A study of sexual satisfaction and frequency of sex among Hong Kong Chinese couples.

A happy marriage often is associated with satisfying sexual activity (Greeley, 1991; Laumann, Gagnon, Michael, & Michaels, 1994). Prior studies on marital sexuality consistently have found that frequency with which a married couple engages in sex with each other is related to marital sexual satisfaction (e.g., Call, Sprecher, & Schwartz, 1995; Haavio-Mannila & Kontula, 1997). Attempting to disentangle the causal relationship between frequency of sex and marital sexual satisfaction however, could prove to be problematic. The couples could be more sexually satisfied simply because they had sex more frequently or they also could tell us that the couples who had sex less frequently did so because it tended to be somehow less pleasurable and satisfying for them (Harvey, Wenzel, & Sprecher, 2004).

Hence, although frequency of sex and sexual satisfaction are closely related, one cannot automatically assume that the two are the same. Frequency of engagement in sexual activity is a physical or factual measure of behavior, while marital sexual satisfaction is a psychological or subjective measure of perception. For these reasons, in the present study we considered sexual satisfaction and frequency of sex to be two correlated, but distinct, variables.

There have been quite a number of studies in Western societies about the causes and consequences of sexual satisfaction in marriage. Little is known about the predictors of marital sexual relationships in Chinese societies (cf. So & Cheung, 2005). By using a large community data set from the Family Planning Association of Hong Kong (FPAHK, 2003) in 2002, we set our research objective to investigate whether the predictors found in Western societies also are applicable in a Chinese society. The data set contains 1,124 pairs of couples (as opposed to unrelated men and women in most studies). The present study addresses whether the predictors of sexual satisfaction are the same for wives and husbands. This gives us a better understanding of what predicts sexual satisfaction within marriage.

In the following sections, we first review factors associated with sexual satisfaction and the frequency of sex in marriages. We then propose a structural equation model (SEM) to test the research hypotheses. The methods and data used in the present study then are introduced. A discussion of the findings and their implications will be addressed.

Factors Associated With Sexual Satisfaction and Frequency of Sex

In this session, we review the factors associated with marital sexual satisfaction or frequency of sex or both. The variables can be grouped as sociodemographic variables, family variables, medical history related to sex, and the psychological variables. Sexual satisfaction usually has been found to be related to the physical aspects of sexuality such as frequency of sexual behavior and the timing of orgasms (Call et al., 1995; Donnelly, 1993; Marsiglio & Donnelly, 1991); to psychological factors involving the couple such as a woman's perspective of how close she is to her partner (Byers & Demmons, 1999; Young, Denny, Luquis, & Young, 1998); to socioeconomic factors such as socioeconomic status and family income (Perrone & Worthington, 2001); and to demographic factors such as age at marriage, educational level, number of children, religiosity, gender, and ethnicity (Call et al., 1995; Whyte, 1990).

Sociodemographic variables. In Western societies, age frequently has been found to be negatively correlated with sexual satisfaction (e.g., Chevret, Jaudinot, Sullivan, Marrel, & De Gendre, 2004a; 2004b; Ojanlatva, Helenius, Rautava, Ahvenainen, & Koskenvuo, 2003; Nicolosi, Moreira, Villa, & Glasser, 2004) and frequency of sex (Doddridge, Schumm, & Berger, 1987; Kahn & Udry, 1986; Marsiglio & Donnelly, 1991; Mazur, Mueller, Krause, Booth, & 2002; Trussell & Westoff, 1980). Thus, married couples tend to have less sex and to feel less sexually satisfied as they get older. Similar findings have been reported also in Chinese societies (Guo & Huang, 2005; Renaud, Byers, & Pan, 1997). That said, Christopher and Sprecher (2000) have contended that sexual satisfaction might change with age but does not decline as rapidly as does the frequency of sex. Gender also plays a role in the effect of age on sexual satisfaction and frequency of sex. Laumann et al. (1994) reported that women over the age of 40 are less sexually satisfied than women under 40; however, men did not experience the same drop in sexual satisfaction with age.

Couples with a higher educational level generally tend to be more sexually satisfied (Guo & Huang, 2005; Laumann et al., 1994; Liu, 2003), and there is a positive correlation between educational level and frequency of sex (Call, Sprecher, & Schwartz, 1995). Ojanlatva et al. (2003) found, however, that participants who have less education are more satisfied sexually.

In addition, it is generally noted that working couples have less time to engage in intimate contact (Hyde, DeLamater, Plant, & Byrd, 1996). For example, Olson and DeFrain (1994) found that married couples in which both partners are holding full-time jobs engage in sex less frequently than those in which one partner does not work full time outside the home. In the longitudinal study of Hyde et al. (1996) involving 500 women and their husbands, however, no significant differences were found between homemakers and women employed part time, full time, and high full time (i.e., working overtime) on measures of frequency of sex, sexual satisfaction, and decreased sexual desire. Instead, fatigue was associated with sexual outcomes by both employed women and homemakers. Specifically, Hyde et al. found that role quality, especially job role quality, was related to sexual satisfaction and frequency of sex. Similar findings have been reported also in other studies (e.g., Henderson-King & Veroff, 1994; Oggins, Leber, & Veroff, 1993). Based on the assumption that longer working hours lead to higher incomes, these researchers found that higher household incomes were associated with less sexual satisfaction for women. Furthermore, remarriage was found to be correlated with a higher frequency of sex (Call, Sprecher, & Schwartz, 1995).

Family variables. Length of marriage has been shown to be negatively correlated with sexual satisfaction (Liu, 2003) and frequency of sex (Liu, 2000; Stafford, Kline, & Rankin, 2004; Udry, 1980, 1993). As length of marriage is highly correlated with age, however, it is difficult to disentangle the effect of age from that of length of marriage. Guo et al. (1999) found similar patterns among the Chinese, while Zhou (1993) found the same correlation for Chinese wives, but not for husbands.

Regarding the association between number of children and marital sexual satisfaction, studies consistently have indicated that couples with children engage in sex less frequently than those without children (Blumstein & Schwartz, 1983; Greenblat, 1983; James, 1981). O'Fallon, Tsui, and Adewuyi (2003) found that couples without children or with one child engage in sex more frequently than couples with five or more children. Call, Sprecher, and Schwartz (1995) found that couples with very young children (aged 0--4) engage in sex less frequently than those with older children (aged 5-18). Donnelly (1993) found a positive correlation, however, between number of children and frequency of sex.

