Unwanted sexual activity among married women in urban China.

This article examines one of the few population-based surveys of unwanted sexual activity anywhere. Plausibly, unwanted sexual activities within marriage could be common in places such as China. Long dominated by a patriarchal sexual regime, Chinese women and men were taught that women have few sexual desires of their own, that male's interests should predominate, and that marital sex is synonymous with reproduction (Pan, 1993; Ruan, 1991). Despite greater equality for women in the public sphere over the last half-century, repression of many forms of sexual expression remained common (Renaud, Byers, & Pan, 1997). More recently, with economic reform, foreign contact, and a more open media, sexual norms have liberalized and there is more emphasis on sexual satisfaction for women (Bullough & Ruan, 1994; Evans 1995; Higgins, Zheng, Liu, & Sun, 2002; Zhang, 1999). Thus, precisely what one should expect in China is unclear. With data from a survey that is representative of the urban adult population aged 2064, this study provides the first national estimates of unwanted sexual activity and its correlates.

Our analysis is of unwanted sexual activity--specifically sex that does not necessarily involve use of physical force (Basile, 1999; Sprecher et al., 1994). One way to characterize the phenomenon is that of "... situations in which a person freely consents to sexual activity with a partner without experiencing a concomitant desire for the initiated sexual activity. In a sense, they feign sexual desire or interest (O'Sullivan & Allgeier, 1998, p. 234)." The reasons women "go along" vary. Wives often report that unwanted intercourse is related not so much to force or threats of force as to a sense of wifely duty (what Finkelhor & Yllo (1985) label "social coercion") and to the fear that their husband will seek other partners if he is not satisfied sexually at home (Im-em, Archvanitkul, & KIanchanachitra, 2004; Knodel et. al., 1999). In one U.S. national study, 34% of women reported unwanted sex with a husband or partner in their lifetime, and 20% of women reported unwanted marital sex with their current husband ever (Basile, 2002). Reasons for consenting included: it was her "duty", he begged and pleaded, he spent money on her, he back rubbed or kissed her, he bullied and humiliated her, he threatened to hurt her, he used physical force on her. While 4-9% of women were threatened or forced by the current husband, other reasons were common. In a random sample of San Francisco women, 26% reported ever having unwanted sex with her husband--14% reported spousal rape (Russell, 1982). In a largely urban study of women aged 25-49 in Thailand, 16% reported unwanted sex, including by force (2-4%) or out of fear (15%), last year (Im-em et al., 2004).

Methodological issues abound. Most research has been for dating relationships, which may or may not be generalizable to marital relationships (Blythe et al., 2006; Impett & Peplau, 2002; O'Sullivan & Allgeier, 1998; Sprecher, Hatfield, Cortese, Potapova, & Levitskaya, 1994). The negative consequences of forced sex have been documented (Basile, 2002; Blythe, Fortenberry, Temkit, Tu, & Orr, 2006). Sexual coercion victims experience anxiety, depression, relationship and sexual difficulties, and poor health (Becker, Skinner, Abel, & Cichon, 1986; Feldman-Summers, Gordon, & Megher, 1979; Gidycz & Koss, 1991; Mandoki & Burkhart, 1991; though also see Oswald & Russell 2006). Whether similar consequences appear in more general experiences of unwanted sex is uncertain. Our study extends the existing literature by assessing whether unwanted sexual activity in marriage, typically without force, is subject to the same risk factors and has consequences similar to other forms of unwanted sex.

Risk Factors

Previous research on women's sexual relationships suggests multiple risk factors for women's submission to unwanted sexual activity. We group these into four categories: (1) relationship quality, (2) attitudes and expectations, (3) bargaining power, and (4) control variables, including lifestyle, stress, and physical vitality (Koss, 1985; Mandoki & Burkhart, 1991). We examine the effect of all these types of items together, guided by the conceptual model in Figure 1.

[FIGURE 1 OMITTED]

Relationship Quality

A large literature suggests that in sexual behavior relationship quality matters greatly to women (Basson, 2001; Kaplan, 1974; Masters, Johnson, & Kolodny, 1995). Various aspects of marital relationship have been examined. Both daily intimacy and various sexual arousal techniques are reported to increase the likelihood of women enjoying sex (Basson, 2001). In a Chinese study, when women reported no more than one minute of foreplay only 40% rated their sexual satisfaction as good or very good. In contrast, when women reported more than 20 minutes of foreplay, 73% rated their sexual satisfaction as good or very good (Liu, Ng, Zhou, & Haelerle, 1997). In Thailand, women were more likely to report unwanted sex when hit by their partner (Im-em et al., 2004). Western feminist theories argue that in patriarchal society, men have been mis-socialized--to control and dominate and not to express intimacy or to know what a woman needs to be sexually satisfied (Blumstein & Schwartz, 1983; DeMaris, 1997; Hochschild, 1989; Schwartz, 2000; Schwartz & Rutter, 1998).

Attitudes/Expectations

There are two contradictory hypotheses about the connection between women's sex-role attitudes and unwanted sexual activity (Ageton, 1988). Social control theory suggests that traditional socialization puts women at greater risk because of passivity and the tendency to accede to male demands. At the same time, nontraditional women may be at higher risk for sexual victimization because of nonconformity and violation of traditional values. Empirical evidence, mainly focused on forced sex, has been mixed. While some studies found that victims of sexual assault were more likely to hold nontraditional sex-role attitudes (Muehlenhard & Linton, 1987), others observed that college women subscribing to traditional sex-role attitudes were more likely to report sexual coercion (Murnen & Byrne, 1991), and still others found no relationship between sexual victimization and measures of hyperfemininity and hypermasculinity (Koss, 1985; O'Sullivan & Allgeier, 1998). Kalof (2000) speculates that a connection might exist between attitudes and perceptions of aggressive behavior, which could lead to similar coercive experiences being reported differently by women who hold dissimilar views. From the literature on raising women's consciousness, we would expect nontraditional women to be more discriminating about how they are treated, which, in turn, might make them more critical and more likely to report unwanted sexual activity.

Bargaining Power

Using personal resources to bargain over contested issues is one strategy a woman might use to gain sexual empowerment. Earning an independent income from her husband over which she has control can lead to increased decision making authority in the household (Wolff, Blanc, & Gage, 2000). Does this autonomy spill over into the bedroom, providing income-earning women a stronger voice in negotiating sexual activity with their husband? China provides a setting to test this relative resources model because urban Chinese women often approximate their husband in education, occupation, and income (Tang & Parish, 2000). The consequences of more bargaining power need not all be happy ones. More resources for the woman have the potential to challenge traditional meanings attached to male superiority in ways that could elicit a backlash from male partners. This in turn might lead to fear of violence, withdrawal of economic support, and infidelity (Fox, Benson, DeMaris, & Van Wyk, 2002; Rivers et al., 1998). These tendencies could be further exacerbated by growing prostitution, which provides men with alternative sexual outlets. With this alternative outlet, women could less often use withdrawal of sex as a tool to control men (Im-em et al., 2004; Knodel et al., 1999). Increased housing availability and increased residential mobility might also cause women often to be isolated from family and friends and, therefore, lacking in the social support networks that might help protect them from unwanted sexual activity (Dobash & Dobash, 1979; Shields & Shields, 1983; Smuts, 1995).

Control Variables

Other items that emerged as important in previous research include lifestyle, stress, and physical vitality. Studies of college students report that more sexually experienced women are at higher risk of sexual coercion (Koss, 1985; Koss & Dinero, 1989; Lottes & Weinberg, 1997). Although no studies have looked the relationship between women's sexual experience and unwanted marital sex, it is plausible that for a married woman, previously unsuccessful relationships may lead her to be more accommodating to her husband's sexual demands in order to maintain the marriage. Previous research also found that unwanted sexual activity occurs more often in relationships involving drinking (Abbey, Ross, McDuffie, & McCauslan, 1996; Muehlenhard & Linton, 1987; Testa, Vanzile-Tamsen, & Livingston, 2004). In addition, young children in the home may impose a major burden on women's sexual life (Liu et al., 1997; Pimentel, 2000). Interest in sex may also decline for women with physical and psychological well-being problems, and simply age and hormonal changes (DeLamater & Sill, 2005; Everaerd, Laan, Both, & van der Velde, 2000; Trudel, Turgeon, & Piche, 2000). When interest in sex declines, women may become less willing to engage in sex and as a result, they may be more likely to perceive any sexual advance from their husband as unwanted.

