics (see Model 1 in Table 3). However, this effect vanished after controlling for age at first sexual debut in Model 2. A cross-classification analysis of religion and age at sexual debut showed that a higher proportion of women who had no religious affiliation experienced sexual debut earlier than those identifying as Roman Catholics. Thus, it appears that Roman Catholics may have lower odds of experiencing first forced sexual intercourse, mainly because of delays in the timing of their sexual debut. This finding corroborates the protective role of religion not only on the timing of sexual debut, but also possibly the timing of sexual violence among Filipino women. The finding is consistent with a study by Wells et al. (2013) that also demonstrated that Roman Catholic traditions provided religious or special services and resources that discouraged violence against Filipino women. In a related study, Ellison et al. (1999) found that regular attendance at religious services was inversely associated with self-reported perpetration of domestic violence for both men and women.

On the relationship between SES and forced first sexual intercourse, we found women from the richer households were less likely to encounter forced sexual debut than those from the poorest households. This finding partially demonstrates how socioeconomic inequality affects the sexual health of women; in particular, forced sexual intercourse. Socioeconomic inequality may increase women’s financial dependence on men, which often leads to sexual exploitation (Tenkorang and Owusu 2013; UNICEF 2000). This may be the case in the Philippines where there is a strong traditional gender stereotype that men should fulfill their educational needs to be future breadwinners, while women rely on their husbands’ economic successes (Aguilar 2008). Also, we found women in rural areas were more likely to report first sexual intercourse as forced compared to women in urban areas. This may reflect socioeconomic inequalities between rural and urban areas in the Philippines. Such differences may be linked to a lack of access to information and resources about sexual and reproductive rights, making women in rural areas particularly vulnerable to forced first sexual intercourse (Wibulpolprasert 1999).

Several policy implications are worth mentioning. First, it is clear from our results that there is a class dimension to forced sexual intercourse in the Philippines, as the poorest women are particularly vulnerable. This means providing economic opportunities for poorer women may be important in reducing the prevalence of sexual violence. Also, given that most teenagers in the Philippines learn about sexuality from their parents and friends (de Irala et al. 2009), interested stakeholders could deliver formal sex education to inform youth about the risks of early sexual initiation, including how this is linked to forced sexual intercourse. Rural dwellers in the Philippines may benefit immensely from information related to their own sexual and reproductive health. Importantly, given ethnic differences in self-reported forced sexual debut and the gender expectations that might inform these differences, community-based programs may play an important role in encouraging redefinition of gender roles and socializations among women and men. This study may also be helpful in creating awareness about the vital role that religion can play in the fight against all forms of violence against women. Religious and professional premarital counselling could also create spaces for educating couples on the laws on anti-rape and anti-violence against women, perhaps encouraging reporting of forced sexual debut in the Philippines.

While these findings may be useful for policymakers and interested stakeholders, there are some noteworthy limitations. First, the PDHS consists of cross-sectional data collected contemporaneously, indicating that causal relationships between the dependent and independent variables cannot be established. Therefore, our findings are limited to statistical associations. Use of longitudinal and qualitative data may better address this concern for future research. Concerns are also raised about the reliability of surveys based on self-reports, especially when they border on sensitive and stigmatized issues such as sexual violence. It is thus possible that forced sexual debut is underreported. Also, as women’s perceptions of sexuality are culture-specific (Population Council 2004), it is important to note that the definition of forced or wanted sexual intercourse in the Philippines is not generalizable to other cultures. A mixed-methods approach may be crucial to capturing the meanings of forced sex and prevalence in follow-up studies.

Despite these limitations, this paper is one of the few that used a large nationally representative dataset to explore the issue of forced first sexual intercourse in the Philippines. This paper reinforces ideas of ethno-cultural norms and socio-economic inequalities as important predictors of sexual violence in the Philippines and adds to the plethora of evidence that underscores gendered violence as a function of structural inequalities stemming from cultural and socio-economic differences.