Sexual problems. Having a medical history of sexual problems affects sexual satisfaction (Dunn, Croft, & Hackett, 2000; MacNeil & Byers, 1997; Woody & D'Souza, 1997). Chevret et al. (2004a, 2004b) found that female partners of men with erectile dysfunction had significantly lower levels of sexual satisfaction than those whose partners did not suffer from the condition. Byers and Grenier (2003) reported similar findings for men with rapid ejaculation.

Interest in sex. Not surprisingly, interest in sex is associated with sexual satisfaction and sexual frequency. Couples who are interested in sex tend to engage in sex more frequently than those who are not (Greeley, 1991). Call et al. (1995) suggested that the decline in interest in sex as the length of a marriage increases is due to a habituation effect, resulting from the increased accessibility of a sexual partner and the predictability in sexual behavior with that partner over time.

Compared with knowledge about sexual satisfaction and frequencies of sex in Western societies, little is known about sexual satisfaction among Chinese couples. The aim of the present study was to fill this knowledge gap by studying marital sexual satisfaction and frequency of sex using a representative sample of 1,124 couples in Hong Kong. Our research questions were as follows: (1) "What are the factors that influence sexual satisfaction for married Hong Kong Chinese couples?," (2) "What are the factors that influence frequency of sex for married Hong Kong Chinese couples?," and (3) "What are similarities and differences in the factors associated with marital sexual satisfaction and frequency of sex among married Hong Kong Chinese couples?"

It should be noted that we did not include marital satisfaction as an independent variable in our study because the strong association between marital satisfaction and marital sexual satisfaction has been well demonstrated (see Christopher & Sprecher, 2000, for a comprehensive review) and because the collinearity between marital and sexual satisfaction would have obscured the relationships among the individual correlates, frequency of sex, and sexual satisfaction.

Method

Data

This study was part of a larger study, the eighth knowledge, attitude and practice (KAP) survey conducted by the FPAHK in 2002. The KAP survey is the longest-running community-based survey on family planning in Hong Kong. The FPAHK carried out the first KAP survey in 1967; since then, the survey has been conducted every 5 years. The KAP surveys contain rich data on current patterns in family planning and related issues. Marriage and sexual satisfaction are two key elements in the 2002 study.

Using a stratified sampling technique, participants were selected randomly across the territory so as to ensure the representativeness of the data. In the 2002 study, about 1,600 married or cohabiting women aged from 15 to 49 were successfully interviewed. In addition, the spouses of the female participants also were invited to participate in the survey. The result was 1,147 successful interviews with spouses. Since the focus of this study was on married couples, 23 pairs of cohabiting couples were excluded from the analysis. Thus, the total number of pairs of couples was 1,124.

For most variables, less than 5% of the data was missing. There were three variables for which a higher percentage of the data was missing: full-time work of wives (7.7%), full-time work of husbands (9.8%), and frequency of sex that month reported by wives (15%). The full information maximum likelihood method implemented in LISREL 8.72 (Joreskog & Sorbom, 2005) was used to handle the missing data (see Acock, 2005; Cheung, 2007). Descriptive statistics on the social demographic background of the participants are shown in Table 1.

Methods of Handling Couples' Data

It has been widely acknowledged that data collected from a single family member, say a husband or wife, are not optimal for family or marriage research (Szinovacz, 1983), especially in sensitive issues such as sexuality (e.g., Szinovacz & Egley, 1995). Data on couples is preferable, but specific issues on the handling of nonindependent data are involved. Common statistical methods such as multiple regressions, which ignore the dependence in the data between husbands and wives, are likely to result in seriously inflated Type I errors (Raudenbush & Bryk, 2002).

Structural equation modeling (Joreskog & Sorbom, 1996) is a better approach to analyzing data involving couples. There have been several studies showing how SEM can be used to analyze such data (e.g., Newsom, 2002; Wendorf, 2002). The main advantage of using SEM in this study is that predictors for sexual satisfaction and the frequency of sex can be tested simultaneously. This provides a way of investigating the similarities and differences in the predictors for sexual satisfaction and the frequency of sex. Moreover, SEM provides effective strategies to handle missing data, which can be a serious problem in family and marriage research (see Acock, 2005).

Measurement

Because we measured sexual satisfaction for both husbands and wives, together with frequency of sex (which should be the same for each couple), there were three dependent variables in the analysis, namely, the sexual satisfaction of husbands, the sexual satisfaction of wives, and the frequency with which the couple has sex. The sexual satisfaction of the husbands and that of the wives was measured by a single-item question, "Are you satisfied with your sexual life?" Responses ranged from "very unsatisfied" (1) to "very satisfied" (5). The frequency of sex in the month of the interview was measured by asking the wives, "In the past 30 days, how many times have you had sex?" This represents the frequency of sex for the couple in that month. Most independent variables were measured directly because they were related to demographics. Interest in sex, however, was gauged by asking the question, "Are you interested in your sexual life?" and measuring the responses with a 5-point Likert scale. Moreover, because this question is considered sensitive in Chinese societies, the respondents were told to mark their response to this question privately to avoid replying in a biased manner. The explanatory variables can be grouped into several categories.

(1) Demographic variables: Age, place of birth (Hong Kong SAR vs Mainland China), educational level, whether or not one is engaged in full-time work, and whether or not one has been married more than once. These variables were measured separately for the husbands and wives. Thus, there were a total of two sets of these demographic variables.

(2) Family variables: Family income, number of children, and years of marriage. These variables were assumed to be the same for the couples.

(3) Medical history related to sex: Whether or not one previously has sought medical help related to sex (husbands and wives) and whether or not one has had an abortion (wives only). There was only one or two studies addressing the impact of abortion history on sexuality in Chinese populations. Fok, Siu, and Lau (2005) found that women who attended a postabortion clinic reported decreased sexual interest, sexual satisfaction, and frequency of sex after the abortion. The present study explores a possible association among these variables by using a large community sample.

(4) Psychological variable: Interest in sex. This was measured separately for husbands and wives by a 5-point Likert scale.