Consequences

Our question is whether, even largely in the absence of force, unwanted sexual activity has consequences for psychological well-being. Some studies find that psychological abuse alone (e.g., verbal abuse and abuse of power and control) is associated with negative physical and psychological well-being (Coker et al., 2002; Smith et al., 2002). One study of college students found that despite reports of some positive outcomes associated with their consent to unwanted sex, including promoting of intimacy in the relationship, many reported negative outcomes such as emotional discomfort, feeling uncomfortable about engaging in "meaningless sex," and feeling disappointed in oneself (O'Sullivan & Allgeier, 1998). Whether such discomfort has long run consequences is less clear.

What are the hypotheses that emerge from this literature review? They are summed up in Figure 1. Or, in words, the hypotheses are, first, that reports of unwanted sexual activity have multiple determinants ranging from relationship quality to physical vitality, with relationship quality, attitudes, and bargaining power being particularly important for women. Second, we hypothesize that even less coercive forms of unwanted sexual activity have negative consequences for psychological well-being, net of the feedback path from psychological distress to reports of unwanted activity, or, in short, that milder forms of unwanted sexual activity deserve serious attention.

Method

Participants

Data come from the 1999-2000 Chinese Health and Family Life Survey (CHFLS). With the exclusion of Tibet and Hong Kong, the sample is nationally representative of the adult population of China aged 20-64. Following standard procedures for complex samples (Levy & Lemeshow, 1999), the probabilistic sample was drawn from 14 strata and 48 primary sampling units (counties and city districts), with probabilities of selection proportional to population size at each of the four sampling steps down to the individual. Cities with a high prevalence of sexually transmitted diseases were oversampled because one of the original goals of CHFLS was to understand the consequences of changing sexual behavior on sexually transmitted disease. Among the sampled individuals, 3,821 completed the interview, yielding a final response rate of 76%.

Translation and cross-cultural adaptation of the questionnaire to Chinese was performed by a forward translation from English to Chinese followed by an independent backward translation from Chinese into English. The Chinese questionnaire was pretested in China in three field trials. Also, 50 husband-wife pairs were given shortened versions of the questionnaire which produced modest agreements about shared sexual behavior (Kappa = .35). Another 50 respondents had repeat interviews after a gap of two months and the agreements on the 21 items about sexual behavior were substantial (Kappa = .75).

Most interviewers were trained middle-aged social workers and researchers who remained with the project throughout the interview period of one year. In the field, interviewers were of the same sex as the respondent. For the sake of privacy, interviews took place outside the homes of the respondents, normally in a private room in a hotel in big cities or in a meeting facility in smaller locales. Oral and computer-entered consent was obtained prior to the hour-long interview. The first part of the interview that included basic demographic questions was computer assisted face-to-face interview--the interviewers read out the questions from the laptop computer to the respondent and entered the answers into the computer. The second part of the interview that included sensitive sexual behavior questions was self-administered--with the interviewer lingering in the hall or nearby to help in case of difficulty. In urban areas, 9% of respondents (older, less literate) needed constant help during this final phase of the interview. Another 62% needed rare or occasional help with the computer. A total of 5% of urban respondents got the questions via audio from earphones attached to the laptop computer, including audio in a local dialect. Interviewers mostly used the Mandarin version of Chinese. For 95% of respondents, the self-administered part of the interview was based on reading Chinese characters, with the characters spanning multiple dialects in Chinese. The methods were approved by institutional review boards at the University of Chicago, Chicago, Illinois, and Renmin University, Beijing, China. Details about sample design and the final questionnaire used in this study are available at [details to be supplied after review process]. The sample for this particular analysis included 1,538 urban women, for whom we had a large sample size and stable estimates. The urban respondents were further limited to 1,283 women who were currently married and of those, to the 1,181 women who were sexually active. Further limiting the data to those with data on our key analytical variables, the final sample included 1,127 respondents.

Measures

Unwanted Sexual Activity. The interview had three questions that were combined into a single, composite index of unwanted sexual activity: (a) Unwanted intercourse. "Up till now, has it occurred that you were unwilling but still had to concede to have sex with someone (including between husband and wife)?" The responses were "yes" or "no." Those who answered "yes" were further asked at what age(s) these events took place and the relationship of that person to her. So as to allow for potential confusion among two types of age accounting in China, we included any event that occurred in the last two years as 1 and all other responses as 0. (b) Unwanted act. Following a set of questions about specific sexual acts--oral sex, genital touching, anal sex, etc. during the year--women were asked the following: "Some people insist that their partner does certain things or adopt certain styles in sex that their partner does not like. In the past 12 months, did your partner often ask you to do things that you did not like to do during sexual intercourse?" The responses were coded 0 ("never") and 1 ("sometimes" or "often"). (c) Sex to satisfy partner. "Some people do not want to have sex themselves, but they do it to satisfy their partner. In the past 12 months, how often did this situation occur to you?" The responses were coded 0 ("never") and 1 ("rarely," "sometimes," or "often"). The sum of the three items provides an index of unwanted sexual activity. In the analyses presented here, this index is treated as a continuous variable. Additional analyses using an ordered probit specification yielded similar results. Also, separate analyses of each of the three constituent items in our summary scale produced results that largely parallel the summary results presented in the body of the text (see Appendices A and B). Throughout, we use the three terms unwanted intercourse, unwanted act, and sex to satisfy partner to describe the three constituent elements described above and then the term "unwanted sexual activities" to describe the composite scale that sums together all three individual items.

Psychological Distress. A 4-item short form of Zung's Self-Rating Depression Scale (SDS) (Zung, 1965), validated and translated in China (Leung, Lue, Lee, & Tang, 1998), was used to measure psychological well-being. The four items each asked about conditions over the last three months: (a) "In the past three months, did you sleep well or poorly at night?" (b) "... did you often feel depressed or bored?" (c) "...did you feel fatigued for reasons unknown to you?" and (d) "... have have you felt more irritable than usual?" The three-step answers to the first item ranged from 1 (usually slept well) to 3 (always slept poorly) and the answers to the remaining three items ranged from 1 (never) to 3 (often). Factor analysis showed the four items loaded on one factor. The factor scores were rescaled to range from 1 to 3 with higher values indicating higher levels of psychological distress; Cronbach's alpha is .64.

Relationship Quality. We examined five aspects of a couple's relationship, with coding such that a high score implied a risk factor for unwanted sex: (a) "Any hitting" was coded 1 (as opposed to 0) if she answered "yes" to either of the questions: "...has your partner ever hit you?" or "... have you ever hit your partner?" (b) "Daily intimacy absent" was from the question: "In the past 12 months, other than during sex, how often did you and your partner have intimate contact (including kissing, embracing, and caressing)?" Responses ranged from 1 (often) to 4 (never). (c) "Foreplay insufficient" was based on the question: "Some people hug, kiss, and caress their partner a lot before they have sexual intercourse. How did your partner do in this respect?" Responses ranged from 1 (enough) to 3 (none). (d) "Husband doesn't know wife's (sexual) need" was based on her response to the question: "In the past 12 months, did your partner know how to please and satisfy you during sexual intercourse?" Responses ranged from 1 (knew all) to 4 (knew none). (e) "Genital caressing of wife absent" was based on the question: "In the past 12 months, did your partner often caress your genitals during sexual intercourse?" The coding was 1 (never) and 0 ("often" or "sometimes").

Attitudes/Expectations. The three indicators of attitudes and expectations were as follows: (a) "Thinks sex is dirty" was based on the question: "Some people feel dirty about their partner's genital, secretion, semen, menstrual blood, and so on. Have you ever felt dirty of these things?" Responses ranged from 1 (never) to 3 (often). (b) Education was simply her education in years. It was used as a proxy for raised expectations and consciousness about sex life (Inkeles, 1973). (c) "Permissive sex attitudes" was a scale based on four items: (i) "Some say that one can have sex just for pleasure with someone whom he or she is not in love with." (ii) "Some say that it is OK to have sex with someone other than your spouse after marriage." (iii) "Nowadays in our society, some married people have sex with those other than their spouse. Do you think that each such case should be treated individually or that these people should all be punished?" (iv) "Nowadays in our society, some couples have sex when they are dating, and they eventually get married. Is this a moral issue?" The scale was the average of responses to the four items, with a range of 1 to 4 and a reliability coefficient of .61.