Proposed Model for Testing

We proposed a path model for testing that is grounded in the existing literature. Figure 1 shows the model (the differences between the solid and dashed lines should be ignored for the time being). Measures of the demographic variables, family variables, variables on medical history, and the psychological variable of the husbands and wives are the potential predictors of the sexual satisfaction of the husbands and wives. The third dependent variable is the frequency with which the couples engage in sex. This variable was regressed on all of the independent variables.

LISREL 8.72 (Joreskog & Sorbom, 2005) with the maximum likelihood method as the estimation method was used for the path analysis. To assess whether or not the proposed model fits the data, the guidelines proposed by Hu and Bentler (1999) were used. They suggested that a good fitted model should have goodness-of-fit indices, for instance, a normed fit index (NFI) and a comparative fit index (CFI) larger than 0.95 or residual-based indices, such as a standardized root mean square residual (RMSEA) smaller than 0.06 or a standardized rootmean-square residual (SRMR) smaller than 0.08. Since a large sample was used (N = 1,124) and many path coefficients were compared, we used [alpha] = .01 in the following analysis to control for the familywise Type I error.

[FIGURE 1 OMITTED]

Results

Descriptive Statistics

The sample statistics and the correlations of the variables are shown in Table 2. Several observations are noteworthy. First, the correlation between the sexual satisfaction reported by the husbands and wives was moderate (r = .32, p < .001). This indicates that married couples may not necessarily have the same level of sexual satisfaction. These couples tended to be satisfied with their marital sexual relationship, however with the husbands (mean = 3.84 out of a maximum of 5) being slightly more satisfied than the wives (mean = 3.69) with a t(1,060) = 6.34, p < .001.

Second, the correlation between the frequency of sex during that month and sexual satisfaction was stronger for the wives (r = .38, p < .001) than for the husbands (r = .25, p < .001). The difference between them was statistically significant at p < .001 according to the method suggested by Cheung and Chan (2004). One possible explanation for this is that the question on the frequency of sex in that month was asked of the wives. Thus, it is not surprising that the association is stronger for the variables related to wives.

Table 3 shows the frequency of sex broken down by sex and by age groups. The general observation is that the frequency of sex dropped when age increased. The gamma statistics, which indicate the association between two ordinal variables, for the wives and the husbands were, respectively, -0.179, p < .001 and -0.209, p < .001. These provide empirical support to the observation that the frequency of sex drops when age increases.

Path Model

The goodness-of-fit indices for the proposed path model were [chi square](14, N= 1,124) = 73.4, p <.001, NFI = 0.99, CFI = 0.99, RMSEA = 0.06, SRMR = 0.02. They indicate that the model fitted the data very well. The [R.sup.2] for the frequency of sex in that month and the sexual satisfaction of the wives and husbands were. 16, .27, and. 19, respectively. The significant and nonsignificant paths ([alpha] = .01) are denoted with the solid and dashed lines, respectively in Figure 1. In order to better compare the estimates, the standardized path coefficients are shown in Table 4.

The results show that the strongest predictor of the sexual satisfaction of husbands and wives and of the frequency of sex was interest in sex. Indeed, interest in sex of the husbands and of wives was the sole significant predictor of sexual satisfaction and frequency of sex. On the other hand, there were differences between the husbands and wives. For the husbands, age was negatively correlated with frequency of sex, but not with sexual satisfaction. This indicates that the frequency of sex decreased when husbands became older, but their sexual satisfaction did not necessarily decline. It should be noted that the same effect was not observed in the wives. Wives with a full-time job had sex less frequently than those without a full-time job, while this effect was not observed in the husbands. Other variables were not related to sexual satisfaction or frequency of sex. These were place of birth, educational level, whether or not one had been married more than once, whether or not one had obtained medical help on sex-related matters, whether or not a woman had a history of abortion, number of children, years of marriage, and family income.

Discussion

Most previous studies on marital sexual satisfaction were conducted in Western societies. Our study attempted to identify predictors of sexual satisfaction and frequency of sex within marriage in Hong Kong. We found that Hong Kong Chinese couples generally are satisfied with their sexual life as a whole, and that husbands are slightly more satisfied than wives. Frequency of sex among couples, however was found to be relatively low. Among the predicting variables, interest in sex was the only variable significantly associated with both marital sexual satisfaction and frequency of sex. Moreover, we also have found some consistent and inconsistent findings compared with previous studies in Western and Eastern societies.

Consistent with past findings (e.g., Greeley, 1991), frequency of sex and sexual satisfaction among both wives and husbands are positively correlated, but the correlation we observed is not as strong as might be expected. The finding that the frequency of sex among Hong Kong married couples is low, although they reported being satisfied with their marital sexual relationship, is interesting. This discrepancy may suggest that, although most couples are satisfied with their marital sexual relationship, they do not have enough time or energy to engage in more sexual activity with their partners. This would not be surprising given the busy and hectic lifestyle in Hong Kong.

Age plays a negative role in the frequency with which husbands engage in sex, but age is not an important factor in predicting frequency of sex and sexual satisfaction for wives. This is consistent with most empirical findings (e.g., Chevret et al., 2004a, 2004b; Renaud, Byers, & Pan, 1997) showing that frequency of sex drops over age for men.

In Western societies, working couples with full-time jobs are less likely to engage in sexual activities (e.g., Hyde et al., 1996; Olson & DeFrain, 1994). One interesting culturally specific finding in this study is that wives with a full-time job have less sex than wives without a full-time job, but this effect was not observed with the husbands. The possible explanation for this may lie in the fact that, in Hong Kong, wives still typically are expected to be responsible for most household chores as well as for most of the parenting duties. Even if they are engaged in fulltime employment, their involvement in parenting and domestic duties remains high (Chu, 1997).

Chu (1997) has stated also that the stereotypical perception is that the career world is still ultimately a man's place, whereas the home is the place where women should concentrate their efforts. Wives may feel that they are fully responsible for the parenting of their children and for the domestic work, leading to high levels of stress. This, in turn, may lead to a lower frequency of sex. On the other hand, the age of husbands is negatively related to the frequency with which they engage in sex; however, this is not the case for wives at all.