Bargaining Power. Five indices tapped different ways in which a woman's bargaining power might be weakened: (a) Wife's share of the couple's joint earnings was recoded into three categories: 0-24%, 25-69%, and 70-100%, based on preliminary analysis of different cutting points. (b) Parental economic background was based on the question: "When you got married (or began to live together) with your partner, whose family had relatively better material conditions?" The three responses were: partner's family, both about the same, and my family. For wife's earnings share and parental background, the middle categories were used as the reference so that we could capture a potential curvilinear relationship between these variables and unwanted sexual activity. (c) "Other adults not in household" indicated whether there was any adult other than the husband and wife (e.g., her or his parents) in the household who might dampen husband-to-wife demands and abuse. It was coded 1 if no

such adults were in the household, and 0 otherwise. (d) "Age gap" was the husband's age minus the wife's age, with extreme values truncated to a range of -4 to 11 years. (e) "Paid sex of men in locality" was the average of reports by men (not the participant's husband) in the sample for each community, with the hypothesis being that when commercial sex was readily available women had less ability to resist their husband's sexual demands. This variable was coded to have 10% point units and to range from 0 to 2.6 (i.e., 0% to 26%).

Control Variables. Respondents' self-reported number of sexual partners in their lifetime was highly skewed with nearly 92% reporting only one partner, thus its logarithm form was used in the analytical models. "Alcohol consumption" was based on how often the respondent drank in the past year; we compare those who had drunk alcohol with those who did not drink at all. We also included a variable indicating whether any children aged 0 to 6 were living in the household. "Sexual dysfunctions" was an index of four items asking the respondent whether during sex in the last 12 months she ever had the following experiences: (i) couldn't get any pleasure (uncomfortable, unaroused); (ii) any genital pain; (iii) vaginal dryness and insufficient lubrication; and (iv) inability to achieve an orgasm. The responses to each question ranged from 1 (never) to 3 (yes, for more than two months). The index was the average of the responses to each of the four items, with higher values associated with more sexual dysfunctions. We divided respondent's age by ten so that each unit represented a ten-year increase in age.

Correlates of Psychological Distress. For the analysis of the effect of unwanted sexual activity on psychological well-being we included other correlates of psychological well-being as control variables. In addition to age and education, we also included self-rated physical health, the total of the couple's joint earnings, perceived daily care, desire to lose weight, body mass index, self-rated attractiveness, and stressful living conditions (Luo, Parish, & Laumann, 2005). Respondents were asked to rate their physical health on a five-point scale ranging from 1 (poor) to 5 (excellent). Couple's joint earnings were log transformed. Perceived daily care was based on the question: "Is your partner sufficiently considerate and affectionate to you in daily life?" with responses ranging from 1 (none) to 4 (sufficient). When asked whether they wanted to lose weight, responses ranged from 1 (somewhat, very much) to 0 (no). Body mass index was weight in kilograms divided by the square of height in meters. Self-rated attractiveness was based on the question, "In your opinion, are you attractive in the eyes of the opposite sex?" with responses ranging from 1 (not at all) to 4 (very much). Stressful living conditions were indexed by the question: "How seriously do your living conditions affect your sex life?" with responses ranging from 1 (no interference) to 3 (a lot).

Statistical Procedures

All analyses were weighted, first using population weights that adjusted for the intentional oversampling of coastal and urban settings with a high prevalence of sexually transmitted diseases. After comparison of the resulting age distribution to census results for 2000, we adjusted weights by age to compensate for the smaller number of usable interviews of 20-29-year-olds and 50-64-year-olds. With these adjustments, the percentage distributions by age, occupation, urban residence and education closely paralleled those in the national census. We adjusted standard errors in our regression analyses for sample stratification (sampling strata independently) and clustering (sampling individuals within each of 48 primary sampling units) (Skinner, Holt, & Smith, 1989).

We ran three sets of analyses: The first set regressed the unwanted sexual activity index on age and a single additional correlate, thereby examining the effect of a particular variable of interest on unwanted sexual activity when only age was controlled. A similar set of regressions were performed for psychological distress. A second set of analyses had two separate multivariate regressions, with unwanted sexual activity and psychological distress separately regressed on all their respective correlates. These two regressions were estimated with Ordinary Least Squares regressions using STATA 9.0. A third set of analyses estimated a simultaneous equations model with two equations: the unwanted sexual activity index regressed on all its correlates in a single equation and psychological distress regressed on all its correlates in another single equation (see Figure 1). This model added sexual dysfunctions to the equation for unwanted sexual activity and it also specified a reciprocal relationship between unwanted sexual activity and psychological distress. This model was estimated with maximum likelihood parameter estimates with robust standard errors using Mplus 3.1--a structural equation modeling program that can adjust for complex sample design (Muthen & Muthen, 2004).

Results

Prevalence of Unwanted Sexual Activity

The first activity was unwanted intercourse with her husband, with this intercourse including both forced and unforced events. The survey had two indicators of the relative use of force and harm in unwanted sex. Following a question about whether a woman ever experienced sex that she did not want, she was asked whether this unwanted sex harmed her. In the total sample of currently married, sexually active urban women 32% reported ever engaging in unwanted sex with her spouse--and the constituent percentages in this 32% total were 19% no harm, 11% some harm, and 2% considerable harm. Or otherwise stated, of those reporting unwanted sex, a total of 40% reported any degree of harm and 6% reported considerable harm in past events.

Second, in the final 11 field sites of the study, 300 urban women responded to a question about whether during the years of her relationship with her husband he had ever physically forced (qiangpo) her to have intercourse that she did not want. Among those who reported any unwanted sex in the past, one-fifth reported that at some point in the past this unwanted intercourse involved force. Thus, the unwanted intercourse analyzed here is a mixture of forced and non-forced, with most events not involving force.

The survey also provided information on how frequently unwanted intercourse had occurred. Most women reported that unwilling intercourse occurred several (48%) or many (43%) times during her relationship with her husband, leaving few women (9%) reporting that unwilling intercourse occurred only once. Among women who experienced unwilling intercourse, the median span of time for occurrences of unwilling intercourse was five years, which equated to 67% of the duration of her relationship with her husband. Thus, when reported, unwanted intercourse tended to be a "chronic" condition. In total, 21% reported that one or more of these events occurred during the previous year (Table 1).

The second unwanted sexual activity last year was being compelled to perform sexual acts that she did not like. In all, 22% reported this type of experience (Table 1). The third unwanted sexual activity was performing sex just to satisfy her husband. This activity was common, with percentages of 28% never, 26% rarely, 40% sometimes, and 6% often (Table 1). With all but the never response coded as 1 and then summed with the other two measures, the composite sexual activity index had a median of 1.0 and a mean of 1.16 unwanted events last year (Table 1).

Correlates of Unwanted Sexual Activity

Relationship Quality. With only a few exceptions, measures of marital quality (and sexual intimacy) were related to unwanted sexual activity in predictable ways (Table 2a). Examined singly (with only age controlled) in column one, all the relationship quality items were statistically significant. These patterns were sustained in the multivariate analysis of Model II. When sexual dysfunctions and psychological distress were added in Model III, only hitting lost statistical significance. The role of intimacy stood out. In the final two models, we used a composite measure of intimacy that was the average of the standardized values of insufficient foreplay and absence of daily intimacy, rescaled to range from 1 to 4. The effects of this composite variable were sizable. Recall that the unwanted sexual activity index ranged from 0 to 3. Complete absence of daily intimacy and foreplay (in comparison to ample intimacy and foreplay) increased unwanted sexual activities by .30 points (b=.10, p < .01). The consequences of the husband's complete ignorance of his wife's sexual were also considerable, increasing reports of unwanted sexual activities by .21 points (b = .07, p < .001). Thus, relationship quality was important--the only surprise being that instead of exacerbating reports of unwanted sexual activity, absence of genital caressing decreased those reports.

Should "any hitting" have been grouped with the bargaining items below? Potentially. Among women who reported any physical slapping or other hitting ever in the relationship, only one-fifth of these reports were of the wife hitting her husband without his hitting her. In short, most hitting involves him hitting her. However, hitting in either direction has similar consequences--similar coefficients--for reports of unwanted sexual activity. Hitting of either type is highly related to reports of daily intimacy. And, in any event, hitting fails to have a statistically significant effect net of other conditions in the final, full Model III. Thus, exactly where hitting should be classified is to some extent a moot point.