Interest in sex is the sole variable that predicts both marital sexual satisfaction and frequency of sex. This finding demonstrates that sexual satisfaction is more than just a biological reflex. It also suggests that mutuality in initiating sex seems to contribute to the sexual satisfaction of both partners (Blumstein & Schwartz, 1983), as those who are interested in sex are more likely to initiate sex, contributing to the sexual satisfaction of both the individual and his or her partner.

Although only a handful of marital sexual studies have been conducted in Chinese societies, there is one very large scale community-based study conducted in China by Liu and his colleagues (1997). They conducted the community-based study from February 1989 to April 1990 and sent out 21,500 questionnaires to secondary students, university students, married couples, and sexual offenders. Out of those, 7,602 were married adults, and 6,210 (81.7%) lived in 15 urban areas. They found that the frequency of sex per month among the participants was higher than in our samples (M=4.66, SD=3.72, M = 2.04, SD = 1.14, respectively). They also found that women are more sexually satisfied than men; on contrast, our study found that husbands are more satisfied than wives. It should be noted, however, that Liu et al. (1997) used unrelated men and women in their study while couples from the same family were used in our study.

Regarding the gender difference of sexual satisfaction, we found that husbands are more sexually satisfied than wives. Zhou (1993) found similar patterns in a survey. It is speculated that more emphasis is still placed on the sexual pleasure of Chinese men than that of Chinese women (Evans, 1995; Pan, 1993; cf. Renaud, Byers, and Pan, 1997, who found that women reported higher sexual satisfaction than men).

Some of the variables that have been shown to be related to sexual satisfaction in previous studies in the Western societies were found nonsignificant in the present study. These nonsignificant variables include place of birth, educational level, whether or not one had been married more than once, whether or not one had obtained medical help on sex-related matters, number of children, years of marriage, and family income.

As there is no strong theoretical reason to suggest that place of birth (Hong Kong SAR vs. Mainland China) is associated with sexual satisfaction; the nonsignificance of this variable is not surprising. This variable merely served as a control variable in our analysis. Most findings in Western societies show that educational level is related positively to sexual satisfaction (Call et al., 1995; Guo & Huang, 2005; Laumann et al., 1994; Liu, 2003). Our correlational analysis in Table 1 also shows a positive association between educational level and sexual satisfaction (r = .10 for wives and r = .08 for husbands, both are significant at [alpha] < .05). They do not reach significance, however, in the SEM. This suggests that after controlling other variables, educational level does not have a unique contribution in predicting sexual satisfaction.

It is expected that whether or not one had been married more than once and whether or not one had obtained medical help on sex-related matters might be related to sexual satisfaction. Both variables are not significantly associated with sexual satisfaction in our analysis. One speculation is that only around or less than 4% of the participants have been married more than once or have obtained medical help on sex-related matters. Thus, the statistical analysis may not be powerful enough to detect any meaningful differences. Regarding the effect of the number of children, the findings in previous studies were mixed. Some studies (e.g., Blumstein & Schwartz, 1983; Greenblat, 1983; James, 1981) suggest that the number of children is correlated negatively with frequency of sex, while Donnelly (1993) found a positive correlation between number of children and frequency of sex. In our study, we did not find any strong association between the number of children and sexual satisfaction or frequency of sex.

One interesting finding is that abortion history is not related to frequency of sex, wives' and husbands' interest in sex, and sexual satisfaction. The finding is different from Fok et al. (2005), who found that interest in sex, sexual satisfaction, and frequency of sex decreased for women with induced abortion. Several points should be noted when comparing these two findings. Our data were drawn from a large community survey, while the samples of Fok et al. (2005) were 140 women attending the postabortion clinic. About one-third of the participants in Fok et al. (2005) were single or unmarried, while all participants in the present study were married.

Coleman et al. (2005) reviewed many theoretical and empirical studies on abortion. They suggested that abortion might have positive and negative aspects on women's experiences. For example, perception of support from one's partner and family members is important to postpsychological adjustment after abortion. As many of the abortions in the present study were due to the unplanned pregnancy within marriage, it is expected that these women might have received some forms of support from their husbands and other family members.

Another possible explanation is that women with high interest in sex could lead to more sex and unplanned pregnancy and abortion. Thus, the association between abortion history and sexual satisfaction (and frequency of sex) is positive for some women. On the other hand, as abortion is not a pleasant experience, this could lead to a reduction in sexual satisfaction and frequency of sex for other women. When these two mechanisms are combined, the net effect is nil. Further empirical studies, however, are required to test whether these two speculations are tenable or not.

Limitations

This study has much strength because it makes use of a large-scale representative Hong Kong Chinese sample and because of the statistical techniques it employs in handling data involving couples. However, there are several limitations to this study worth mentioning. Although it has been stated that using a single item to measure the subjective feeling of satisfaction with sex is valid, there are other aspects to measuring sexual satisfaction that may have been overlooked. For example, instead of asking participants to rate their own subjective perspective on sexual satisfaction, Renaud et al. (1997) used orgasmic consistency as an indictor of sexual satisfaction. Other researchers have used various multi-item scales to measure sexual satisfaction. The use of such scales allows the researcher to consider various components or dimensions of sexual relationships. Future research may be attempted using multi-item scales to measure sexual satisfaction.

Another limitation is that only the wives' frequency of sex was asked in this study. One possible consequence of missing the husbands' frequency of sex is that the correlations between variables reported by wives and frequency of sex tend to be stronger than those reported by husbands. When we compare the standardized parameter estimates on frequency of sex of wives and of husbands in Table 4, it seems that they are quite comparable. Interest in sex is significant in predicting frequency of sex for both wives and husbands, while full-time work and age are significant for wives and husbands, respectively. These results seem to support that missing the husbands' frequency of sex may not have a large impact on the results.

Without the husbands' frequency of sex, however it is not possible to test whether the responses of the couples are consistent or not. The consistency of frequency of sex reported by the couples may be of theoretical interest to researchers. Further studies need to address this issue. There were several reasons for not separately asking husbands' report of frequency of sex after such question had been asked from the wives. First, sexual activities outside marriage are not very common in Hong Kong. Second, including questions on both partners' frequencies of sex during the same month may make the spouses worry about their responses during the interview. If their answers are not exactly the same, misunderstanding may arise from the suspicion that one of them may have sexual activities outside the marriage. We did not want to give the impression to our respondents that there could be discrepancy in their answers, and we had to avoid upsetting the couple as we needed their full cooperation in the interview. Finally, frequency of sex during that month is an objective event. It is reasonable to assume that the recall should not be subject to much error. In future research, special methods of questioning will need to be designed to minimize the sensitivity of this issue.

In addition, as often has been discussed by local researchers and scholars, cultural aspects are important in any study. For instance, some variables that should be examined in studies on Hong Kong Chinese relationships are the relationship individuals have with the parents or relatives of their spouses, living conditions, and the frequency of coupling activities.

It is of interest to compare how the responses on sexual satisfaction change over time. For example, Haavio-Mannila and Kontula (1997) found that sexual satisfaction has increased greatly, particularly among women from 1971 to 1992 in Finland. The current study utilizes data from the recent KAP Survey. Since the FPAHK carried out the KAP survey every 5 years (a new survey recently took place in 2007), further studies may compare how sexual satisfaction changes over time in Hong Kong.

We thank Cyprus Hoi-yan Lam for reviewing the literature review and the Family Planning Association of Hong Kong for access to the KAP survey data. This research was supported by the Health and Health Services Research Fund (HSRC/HCPF #02030581), Hong Kong SAR.

References

Acock, A. C. (2005). Working with missing values. Journal of Marriage and the Family, 67, 1012-1028.

Blumstein, P. & Schwartz, P. (1983). American couples: Money, work, sex. New York: Morrow.

Byers, E. S. & Demmons, S. (1999). Sexual satisfaction and sexual self disclosure within dating relationships. Journal of Sex Research, 36, 180-190.

Byers, E. S. & Grenier, G. (2003). Premature or rapid ejaculation: Heterosexual couples' perceptions of men's ejaculatory behavior. Archives of Sexual Behavior, 32, 261-270.

Call, V., Sprecher, S., & Schwartz, P. (1995). The incidence and frequency of marital sex in a national sample. Journal of Marriage and the Family, 57, 639-652.

Cheung, M. W. L. (2007). Comparison of methods of handling missing time-invariant covariates in latent growth models under the assumption of missing completely at random. Organizational Research Methods, 10, 609-634.

Cheung, M. W. L. & Chan, W. (2004). Testing dependent correlation coefficients via structural equation modeling. Organizational Research Methods, 7, 206-223.

Chevret, M., Jaudinot, E., Sullivan, K., Marrel, A., & De Gendre, A. S. (2004a). Quality of sexual life and satisfaction in female partners of men with ED: Psychometric validation of the index of sexual life (ISL) questionnaire. Journal of Sex & Marital Therapy, 30, 141-155.

Chevret, M., Jaudinot, E., Sullivan, K., Marrel, A., & De Gendre, A. S. (2004b). Impact of erectile dysfunction (ED) on sexual life of female partners: Assessment with the index of sexual life (ISL) questionnaire. Journal of Sex & Marital Therapy, 30, 157-172.

Christopher, F. S. & Sprecher, S. (2000). Sexuality in marriage, dating, and other relationships: A decade review. Journal of Marriage and Family, 62, 999-1017.

Chu, W. C. (1997). Who is doing what? The implication behind housework sharing. In S. K. Lau, M. K. Lee, P. S. Wan, & S. L. Wong (Eds.), Indicators of social development: Hong Kong 1995 (pp. 203-232). Hong Kong SAR: Hong Kong Institute of Asia-Pacific Studies, Chinese University of Hong Kong.

Coleman, P. K., Reardon, D. C., Strahan, T., & Cougle, J. R. (2005). The psychology of abortion: A review and suggestions for future research. Psychology & Health, 20, 237-271.

Doddridge, R., Schumm, W. R., & Berger, M. B. (1987). Factors related to decline in preferred frequency of sexual intercourse among young couples. Psychological Reports, 60, 391-395.

Donnelly, D. A. (1993). Sexually inactive marriages. The Journal of Sex Research, 30, 171-179.

Dunn, K. M., Croft, P. R., & Hackett, G. I. (2000). Satisfaction in the sex life of a general population sample. Journal of Sex & Marital Therapy, 26, 141-151.

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

Family Planning Association of Hong Kong (FPAHK). (2003). Knowledge, attitude and practice (KAP). Hong Kong: Author.

Fok, W. Y., Siu, S.-S. N., & Lau, T. K. (2005). Sexual dysfunction after a first trimester induced abortion in a Chinese population. European Journal of Obstetrics & Gynecology and Reproductive Biology, 126, 255-258.

Greeley, A. M. (1991). Faithful attraction: Discovering intimacy, love, and fidelity in American marriage. New York: Doherty.

Greenblat, C. S. (1983). The salience of sexuality in the early years of marriage. Journal of Marriage and the Family, 45, 289-299.

Guo, B. & Huang, J. (2005). Marital and sexual satisfaction in Chinese families: Exploring the moderating effects. Journal of Sex & Marital Therapy, 31, 21-29.

Guo, Y. N., Ng, E. M. L., Lou, C. H., & Zhao, H. G. (1999). A dynamic analysis of the marital and sexual satisfaction of couples in Shanghai six years after marriage. Population Information, 2, 25-31. (In Chinese)

Haavio-Mannila, E. & Kontula, O. (1997). Correlates of increased sexual satisfaction. Archives of Sexual Behaviour, 26, 399-419.

Harvey, J. H., Wenzel, A., & Sprecher, S. (2004). The handbook of sexuality in close relationships. Mahwah, NJ: Lawerence Erlbaum Associates Inc.

Henderson-King, D. H. & Veroff, J. (1994). Sexual satisfaction and marital well-being in the first years of marriages. Journal of Social and Personal Relationships, 11, 509-534.

Hu, L. & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

Hyde, J. S., DeLamater, J. D., Plant, E. A., & Byrd, J. M. (1996). Sexuality during pregnancy and the year postpartum. The Journal of Sex Research, 33, 143-151.

James, W. H. (1981). The honeymoon effect on marital coitus. The Journal of Sex Research, 17, 114-123.

Joreskog, K. G. & Sorbom, D. (1996). LISREL 8: User's reference guide. Chicago: Scientific Software International.