Attitudes/Expectations. Several measures of sexual attitudes were related to unwanted sexual activity, both singly in Model I and in the multivariate Models II and III (Table 2a). The most robust influences were education and the belief that sex was dirty, with each statistically significant across all three models. The final multivariate model showed that an additional five years of education increased unwanted sexual activity by .20 points (b = .04, p < .001). Similarly, women with the most extreme beliefs about sex being dirty were .30 points (b =. 15, p < .001) more likely to report unwanted sexual activity. Though consistent in direction across all three models, permissive sex attitudes were significant at the .05 only when considered singly in model I.

Bargaining Power. Impaired bargaining power put women at risk. Net of all other factors, reports of unwanted sexual activity increased when women had an income share of less than 25% (b = .09, p < .05), when their husband's parents were from a better economic background (b = .14, p < .001), and when there were no adults other than the couple in the household (b =.14, p < .05). There were some exceptions to these patterns: Unwanted sexual activity declined only when her income share was at intermediate levels. At the highest level, when she earned 70% or more of the couple's income, unwanted sexual activity increased (b = .31, p < .05). Her parent's economic status failed to improve patterns of unwanted sexual activity. Contrary to expectations, a husband older than the wife reduced reports of unwanted sexual activity (b = -.02, p < .001). Also, high levels of paid sex in a locale were at least neutral for reports of unwanted sexual activity.

Control Variables. Number of sexual partners in her lifetime, alcohol consumption and sexual dysfunctions were positively associated with unwanted sexual activity. Even though only marginally significant in the full multivariate model, women with more sexual partners were more likely to report unwanted sexual activity with their spouse (b = .14, p < .10). The effect of alcohol consumption was substantial--women who consumed alcohol were .12 (b= .12, p< .001) points higher in unwanted sexual activities than women who consumed no alcohol. Unwanted sexual activities also increased with the presence of a small child (b = .17, p < .05).

Psychological distress was a strong correlate of unwanted sexual activity. In Model IIIa when the effects of other variables and a feedback effect of unwanted sexual activity on psychological distress were taken into account, women with the highest level of psychological distress reported .84 (b = .42, p < .01) more instances of unwanted sexual activity (Table 2a). Women who experienced persistent sexual dysfunction reported .74 (b= .37, p < .001) more instances of unwanted sexual activities.

Consequences of Unwanted Sexual Activity

Regression coefficients of psychological distress on unwanted sexual activities and other covariates from both the simple models and the simultaneous equations model are presented in Table 2b. Unwanted sexual activities had a significant effect on psychological distress in all three models. The effect was strongest when all other covariates were controlled and the feedback effect of psychological distress on unwanted sexual activities also taken into account. In this situation, an additional unwanted sexual activity increased psychological distress by nearly .20 (p < .001) on a scale that ran from 1 to 3 (Model IIIb). Other variables were included as controls and the results for these variables were generally consistent with an existing literature. Although the coefficients varied depending on which variables were included in the model, the general pattern was that psychological distress was positively correlated with desire to lose weight and stressful living conditions. It was negatively correlated with physical health, education, family income, daily care, body mass index, and self-rated physical attractiveness. In Model IIIb, the age coefficients describe a significant decline in distress after age 40.

To check on the robustness of the above patterns, we also split out the three constituent items in our unwanted sexual activity index. The multivariate results for each item largely paralleled those for the summary activity index (Appendices A and B). In the results for each type of unwanted sexual activity, fewer background items were significant (Appendix A). In the third equation for "sex to satisfy partner" her belief that sex is dirty, her education, her permissive sex values, and his understanding of her sexual needs appeared more consistently important than in the other two equations. However, for most background conditions, the sign of the coefficient was the same across equations. A separate analysis of the statistical significance of the differences of coefficients across these equations also found that few of the differences were statistically different at the p < .05 level. Also, the analysis of the determinants of psychological distress found that specific unwanted sexual activities were significant in each of the three equations (Appendix B). Thus, by-and-large, the analysis for specific items found results that paralleled the results for the summary unwanted activities index--at least in sign of the coefficients, if not always in level of statistical significance.

Discussion

This paper identified the prevalence, risk factors, and consequences of unwanted sexual activity among urban Chinese women.

Prevalence

The prevalence data included three types of reports from currently married urban Chinese women: The first was a report of ever engaging in unwanted spousal intercourse, with 32% (95% CI, 28-35) reporting at least one (often more) experience(s) during the lifetime of the relationship and 21% (CI, 19-23) reporting such an experience last year. The 32% lifetime figure is somewhat higher than the 16 to 26% reported in studies from the U.S. and Thailand (Basile, 2002; Im-em et al., 2004; Russell, 1982). Moreover, in a limited subset of 300 women questioned at the end of the field work, one-fifth of the women reporting ever experiencing unwanted intercourse with their husband also reported that their husband had forced them to have intercourse. This would produce an urban total of about 6% (32% x .20) of women ever experiencing forced intercourse with her husband. Because of variations in surveys methods used, comparisons to other countries are difficult. Accepting this caution, China's 6% for forced sex is modest compared to other large-sample, nationally representative surveys, which have found prevalence for attempted or forced sex with an intimate partner to range from about 6% to 14% (Krug et al., 2002, p. 152). Thus, while not exceptionally large in international comparison, force was a component for a subset of Chinese women.

In the Chinese data, the other two types of reports for experiences last year were for unwanted sex act(s) (22%) and for sometimes or often agreeing to have sex just to satisfy the husband (72% at least sometime). Thus, most women reported at least one of the three types of unwanted sexual activity during the past year.

Risk Factors

With controls for lifestyle, stress, prior sexual experience and physical vitality, the analysis of risk factors focused on three types of factors--relationship quality, attitudes/expectations, and bargaining. Results for items indexing these factors were often consistent with the extant literature. But there were also some patterns that were inconsistent, and it is these inconsistencies that invite special comment.

Unsurprisingly, relationship quality was important--though it was striking how much this importance was restricted to affective matters rather than "sexual techniques." Specifically, fewer women reported unwanted sexual activity when their husband provided daily intimacy and gave attention to foreplay and their specific sexual needs. Conversely, more women reported unwanted sexual activity when their husband touched them genitally. Additional analyses showed that other varieties of sex positions produced similar results. Entered one-by-one in the same equation as in Model IIIa, her being on top and oral sex provided by the man to the woman both produced not fewer but more reports of unwanted sexual activity. An index based on the count of multiple techniques--woman-on-top, kneeling position, caressing of her breasts, oral sex provided for her, caressing of her genitals, and anal sex--significantly exacerbated reports of unwanted sexual activity (OR = 1.29, p < .01). In short, a simple increase in "techniques" was counterproductive in this sample.

We suspect several things may have been occurring: Separate analyses found that among currently married, sexually active urban men of all ages, 40% reported having viewed pornography last year. (The percentage for the comparable group of women was 23%). Also among these men and net of age, education, and region, reports of the variety of sexual techniques used with his wife increased significantly when he also reported having watched pornography during the year. This pattern is consistent with men learning many of their sexual techniques from viewing porn. If so, wives may have interpreted these actions as too abrupt, too rough, and inconsistent with their traditional sexual scripts for love making. In short, more than techniques need to be changed.

The influences of attitudes and expectations were less than straightforward. Unsurprisingly, belief that sex was dirty produced more reports of unwanted sexual activity. Less obviously, more education and more permissive sex values (e.g., more approval of premarital sex) also increased reports of unwanted sexual activity. Two things may have occurred: First, more educated women with more permissive values may have experienced "consciousness raising" and, as a result, have had higher standards as to what was an acceptable set of behaviors and rewards during sex. Second, nontraditional women may have been at higher risk for unwanted sexual activity because of their violation of traditional values. That is, their husband may have taken more liberties with her because of what he saw as her greater openness to sex (Ageton, 1988; Muehlenhard & Linton, 1987).

Several dimensions of bargaining were significant--a woman reported more unwanted sexual activity when her income share was low, when her husband's parents were more economically advantaged than hers, and when there were no adults in the household other than the couple. Several other possible dimensions of bargaining were not significant--or significant in the "wrong" direction. Nevertheless, on further reflection, several of these "deviations" were not exceptional after all. First, her report of more unwanted sexual activity when her income share was high was consistent with a literature on spousal violence, which reports that when married to low income men, high income women are at risk (e.g., Wolff et al., 2000; Rivers et al., 1998). An alternative explanation is that the earnings-superior wives endured more unwanted sexual activity as a strategy to preserve their marriage despite their non-normative, higher-than-his wages. Some studies found that women with superior earnings take steps to neutralize their earnings power in the context of marriage by shouldering an extra burden of childcare and housework (Greenstein, 2000; Evertsson & Nermo, 2005). Thus, much as in spousal violence and housework, the "sweet spot" for spousal relations is when her income is approximately equal to his--a theme reminiscent of discussions of "peer marriage" (Schwartz, 1994). Second, the economic status of the husband's and the wife's parents were not related in a linear way to unwanted sexual activity. Her parent's superior economic status did not further reduce unwanted sexual activity. Instead, when her parents were at least as well off economically as his parents, unwanted sexual activity diminished.

The third exception to the bargaining hypothesis was that older husbands were not a threat but instead a source of fewer reports of unwanted sexual activity. It is possibly because older husbands were sexually less demanding--a simple age rather than relative age effect. It is also possible that women with younger husbands submitted to unwanted sexual activity for reasons that amount to discounting their non-normative marital age gradient--thereby "proving" that they could "keep up with" the sexual appetites of their younger husband as well as could younger women. Another exception to the bargaining hypothesis was that the consequence of men's frequent participation in paid sex in a community was at least neutral. And, with only age in the model in model I, it may have even been protective--suggesting a possible hypothesis that commercial sex in the community provided a relief valve for men who would otherwise make unwelcome demands on his wife.

Several other factors were important as controls for potential confounding influences on the patterns described above. As for sexual experience, lifestyle, and stress, the significant effect of prior sexual experience was consistent with a model suggesting that early sexual experiences trigger sexual attitudes and behavior in later life, which in turn affect women's experience of unwanted sexual activity (Koss & Dinero, 1989; Mandoki & Burkhart, 1991). Alcohol consumption, rare among many women, was significant at the p < .001 level in the final multivariate model. Women who drank were more susceptible to unwanted sexual activity. Presence of a pre-school child also increased reports of unwanted sexual activity. As for physical impediments, psychological distress, sexual dysfunctions and advancing age all increased reports of unwanted sexual activity--with age possibly indexing more a set of cohort effects (being socialized at an earlier time) as opposed to simple physical aging.

Sequelae 

One issue is whether the forms of unwanted sexual activity reported here--for example, engaging in sex only to please to one's husband--were so mild in form that they were of little consequence for a woman. The analysis of psychological distress suggested that these forms of unwanted sexual activity were of some consequence. Even after accounting for the potential feedback effects from psychological distress to unwanted sex, the significant negative effect of unwanted sexual activity on psychological well-being persisted. Though perhaps not as extreme as the psychological consequences of rape and battering (Campbell, 2002; Gidycz & Koss, 1991; Mandoki & Burkhart, 1991), significant effects were present nevertheless--suggesting that unwanted sexual activity is a phenomenon meriting further inquiry.

Limitations

Our study had several limitations. We had insufficient details about forced activity to differentiate types of unwanted sexual activity in marriage (Basile, 1999, 2002; Finkelhor & Yllo, 1985). We had insufficient information on the attitudes and behavior of the husband, factors that were ultimately responsible for women's unwanted sexual activity (Hall, Sue, Narang, & Lilly, 2000; Malamuth, 2003; Mandoki & Burkhart, 1991). Moreover, as in all studies based on cross-sectional data, we could not rule out strong feedback effects for example, hitting could have been as much the cause as the result of unwanted sexual activity. The structural equation model attempted to deal with this issue, but only for two dependent variables. Finally, as in all research that uses only people who are currently sexually active, our results could suffer from selection bias--that is, the people who experienced the most sexual problems may have simply quit having sex and therefore have no longer been available for observation. In separate, unreported analyses we attempted to deal with this issue using Heckman selection models. These produced results largely similar to those reported here. Nevertheless, that analysis was less thoroughgoing than what was reported here and problems could still remain.

Summary

In urban China, the proportion of married women experiencing unwanted sexual activity within their marriage was as high, or higher, than in other societies. Even when physical force was not the dominant cause of unwanted sexual activity, this activity had significant negative consequences for psychological well-being. Multiple elements were important for women's reports of unwanted sexual activity. The results suggested that husbands should be more caring of his wife in daily life, take more time in foreplay, and attend to her specific sexual needs. She needs to become more open to sexual experiences, losing traditional attitudes such as "sex is dirty." And, she benefits from a better bargaining position--including an income more nearly equal to that of her husband, coming from a similar economic background, and potentially having his or her parents in the household. These are all patterns that one might expect from past literature. What the results add is the warning that simple learning of new sexual techniques (whether through pornography or elsewhere) may be counter-productive when unaccompanied by affective care. Her rising expectations, driven in part by more education and exposure to modern values at sex may make her in the short run feel even more disappointed in her sex life.

Can these results be generalized to other societies? Some of the results may be particularly applicable to societies where women's consciousness of sexual issues has recently increased. In those societies, the special patterns that are likely to hold are that educated women with more permissive sexual values are likely to be more critical about their sexual experiences. Also, in some of these societies, learning new sexual practices primarily through pornography may provide for a troubled move towards more liberal sexual practices. However, on many dimensions, we expect that the Chinese results are similar to what will be found for other societies. Chinese urban prevalence levels for unwanted intercourse and forced intercourse are not all that different from other societies, suggesting that these phenomena are not that exceptional in China. In all societies, the results are consistent with an old emphasis on affective care being at least as important as learning new sexual techniques and that for women sexual rewards are multifaceted, requiring more than a single approach to resolution.

Appendix A. Determinants of Three Measures of Unwanted Sexual
Activity

                                                   Unwanted
                                                 Intercourse (a)

Variable                                         b            t

Relationship quality
  Any hitting                                   .17 *        2.22
  Intimacy and foreplay insufficient            .19 *        2.42
  Husband doesn't know wife's need              .12          1.68
  Genital caressing of wife absent             -.27 *       -2.53
Attitudes/Expectations
  Thinks sex is dirty                           .24 *        2.34
  Education (years)                             .05          1.42
  Permissive sex attitudes                      .16 *        2.45
Bargaining
  Wife's share of joint earnings
    0-24%                                       .16 *        2.14
    25-69% (reference)                           --
    70-100%                                     .54 **       2.88
  Parents' economic background
    Husband's better                            .15          1.06
    About equal (reference)                      --
    Wife's better                               .03           .21
  Other adults not in household                 .23 *        2.49
  Age gap (husband--wife) (years)             -0.02 *       -2.09
  Paid sex of men in locality (10%)            -.18         -1.36
Control variables
  No. of sexual partners in lifetime (log)      .17          1.02
  Alcohol consumption                           .11 *        2.00
  Child age 0-6                                 .28 **       2.88
  Psychological distress                        .55          1.79
  Sexual dysfunctions                           .41 **       3.29
  Age (10 years)                                .56          1.19

                                                     Unwanted
                                                    Act(s) (a)

Variable                                         b            T

Relationship quality
  Any hitting                                   .15          1.31
  Intimacy and foreplay insufficient            .08 *        2.06
  Husband doesn't know wife's need              .01           .10
  Genital caressing of wife absent             -.41 ***     -4.27
Attitudes/Expectations
  Thinks sex is dirty                           .12          1.24
  Education (years)                            -.01          -.42
  Permissive sex attitudes                      .19          1.91
Bargaining
  Wife's share of joint earnings
    0-24%                                       .19 *        2.17
    25-69% (reference)                           --
    70-100%                                     .58          1.91
  Parents' economic background
    Husband's better                            .19 *        2.13
    About equal (reference)                      --
    Wife's better                              -.02          -.39
  Other adults not in household                 .07           .93
  Age gap (husband--wife) (years)             -0.05 ***     -3.51
  Paid sex of men in locality (10%)             .06           .90
Control variables
  No. of sexual partners in lifetime (log)      .10           .68
  Alcohol consumption                           .13 *        2.39
  Child age 0-6                                 .20 *        1.97
  Psychological distress                        .45          1.46
  Sexual dysfunctions                           .28 ***      3.37
  Age (10 years)                                .65           .97

                                                  Sex to Satisfy
                                                   Partner (b)