Joreskog, K. G. & Sorbom, D. (2005). LISREL 8.72 [Computer software]. Chicago, IL: Scientific Software International.

Kahn, J. R. & Udry, J. R. (1986). Marital coital frequency: Unnoticed outliers and unspecified interactions lead to erroneous conclusions. American Sociological Review, 51, 734-737.

Laumann, E., Gagnon, J., Michael, R., & Michaels, S. (1994). The social organization of sexuality: Sexual practices in the United States. Chicago: University of Chicago Press.

Liu, C. (2000). A theory of marital sexual life. Journal of Marriage and the Family, 62, 363-374.

Liu, C. (2003). Does quality of marital sex decline with duration? Archives of Sexual Behavior, 32, 55-60.

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

MacNeil, S. & Byers, E. S. (1997). The relationships between sexual problems, communication, and sexual satisfaction. The Canadian Journal of Human Sexuality, 6, 277-283.

Marsiglio, W. & Donnelly, D. (1991). Sexual relations in later life: A national study of married persons. Journal of Gerontology, 46, $338-344.

Mazur, A., Mueller, U., Krause, W., & Booth, A. (2002). Causes of sexual decline in aging married men: Germany and America. International Journal of Impotence Research, 14, 101-106.

Newsom, J. T. (2002). A multilevel structural equation model for dyadic data. Structural Equation Modeling, 9, 431-447.

Nicolosi, A., Moreira Jr., E. D., Villa, M., & Glasser, D. B. (2004). A population study of the association between sexual function, sexual satisfaction and depressive symptoms in men. Journal of Affective Disorders, 82, 235-243.

O'Fallon, J. B., Tsui, A., & Adewuyi, A. (2003). Social and proximate determinants of sexual activity in rural Nigeria. Journal of Biosocial Science, 35, 585-599.

Oggins, J., Leber, D., & Veroff, J. (1993). Race and gender differences in black and white newlyweds' perceptions of sexual and marital relations. Journal of Sex Research, 30, 151-160.

Ojanlatva, A., Helenius, H., Rautava, P., Ahvenainen, J., & Koskenvuo, M. (2003). Importance of and satisfaction with sex life in a large Finnish population. Sex Roles, 48, 543-553.

Olson, D. H. & DeFrain, J. (1994). Marriage and the family: Diversity and strengths. Mountain View. CA: Mayfield.

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

Perrone, K. M. & Worthington, E. L. (2001). Factors influencing ratings of marital quality by individuals within dual-career marriages: A conceptual model. Journal of Counseling Psychology, 48, 3-9.

Raudenbush, S. W. & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage Publications.

Renaud, C., Byers, E. S., & Pan, I. (1997). Sexual and relationship satisfaction in mainland China. Journal of Sex Research, 34, 399-410.

So, H. & Cheung, F. M. (2005). Review of Chinese sex attitudes and applicability of sex therapy for Chinese couples with sexual dysfunction. Journal of Sex Research, 42, 93-101.

Stafford, L., Kline, S. L., & Rankin, C. T. (2004). Married individuals, cohabiters, and cohabiters who marry: A longitudinal study of relational and individual well-being. Journal of Social and Personal Relationships, 21, 231-248.

Szinovacz, M. E. & (1983). Using couple data as a methodological tool: The case of marital violence. Journal of Marriage and the Family, 45, 633-644.

Szinovacz, M. E. & Egley, L. C. (1995). Comparing one-partner and couple data on sensitive marital behaviors: The case of marital violence. Journal of Marriage and the Family, 57, 995-1010.

Trussell, J. & Westoff, C. F. (1980). Contraceptive practice and trends in coital frequency. Family Planning Perspectives, 12, 246-249.

Udry, J. R. (1980). Changes in the frequency of marital intercourse from panel data. Archives of Sexual Behavior, 9, 319-325.

Udry, J. R. (1993). Coitus as demographic behaviour. In R. Gray, H. Leridon, & A. Spira (Eds.). Biomedical and demographic determinants of reproduction (pp. 85-97). Oxford: Clarendon Press.

Wendorf, C. A. (2002). Comparisons of structural equation modeling and hierarchical linear modeling approaches to couples' data. Structural Equation Modeling, 9, 126-140.

Whyte, M. K. (1990). Dating, mating, and marriage. New York: Walter de Gruyter.

Woody, J. D. & D'Souza, H. J. (1997). Sexual interaction in nonclinical couples. The Family Journal: Counseling and Therapy for Couples and Families, 5, 224-230.

Young, M., Denny, G., Luquis, R., & Young, T. (1998). Correlates of sexual satisfaction in marriage. The Canadian Journal of Human Sexuality, 7, 115-127.

Zhou, M. (1993). A survey of sexual states of married, healthy, reproductive age women. Journal of Psychology and Human Sexuality, 6, 15-28.

Correspondence should be addressed to Paul S.-F. Yip, Department of Social Work and Social Administration, 13/F K.K. Leung Building, University of Hong Kong, Pok Fu Lam Road, Hong Kong. E-mail: [e-mail]

Mike W.-L. Cheung

National University of Singapore 

Paul W.-C. Wong

The University of Hong Kong

Ka Y. Liu

University of Oxford

Paul S.-F. Yip

The University of Hong Kong

Susan Yun-sun Fan

Hong Kong

Tai-hing Lam

The University of Hong Kong

Table 1. Descriptive Statistics on Social Demographics

Variables       Categories                           Percentage

Household       $3,999 or below                           1.83
  income        $4,000-$5,999                             2.35
                $6,000-$7,999                             4.10
                $8,000-$9,999                             7.15
                $10,000-$14,999                          21.80
                $15,000-$19,999                          14.73
                $20,000-$24,999                          13.86
                $25,000-$29,999                           8.54
                $30,000-$39,999                           9.85
                $40,000-$59,999                           8.02
                $60,000 or above                          6.89
                Not reported                              0.87
Educational     No schooling/Kindergarten                 1.92
  level (W)     Primary                                  16.91
                Lower secondary                          21.27
                Upper secondary                          42.37
                Matriculation                             4.45
                Technical/vocational training             0.09
                (apprenticeship)
                Technical/vocational training             0.17
                (certificate courses)
                Tertiary (nondegree courses)/             4.62
                associate's degree courses
                Bachelor's degree                         6.45
                Master's/doctoral degree                  1.74
Educational     No schooling/Kindergarten                 1.39
  level (H)     Primary                                  17.00
                Lower secondary                          27.99
                Upper secondary                          29.21
                Matriculation                             4.27
                Technical/vocational training             0.17
                  (apprenticeship)
                Technical/vocational training             1.74
                  (certificate courses)
                Tertiary (nondegree courses)/             6.54
                  associate's degree courses
                Bachelor's degree                         8.63
                Master's/doctoral degree                  3.05
Full-time       Yes                                      59.63
  work (W)      No                                       32.69
                Not reported                              7.67
Full-time       Yes                                      87.71
  work (H)      No                                        2.44
                Not reported                              9.85

Note. "W" and "H" denote variables measured on wives and
husbands, respectively.