Variable                                         b            t

Relationship quality
  Any hitting                                   .15          1.06
  Intimacy and foreplay insufficient            .05           .84
  Husband doesn't know wife's need              .15 **       3.33
  Genital caressing of wife absent             -.23         -1.76
Attitudes/Expectations
  Thinks sex is dirty                           .26 ***      4.22
  Education (years)                             .05 ***      5.33
  Permissive sex attitudes                      .19 ***      4.11
Bargaining
  Wife's share of joint earnings
    0-24%                                       .16          1.58
    25-69% (reference)                           --
    70-100%                                     .34 *        2.02
  Parents' economic background
    Husband's better                            .22 *        2.42
    About equal (reference)                      --
    Wife's better                              -.03          -.30
  Other adults not in household                 .11           .81
  Age gap (husband-wife) (years)               -.02         -1.34
  Paid sex of men in locality (10%)             .01          -.07
Control variables
  No. of sexual partners in lifetime (log)      .14           .90
  Alcohol consumption                           .19 **       3.17
  Child age 0-6                                 .09           .55
  Psychological distress                        .31          1.22
  Sexual dysfunctions                           .72          8.09
  Age (10 years)                                .17           .56

Coefficients are weighted and with standard errors adjusted for
sample design. N = 1,127 currently married, sexually active urban
women. Results are from the first equation of three simultaneous
equations models with one model for each unwanted sexual activity
measure. (a) probit regression coefficients, (b) ordered probit
regression coefficients.

* p < .05, ** p < .01, *** p < .001 (two--tailed test).

Appendix B. Type of Unwanted Sexual Activity and Other
Determinants of Psychological Distress

                                         Psychological
                                            Distres

Variable                                  b          t

Unwanted sex
  Unwanted intercourse                  .16 ***     4.70
  Unwanted act
  Sex to satisfy partner
Other correlates
  Physical health                      -.16 ***    -6.57
  Education (years)                    -.01        -1.74
  Joint earnings (log)                 -.03 *      -2.12
  Perceived care in daily life         -.02         -.75
  Desire to lose weight                 .15 ***     7.22
  Body mass index                      -.02 ***    -3.35
  Attractiveness (1-4)                 -.03        -1.64
  Living conditions hinder sex life     .05 ***     3.64
  Age (10 years)                        .15          .63
  Age squared                          -.04        -1.76

                                         Psychological
                                            Distress

Variable                                  b          T

Unwanted sex
  Unwanted intercourse
  Unwanted act                          .18 **      2.73
  Sex to satisfy partner
Other correlates
  Physical health                      -.16 ***    -8.94
  Education (years)                    -.01        -1.00
  Joint earnings (log)                 -.03        -1.34
  Perceived care in daily life         -.04        -1.34
  Desire to lose weight                 .15 ***    10.22
  Body mass index                      -.02 **     -2.85
  Attractiveness (1-4)                 -.03        -1.60
  Living conditions hinder sex life     .05 ***     3.75
  Age (10 years)                        .12          .43
  Age squared                          -.03        -1.70

                                         Psychological
                                            Distress

Variable                                  b          t

Unwanted sex
  Unwanted intercourse
  Unwanted act
  Sex to satisfy partner                .16 ***     5.69
Other correlates
  Physical health                      -.17 ***   -11.95
  Education (years)                    -.02 ***    -4.10
  Joint earnings (log)                 -.04        -1.80
  Perceived care in daily life         -.04        -1.12
  Desire to lose weight                 .16 ***     8.19
  Body mass index                      -.02 *      -2.52
  Attractiveness (1-4)                 -.03        -1.49
  Living conditions hinder sex life     .05 **      2.99
  Age (10 years)                        .22         1.08
  Age squared                          -.03        -1.43

Appendix B continued from Appendix A. Results are from the second
equation of three simultaneous equations models with one model for each
unwanted sexual activity measure.

* p < .05, ** p < .01, *** p < .001 (two-tailed test).

Primary funding support was provided by the National Institute of Child Health and Human Development grant ROI HD34157. Additional support was provided by the University of Chicago/ NICHD Population Research Center (P30 HD18288). A previous draft was presented at the 2005 Annual Meeting of the Population Association of America, Philadelphia, Pennsylvania, March 31-April 2, 2005.

References

Abbey, A., Ross, L. T., McDuffie, D., & McCauslan, P. (1996). Alcohol, misperception, and sexual assault: How and why are they linked? In D. M. Buss & N. M. Malamuth (Eds.), Sex, power, conflict: Evolutionary and feminist perspectives (pp. 138-161). New York: Oxford University Press.

Ageton, S. S. (1988). Vulnerability to sexual assault. In A. W. Burgess (Ed.), Rape and sexual assault (Vol. II, pp. 221-243). New York: Garland Publishing.

Basile, K. C. (1999). Rape by acquiescence: The ways in which women "give in" to unwanted sex with their husbands. Violence Against Women, 5, 1036-1058.

Basile, K. C. (2002). Prevalence of wife rape and other intimate partner sexual coercion in a nationally representative sample of women. Violence and Victims, 17, 511-524.

Basson, R. (2001). Using a different model for female sexual response to address women's problematic low sexual desire. Journal of Sex and Marital Therapy, 27, 359-403.

Becker, J., Skinner, L., Abel, G., & Cichon, J. (1986). Level of post-assault sexual functioning in rape and incest victims. Archives of Sexual Behavior, 15, 47-49.

Blythe, M. J., Fortenberry, J. D., Temkit, M. H., Tu, W., & Orr, D. P. (2006). Incidence and correlates of unwanted sex in relationships of middle and late adolescent women. Archives of Pediatrics & Adolescent Medicine, 160, 591-595.

Blumstein, P. & Schwartz, P. (1983). American couples. New York: Pocket Books.

Bullough, V. L. & Ruan, F. F. (1994). Marriage, divorce, and sexual relations in contemporary China. Journal of Comparative Family Studies, 25, 383-393.

Campbell, J. C. (2002). Health consequences of intimate partner violence. Lancet, 359, 1331-1336.

Coker, A. L., Davis, K. E., Arias, I., Desai, S., Sanderson, M., Brandt, H. M., & Smith, P. H. (2002). Physical and mental health effects of intimate partner violence for men and women. American Journal of Preventive Medicine, 23, 260-268.

DeLamater, J. D. & Sill, M. (2005). Sexual desire in later life. Journal of Sex Research, 42, 138-149.

DeMaris, A. (1997). Elevated sexual activity in violent marriages: Hypersexuality or sexuality. Journal of Sex Research, 34, 361-373.

Dobash, R. E. & Dobash, R. P. (1979). Violence against wives: A case against the patriarchy. New York: Free Press.

Evans, H. (1995). Defining difference: The "scientific" construction of sexuality and gender in the People's Republic of China. Signs. Journal of Women in Cultural and Society, 20, 357-394.

Everaerd, W., Laan, E., Both, S., & van der Velde, J. (2000). Female sexuality. In L. T. Szuchman & F. Muscarella (Eds.), Psychological perspectives on human sexuality (pp. 101-147). New York: John Wiley.

Evertsson, M. & Nermo, M. (2005). Dependence within families and the division of labor: Comparing Sweden and the United States. Journal of Marriage and the Family, 66, 1272-1286.

Feldman-Summers, S., Gordon, P. E., & Megher, J. R. (1979). The impact of rape on sexual satisfaction. Journal of Early Adolescence, 3, 363-385.

Finkelhor, D. & Yllo, K. (1985). License to rape: Sexual abuse of wives. New York: Holt Rinehart and Winston.

Fox, G. L., Benson, M. L., DeMaris, A. A., & Van Wyk, J. (2002). Economic distress and intimate violence: Testing family stress and resources theories. Journal of Marriage and the Family, 64, 793-807.

Gidycz, C. A. & Koss, M. P. (1991). The effects of acquaintance rape on the female victim. In A. Parrot & L. Bechhofer (Eds.), Acquaintance rape. The hidden crime (pp. 270-283). New York: Wiley.

Greenstein, T. N. (2000). Economic dependence, gender, and the division of labor in the home: A replication and extension. Journal of Marriage and the Family, 62, 322-335.

Hall, G. C. N., Sue, S., Narang, D. S., & Lilly, R. S. (2000). Culture-specific models of men's sexual aggression: Intra- and interpersonal determinants. Cultural Diversity & Ethnic Minority Psychology, 6, 252-267.