Table 2. Descriptive Statistics and Correlations of the Variables

                                             (1)      (2)      (3)

(1)  Family income                           1.00
(2)  No. of children                        -0.23     1.00
(3)  Year of marriage                       -0.16     0.47     1.00
(4)  Abortion history(W)                    -0.10     0.54     0.15
(5)  Age(W)                                 -0.10     0.38     0.82
(6)  Place of birth(W)                      -0.34     0.26     0.10
(7)  Educational level(W)                    0.56    -0.30    -0.37
(8)  Full-time work(W)                       0.34    -0.19    -0.10
(9)  More than one time                     -0.03     0.06    -0.09
     of marriage(W)
(10) Consulted medical                      -0.04     0.00    -0.09
     help on sex-related matters(W)
(11) Interest in sex(W)                      0.17    -0.16    -0.24
(12) Sexual satisfaction(W)                  0.12    -0.09    -0.08
(13) Frequency of sex in                     0.01    -0.06    -0.15
     that month
(14) Age (H)                                -0.25     0.34     0.68
(15) Place of birth (H)                     -0.29     0.24     0.21
(16) Educational level (H)                   0.56    -0.24    -0.30
(17) Full-time work (H)                      0.21    -0.13    -0.03
(18) More than one time                     -0.02    -0.09    -0.10
     of marriage (H)
(19) Consulted medical help                  0.00     0.03     0.02
     on sex-related matters (H)
(20) Interest in sex (H)                     0.09    -0.12    -0.19
(21) Sexual satisfaction (H)                 0.06    -0.05    -0.06
Mean                                         6.60     2.04    13.19
SD                                           2.40     1.39     7.15

                                             (4)      (5)      (6)

(1)  Family income
(2)  No. of children
(3)  Year of marriage
(4)  Abortion history(W)                     1.00
(5)  Age(W)                                  0.10     1.00
(6)  Place of birth(W)                       0.14     0.04     1.00
(7)  Educational level(W)                   -0.10    -0.27    -0.20
(8)  Full-time work(W)                      -0.07    -0.12    -0.16
(9)  More than one time                      0.10     0.02    -0.05
     of marriage(W)
(10) Consulted medical                       0.02    -0.08     0.02
     help on sex-related matters(W)
(11) Interest in sex(W)                     -0.05    -0.24    -0.09
(12) Sexual satisfaction(W)                  0.01     -0.1     0.00
(13) Frequency of sex in                     0.02    -0.18    -0.03
     that month
(14) Age (H)                                 0.10     0.77     0.20
(15) Place of birth (H)                      0.09     0.15     0.49
(16) Educational level (H)                  -0.09    -0.18     -0.2
(17) Full-time work (H)                     -0.03    -0.02    -0.13
(18) More than one time                      0.01    -0.04     0.01
     of marriage (H)
(19) Consulted medical help                  0.05    -0.03     0.00
     on sex-related matters (H)
(20) Interest in sex (H)                    -0.02    -0.20    -0.05
(21) Sexual satisfaction (H)                 0.01    -0.11     0.06
Mean                                         0.25    38.37     0.38
SD                                           0.44     6.58     0.49

                                             (7)      (8)      (9)

(1)  Family income
(2)  No. of children
(3)  Year of marriage
(4)  Abortion history(W)
(5)  Age(W)
(6)  Place of birth(W)
(7)  Educational level(W)                    1.00
(8)  Full-time work(W)                       0.21     1.00
(9)  More than one time                     -0.05    -0.03     1.00
     of marriage(W)
(10) Consulted medical                      -0.05     0.00     0.09
     help on sex-related matters(W)
(11) Interest in sex(W)                      0.20     0.10    -0.01
(12) Sexual satisfaction(W)                  0.10     0.03     0.02
(13) Frequency of sex in                     0.03    -0.03     0.02
     that month
(14) Age (H)                                -0.35    -0.16     0.00
(15) Place of birth (H)                     -0.18    -0.11    -0.04
(16) Educational level (H)                   0.65     0.13     0.01
(17) Full-time work (H)                      0.09     0.08    -0.10
(18) More than one time                     -0.03     0.00     0.13
     of marriage (H)
(19) Consulted medical help                 -0.05    -0.02     0.01
     on sex-related matters (H)
(20) Interest in sex (H)                     0.12     0.03    -0.02
(21) Sexual satisfaction (H)                 0.05     0.03    -0.02
Mean                                         4.04     0.64     0.02
SD                                           2.02     0.48     0.15

                                             (10)     (11)     (12)

(1)  Family income
(2)  No. of children
(3)  Year of marriage
(4)  Abortion history(W)
(5)  Age(W)
(6)  Place of birth(W)
(7)  Educational level(W)
(8)  Full-time work(W)
(9)  More than one time
     of marriage(W)
(10) Consulted medical                       1.00
     help on sex-related matters(W)
(11) Interest in sex(W)                      0.04     1.00
(12) Sexual satisfaction(W)                 -0.04     0.51     1.00
(13) Frequency of sex in                     0.04     0.30     0.38
     that month
(14) Age (H)                                -0.08    -0.20    -0.10
(15) Place of birth (H)                     -0.03    -0.07     0.02
(16) Educational level (H)                  -0.04     0.21     0.12
(17) Full-time work (H)                     -0.03     0.03     0.07
(18) More than one time                      0.03    -0.03    -0.03
     of marriage (H)
(19) Consulted medical help                  0.13    -0.08    -0.06
     on sex-related matters (H)
(20) Interest in sex (H)                     0.08     0.26     0.24
(21) Sexual satisfaction (H)                -0.02     0.24     0.32
Mean                                         0.04     3.32     3.69
SD                                           0.20     0.62     0.68

                                            (13)     (14)     (15)

(1)  Family income
(2)  No. of children
(3)  Year of marriage
(4)  Abortion history(W)
(5)  Age(W)
(6)  Place of birth(W)
(7)  Educational level(W)
(8)  Full-time work(W)
(9)  More than one time
     of marriage(W)
(10) Consulted medical
     help on sex-related matters(W)
(11) Interest in sex(W)
(12) Sexual satisfaction(W)
(13) Frequency of sex in                     1.00
     that month
(14) Age (H)                                -0.19     1.00
(15) Place of birth (H)                     -0.02     0.26     1.00
(16) Educational level (H)                   0.02    -0.29    -0.23
(17) Full-time work (H)                      0.07    -0.07    -0.12
(18) More than one time                      0.00     0.14     0.03
     of marriage (H)
(19) Consulted medical help                 -0.06    -0.02    -0.05
     on sex-related matters (H)
(20) Interest in sex (H)                     0.28    -0.22    -0.11
(21) Sexual satisfaction (H)                 0.25    -0.09     0.02
Mean                                         2.04    42.63     0.36
SD                                           1.14     8.27     0.48

                                            (16)     (17)     (18)

(1)  Family income
(2)  No. of children
(3)  Year of marriage
(4)  Abortion history(W)
(5)  Age(W)
(6)  Place of birth(W)
(7)  Educational level(W)
(8)  Full-time work(W)
(9)  More than one time
     of marriage(W)
(10) Consulted medical
     help on sex-related matters(W)
(11) Interest in sex(W)
(12) Sexual satisfaction(W)
(13) Frequency of sex in
     that month
(14) Age (H)
(15) Place of birth (H)
(16) Educational level (H)                   1.00
(17) Full-time work (H)                      0.12     1.00
(18) More than one time                      0.00     0.03     1.00
     of marriage (H)
(19) Consulted medical help                 -0.03     0.03     0.02
     on sex-related matters (H)
(20) Interest in sex (H)                     0.12     0.06    -0.02
(21) Sexual satisfaction (H)                 0.08     0.02     0.02
Mean                                         4.30     0.97     0.04
SD                                           2.34     0.16     0.19

                                             (19)     (20)     (22)

(1)  Family income
(2)  No. of children
(3)  Year of marriage
(4)  Abortion history(W)
(5)  Age(W)
(6)  Place of birth(W)
(7)  Educational level(W)
(8)  Full-time work(W)
(9)  More than one time
     of marriage(W)
(10) Consulted medical
     help on sex-related matters(W)
(11) Interest in sex(W)
(12) Sexual satisfaction(W)
(13) Frequency of sex in
     that month
(14) Age (H)
(15) Place of birth (H)
(16) Educational level (H)
(17) Full-time work (H)
(18) More than one time
     of marriage (H)
(19) Consulted medical help                  1.00
     on sex-related matters (H)
(20) Interest in sex (H)                     0.00     1.00
(21) Sexual satisfaction (H)                -0.04     0.44     1.00
Mean                                         0.03     3.69     3.84
SD                                           0.18     0.67     0.68

Note. "W" and "H" denote variables measured on wives and husbands,
respectively. Correlations are significant at [alpha] = .05 and
[alpha] = .01 when the absolute correlations are larger than .07
and .09, respectively.

Table 3. Frequency of Sex Broken Down by Sex and by Age
Groups

                   Number of Times of Sex in that Month

Age Group       0 (%)   1-2 (%)   3-4 (%)     5-9 (%)    10 (%) or More

Wife
  19-29          4.2      18.8      27.1        27.1        22.9
  30-39         10.4      18.1      33.6        28.0        10.0
  40-49         10.5      29.0      27.9        25.5         7.0
Husband
  21-30          9.2      19.7      21.1        27.6        22.4
  31-40          8.4      15.4      35.8        28.0        12.3
  41-50         10.5      24.5      29.7        27.3         7.9
  51-60          9.6      40.4      22.3        25.5         2.1
  61-70         22.2      55.6      22.2         0.0         0.0
  71-80         14.3      14.3      42.9        14.3        14.3
Total            9.8      23.0      30.4        26.8         9.9

Table 4. Standardized Parameter Estimates of the Path Model

                                                    Wives

                                   Frequency of Sex        Sexual
Predictors                          in that Month       Satisfaction

Age (a)                                 -0.023             -0.015
Place of birth (a) (0-HK;               -0.028              0.063
  1-Non-HK)
Educational level (a)                   -0.012             -0.019
With full-time work job (a)             -0.086 *           -0.029
  (0-No; 1-Yes)
More than one time of                    0.021              0.050
  marriage (a) (0-No; 1-Yes)
Interest in sex (a)                      0.248 **           0.506 **
Consulted medical help on sex            0.016             -0.051
  related matters before (a)
  (0-No; 1-Yes)
No. of children (b)                     -0.028             -0.089
Abortion history (b)                     0.052              0.063
  (0-No; 1-Yes)
Years of marriage (b)                    0.028              0.068
Family income (b)                       -0.021              0.079

                                                Husbands

                                   Frequency of Sex        Sexual
Predictors                          in that Month       Satisfaction

Age (a)                                -0.132 *           -0.020
Place of birth (a) (0-HK;               0.020              0.073
  1-Non-HK)
Educational level (a)                  -0.072              0.025
With full-time work job (a)             0.039             -0.018
  (0-No; 1-Yes)
More than one time of                   0.029              0.043
  marriage (a) (0-No; 1-Yes)
Interest in sex (a)                     0.181 **           0.430 **
Consulted medical help on sex          -0.045             -0.039
  related matters before (a)
  (0-No; 1-Yes)
No. of children (b)                       --              -0.019
Abortion history (b)                      --               0.025
  (0-No; 1-Yes)
Years of marriage (b)                     --               0.035
Family income (b)                         --               0.036

Note. (a) Variables measured on husbands and wives separately;
(b) Variables based on wives only. Variables with (*) denote significant
for the unstandardized variables. ** p < .001, * p < .01.