Higgins, L. T., Zheng, M., Liu, Y., & Sun, C. H. (2002). Attitudes to marriage and sexual behaviors: A survey of gender and culture differences in China and United Kingdom. Sex Roles, 46, 75-89.

Hochschild, A. (1989). The second shift. New York: Avon Books.

Im-em, W., Archvanitkul, K., & Kanchanachitra, C. (2004). Sexual coercion among women in Thailand. Results from the WHO multi-country study on women's health and life experiences. Paper presented at the Population of America Association, Boston, MA, August 3-5, 2004.

Impett, E. A. & Peplau, L. A. (2002). Why some women consent to unwanted sex with a dating partner: Insights from attachment theory. Psychology of Women Quarterly, 26, 360-370.

Impett, E. A. & Peplau, L. A. (2003). Sexual compliance: Gender, motivational, and relationship perspectives. Journal of Sex Research, 40, 87-100.

Inkeles, A. (1973). The school as a context for modernization. International Journal of Comparative Sociology, 14, 163-179.

Kalof, L. (2000). Vulnerability to sexual coercion among college women: A longitudinal study. Gender Issues, 18, 47-58.

Kaplan, H. S. (1974). The new sex therapy: Active treatment of sexual dysfunctions. New York: Brunner/Mazel.

Knodel, J., Saengtienchai, C., VanLandingham, M., & Lucas, R. (1999). Sexuality, sexual experience, and the good spouse: Views of married Thai men and women. In P. A. Jackson & N. M. Cook, (Eds.), Genders and Sexualities in Modern Thailand (pp. 93-113). Bangkok: Silkworm Books.

Koss, M. P. (1985). The hidden rape victim: Personality, attitudinal, and situational characteristics. Psychology of Women Quarterly, 9, 193-212.

Koss, M. P. & Dinero, T. E. (1989). Discriminant-analysis of risk-factors for sexual victimization among a national sample of college-women. Journal of Consulting and Clinical Psychology, 57, 242-250.

Krug, E. G., Dahlberg, L. L., Mercy, J. A., Zwi, A. B., & Lozano, R., eds. (2002). World report on violence and health. Geneva: World Health Organization.

Leung, K., Lue, B., Lee, M., & Tang, L. (1998). Screening of depression in patients with chronic medical diseases in a primary care setting. Family Practice, 15, 67-75.

Levy, P. S. & Lemeshow, S. (1999). Sampling of populations: Methods and applications (3rd ed.). New York: Wiley.

Liu, D., Ng, M., Zhou, L., & Haeberle, E. J. (1997). Sexual behavior in modern China: Report on the nationwide survey of 20,000 men and women. New York: The Continuum Publishing Company.

Lottes, I. L. & Weinberg, M. S. (1997). Sexual coercion among university students: A comparison of the United States and Sweden. Journal of Sex Research, 34, 67-76.

Luo, Y., Parish, W. L., & Laumann, E. O. (2005). A population-based study of body image concerns among urban Chinese adults. Body Image, 2(4), 333-345.

Malamuth, N. M. (2003). Criminal and noncriminal sexual aggressors: Integrating psychopathy in a hierarchical-mediational confluence model. Annals of New York Academy of Sciences, 989, 33-58.

Mandoki, C. A. & Burkhart, B. R. (1991). Women as victims: Antecedents and consequences of acquaintance rape. In A. Parrot & L. Bechhofer (Eds.), Acquaintance rape: The hidden crime (pp. 176-191). New York: Wiley.

Masters, W. H., Johnson, V. E., & Kolodny, R. C. (1995). Human sexuality. New York: Harper Collins College.

Muehlenhard, C. L. & Linton, M. A. (1987). Date rape and sexual aggression in dating situations: Incidence and risk-factors. Journal of Counseling Psychology, 34, 186-196.

Murnen, S. K. & Byrne, D. (1991). Hyperfemininity: Measurement and initial validation of the construct. Journal of Sex Research, 28, 479-489.

Muthen, L. K. & Muthen, B. O. (2004). Mplus user's guide. Los Angeles, CA: Muthen and Muthen.

O'Sullivan, L. F. & Allgeier, E. R. (1998). Feigning sexual desire: Consenting to unwanted sexual activity in heterosexual dating. Journal of Sex Research, 35, 234-243.

Oswald, D. L. & Russell, B. L. (2006). Perceptions of sexual coercion in heterosexual dating relationships. Journal of Sex Research, 43, 87-95.

Pan, S. (1993). A sex revolution in current China. Journal of Psychology and Human Sexuality, 6, 1-14.

Pimentel, E. E. (2000). Just how do I love thee? Marital relations in urban China. Journal of Marriage and the Family, 62, 32-47.

Renaud, C., Byers, E. S., & Pan, S. (1997). Sexual and relationship satisfaction in mainland China. Journal of Sex Research, 34, 399-415.

Rivers, K., Aggleton, P., Elizondo, J., Hernandez, G., Herrera, G., Mane, P., Niang, C. I., Scott, S., & Setiadi, B. (1998). Gender relations, sexual communication and the female condom. Critical Public Health, 8, 273-290.

Ruan, F. F. (1991). Sex in China. New York: Plenum.

Russell, D. E. H. (1982). Rape in marriage. New York: Macmillan.

Schwartz, P. (1994). Peer marriage: How love between equals really works. New York: Free Press.

Schwartz, P. (2000). Creating sexual pleasure and sexual justice in the twenty-first century. Contemporary Sociology: A Journal of Reviews, 291, 213-219.

Schwartz, P. & Rutter, V. (1998). The gender of sexuality. Thousand Oaks, CA: Pine Forge.

Smuts, B. 1995. The evolutionary origins of patriarchy. Human Nature, 6, 1-32.

Shields, W. M. & Shields, L. M. (1983). Forcible rape: An evolutionary perspective. Ethology and Sociobiology, 4, 115-136.

Skinner, C., Holt, D., & Smith, T. (1989). Analysis of Complex Surveys. New York: Wiley.

Smith, P. H., Thornton, G. E., DeVellis, R., Earp, J., & Coker, A. L. (2002). A population-based study of the prevalence and distinctiveness of battering, physical assault, and sexual assault in intimate relationships. Violence Against Women, 8, 1208-1232.

Sprecher, S., Hatfield, E., Cortese, A., Potapova, E., & Levitskaya, A. (1994). Token resistance to sexual intercourse and consent to unwanted sexual intercourse: College students' dating experiences in three countries. Journal of Sex Research, 31, 125-132.

Tang, W.-F. & Parish, W. L. (2000). Chinese urban life under reform: The changing social contract. New York: Cambridge University Press.

Testa, M., Vanzile-Tamsen, C., & Livingston, J. A. (2004). The role of victim and perpetrator intoxication on sexual assault outcomes. Journal of Studies on Alcohol, 65, 320-329.

Trudel, G., Turgeon, L., & Piche, L. (2000). Marital and sexual aspects of old age. Sexual and Relationship Therapy, 154, 381-406.

Wolff, B., Blanc, A. K., & Gage, A. J. (2000). Who decides? Women's status and negotiation of sex in Uganda. Culture, Health & Sexuality, 2, 303-322.

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

Zung, W. W. K. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63-70.

William L. Parish

University of Chicago and National Opinion Research Center

Ye Luo

University of North Florida and National Opinion Research Center

Edward O. Laumann, Melissa Kew, and Zhiyuan Yu

University of Chicago and National Opinion Research Center

Correspondence should be addressed to William L. Parish, Department of Sociology, University of Chicago, 1126 E. 59th Street, Chicago, IL 60637. E-mail: [e-mail]

Table 1. Descriptive Statistics (means and ranges)

Variable                                    Weighted M   Range

Unwanted sex measures
  Unwanted intercourse (a)                     .21        0-1
  Unwanted act (a)                             .22        0-1
  Sex to satisfy partner
    Never (a)                                  .28        0-1
    Rarely (a)                                 .26        0-1
    Sometimes (a)                              .40        0-1
    Often (a)                                  .06        0-1
  Unwanted sexual activity index               1.16       0-3
Psychological distress
  Psychological distress index                 1.48       1-3
Relationship quality
  Any hitting (a)                              .28        0-1
  Daily intimacy absent                        2.30       1-4
  Foreplay insufficient                        1.61       1-3
  Husband doesn't know wife's need             1.94       1-4
  Genital caressing of wife absent'            .19        0-1
Attitudes/ Expectations
  Thinks sex is dirty                          1.63       1-3
  Education (years)                            9.01      0-16
  Permissive sex attitudes                     1.88       1-4
Bargaining
  Wife's share of joint earnings
    0-24% (a)                                  .25        0-1
    25-69% (a)                                 .70        0-1
    70-100% (a)                                .04        0-1
  Parents' economic background
    Husband's better (a)                       .20        0-1
    About equal (a)                            .43        0-1
    Wife's better (a)                          .37        0-1
  Other adults not in household (a)            .61        0-1
  Age gap (husband-wife) (years)               2.12      -4-11
  Paid sex of men in locality (10%)            .99       0-2.6
Control variables
  No. of sexual partners in lifetime (log)     .75        0-4
  Alcohol consumption                          .35        0-1
  Child age 0-6 (a)                            .23        0-1
  Sexual dysfunction(s)                        1.75       1-3
  Age (10 years)                               3.97      2-6.4
Psychological distress correlates
  Physical health                              3.72       1-5
  Joint earnings (log)                         7.07      0-11
  Perceived care in relationship               3.52       1-4
  Desire to lose weight (a)                    .30        0-1
  Body mass index                             22.92      16-33
  Attractiveness (self-rated)                  2.35       1-4
  Living conditions hinder sex life            1.36       1-3

Note. N = 1,127 currently married, sexually active urban women.

(a) Dummy variable, with 1 signifying the presence of the
characteristic and 0, its absence.

Table 2a. Determinants of Unwanted Sexual Activity

                                                 Model I

Variable                                       b          t

Relationship quality
  Any hitting                                 .41 **      3.25
  Intimacy and foreplay insufficient
    Daily intimacy absent                     .10 *       2.06
    Foreplay insufficient                     .12 **      3.20
  Husband doesn't know wife's need            .10 *       2.65
  Genital caressing of wife absent           -.45 ***    -6.72
Attitudes/ Expectations
  Thinks sex is dirty                         .26 ***     4.60
  Education (years)                           .04 ***     4.01
  Permissive sex attitudes                    .25 **      3.04
Bargaining
  Wife's share of joint earnings
    0-24%                                     .03          .37
    25-69% (reference)                        --
    70-100%                                   .33         1.59
  Parents' economic background
    Husband's better                          .21 **      2.90
    About equal (reference)                   --
    Wife's better                             .04          .54
  Other adults not in household               .18         1.57
  Age gap (husband - wife) (years)           -.03 ***    -3.72
  Paid sex of men in locality (10%)          -.17 *      -2.23
Control variables
  No. of sexual partners in lifetime (log)    .44 ***     3.91
  Alcohol consumption                         .29 **      2.52
  Child age 0-6                               .16         1.31
  Psychological distress                      .43 ***     4.57
  Sexual dysfunctions                         .70 ***     6.64
  Age (10 years)                             -.05        -1.64

                                                Model II

Variable                                       b          t

Relationship quality
  Any hitting                                 .22 *       2.44
  Intimacy and foreplay insufficient          .15 *       2.71
    Daily intimacy absent
    Foreplay insufficient
  Husband doesn't know wife's need            .11 ***     5.08
  Genital caressing of wife absent           -.48 ***    -4.87
Attitudes/ Expectations
  Thinks sex is dirty                         .21 ***     4.75
  Education (years)                           .03 ***     4.07
  Permissive sex attitudes                    .13         1.84
Bargaining
  Wife's share of joint earnings
    0-24%                                     .10         1.64
    25-69% (reference)                         -
    70-100%                                   .21         1.23
  Parents' economic background
    Husband's better                          .15         1.87
    About equal (reference)                   --
    Wife's better                             .00          .10
  Other adults not in household               .14         1.55
  Age gap (husband - wife) (years)           -.03 ***    -3.64
  Paid sex of men in locality (10%)          -.10        -1.98
Control variables
  No. of sexual partners in lifetime (log)    .33 **      3.15
  Alcohol consumption                         .11         1.68
  Child age 0-6                               .21 **      2.82
  Psychological distress                       --
  Sexual dysfunctions                          --
  Age (10 years)                              .06         2.01

                                                Model IIIa

Variable                                       b          t

Relationship quality
  Any hitting                                 .14         1.23
  Intimacy and foreplay insufficient          .10 *       2.48
    Daily intimacy absent
    Foreplay insufficient
  Husband doesn't know wife's need            .07 ***     4.68
  Genital caressing of wife absent           -.35 ***    -7.09
Attitudes/ Expectations
  Thinks sex is dirty                         .15 ***     3.31
  Education (years)                           .03 ***     3.73
  Permissive sex attitudes                    .12         1.66
Bargaining
  Wife's share of joint earnings
    0-24%                                     .09 *       2.09
    25-69% (reference)                        --
    70-100%                                   .31 *       2.22
  Parents' economic background
    Husband's better                          .14 ***     4.06
    About equal (reference)                    --
    Wife's better                             .00         -.06
  Other adults not in household               .14 *       2.00
  Age gap (husband - wife) (years)           -.02 ***    -3.79
  Paid sex of men in locality (10%)          -.04         -.96
Control variables
  No. of sexual partners in lifetime (log)    .14         1.70
  Alcohol consumption                         .12 ***     3.60
  Child age 0-6                               .17 *       2.41
  Psychological distress                      .42 **      2.75
  Sexual dysfunctions                         .37 ***     6.87
  Age (10 years)                              .03         1.06

Note. Regression coefficients, weighted and with standard errors
adjusted for sample design. N= 1,127 currently married, sexually
active urban women. Model I shows results from 19 separate
regressions, each including only age and a single covariate. Model
11 shows results from an ordinary least squares regression with all
variables except psychological distress and sexual dysfunction. Model
IIIa is the first equation of a simultaneous equations model that
includes all variables and the possible endogenous relationship of
psychological distress and unwanted sex (see Table 2b for the second
half of this model). With correlated error terms of equations IIIa
and IIIb, fit statistics for model III are XZ = 16.33, df = 23,
p = .84; CFI = 1.00, RMSEA <.001.
* p < .05, ** p < .01, *** p < .001 (two-tailed test).

Table 2b. Unwanted Sexual Activity and Other Determinants of
Psychological Distress

                                           Model I

Variable                                 b          t

Unwanted sex
  Unwanted sexual activity index       .10 ***      5.60
Other correlates
  Physical health                     -.20 ***     -8.44
  Education (years)                    .00          -.44
  Joint earnings (log)                -.06 **      -2.93
  Perceived care in daily life        -.13 **      -2.59
  Desire to lose weight                .11 ***      7.80
  Body mass index                     -.01         -1.24
  Attractiveness (1-4)                -.07 ***     -3.69
  Living conditions hinder sex life    .13 ***      5.35
  Age (10 years)                       .39          1.84
  Age squared                         -.05         -1.78

                                         Model II

Variable                                 b          t

Unwanted sex
  Unwanted sexual activity index       .06 **       2.96
Other correlates
  Physical health                     -.17 ***     -6.24
  Education (years)                   -.01 *       -2.44
  Joint earnings (log)                -.03         -1.77
  Perceived care in daily life        -.06         -1.42
  Desire to lose weight                .16 ***      9.68
  Body mass index                     -.02 **      -3.14
  Attractiveness (1-4)                -.02          -.97
  Living conditions hinder sex life    .05 *        2.56
  Age (10 years)                       .32          1.58
  Age squared                         -.04         -1.56

                                         Model IIIb

Variable                                 b          t

Unwanted sex
  Unwanted sexual activity index       .20 ***      5.48
Other correlates
  Physical health                     -.16 ***     -6.32
  Education (years)                   -.01 ***     -3.71
  Joint earnings (log)                -.03 *       -2.05
  Perceived care in daily life        -.03          -.84
  Desire to lose weight                .15 ***     10.87
  Body mass index                     -.02 **      -2.89
  Attractiveness (1-4)                -.03         -1.61
  Living conditions hinder sex life    .05 **       3.01
  Age (10 years)                       .28          1.72
  Age squared                         -.03         -1.67

Note. Table continued from Table 2a. In model IIIb "unwanted
sexual activity" incorporates the possible endogenous relationship
of psychological distress and unwanted sexual activity.

* p < .05, **p < .01, *** p < .001 (two-tailed test).