Sexual and relationship satisfaction in mainland China.

After the establishment of the People's Republic of China in 1949 and until recently, there has been official repression of many forms of sexual expression in China. This applied to music, visual arts, and literature deemed as sexual (Evans, 1995; Pan, 1993b); sex education, research, and therapy (Gil, 1993; Herold & Byers, 1994); and public displays of affection such as caressing or kissing, even between married individuals (Pan, 1989b). As sexuality was promoted only for reproduction and the enhancement of the family and society, sexual pleasure seeking was frowned upon (Pan, 1993a). Women in particular were not expected to derive pleasure from sex, and their sexual desires were thought to be dictated by their capacity to bear children (Evans, 1995; Pan, 1993a).

Today, there appears to be greater tolerance and less social control of diverse sexual attitudes and behaviors in China than there has been in the past (Herold & Byers, 1994). However, we are left with an "ignorance regarding existing conditions" (Zhou, 1993, p. 16), as much of what we know about sexuality in China is based on impressions of short-term visitors there rather than empirical data from Chinese citizens. Therefore, the main purposes of the current study were to provide data on sexual behavior in China and to validate instruments and methodology for sex research with a Chinese sample. Specifically, we investigated the relationship satisfaction and sexual satisfaction of married Chinese men and women within the framework of the Interpersonal Exchange Model of Sexual Satisfaction (IEMSS; Lawrance & Byers, 1995, in press).

Characteristics of Sexual Relationships

A limited number of studies have been used to investigate the frequency of marital sexual activity in China. In one study of the sexual behaviors of 1,800 married Chinese women, Zhou (1993) found that the majority (65.9%) engaged in coitus with their husbands between once and twice a week. In contrast, a number of researchers have found low frequencies of non-coital sexual behaviors in Chinese marriages (Pan, 1993a, 1994; Zhou, 1993). Therefore, our first objective was to describe characteristics of the relationships of married Chinese men and women, including their affectionate and sexual behavior, sexual concerns, sexual problems, relation ship satisfaction, and sexual satisfaction. In addition, we aimed to describe a number of aspects of the sexual relationship that have been shown to be related to sexual and relationship satisfaction (Lawrance & Byers, 1995). Specifically, the levels of rewards and costs that the partners exchange in their sexual relationship were investigated. Rewards were defined as pleasurable and gratifying exchanges, whereas costs were defined as exchanges that require. effort and/or produce pain, embarrassment, or anxiety. Similarly, the comparison of the levels of rewards and costs relative to the levels of rewards and costs that the individual expects to receive in a sexual relationship were investigated (hereafter called relative rewards and relative costs). The degree of equality perceived between the participant's levels of rewards and costs and those of his or her partner were also examined. Finally, we examined participants' affective orientation toward sex, or erotophobia-erotophilia.

Gender Differences

Lawrance and Byers (1995) found that, among a Canadian sample of mostly sexually satisfied married and cohabiting individuals, men and women did not differ in their sexual satisfaction nor in their perceptions of how rewarding/costly their sexual relationship was. However, feminist scholars have argued that differences in the socialization of boys and girls and the sexual script for men and women have an impact on men's and women's experiences of sexual satisfaction (Lawrance & Byers, 1995; Poulin, 1992; Tiefer, 1988). In a culture such as that in China, in which more emphasis is placed on the sexual pleasure of men than that of women (Evans, 1995; Pan, 1993a), gender differences may be more apparent. Zhou (1993) found that husbands were more likely than wives to initiate coitus. In addition, whereas the majority of the husbands in this study (97%) experienced ejaculations "every time," only 31% of the wives experienced orgasm "mostly" every time. Although these findings are consistent with a culture in which women are expected to be sexually passive and only men are expected to receive pleasure from sex, further research is needed. Therefore, our second goal was to examine gender differences in the sexual relationships of married Chinese individuals.

Sexual Satisfaction and Relationship Satisfaction

Sexual satisfaction is important in the marriages of Chinese men and women. In one study, Pan (1993a) found that married Chinese women and men rated sexual satisfaction as the third and fourth, respectively, most important factors in marital satisfaction. Similarly, 47% of people seeking marital counseling and wanting to divorce gave sexual difficulties as the primary reason. North American research also has shown that characteristics of the overall relationship, such as relationship satisfaction, love, level of intimacy, and the amount of physical affection, are associated with sexual satisfaction (Lawrance & Byers, 1995; Oggins, Leber, & Veroff, 1993). Further, this relationship may be bi-directional (Henderson-King & Veroff, 1994). Therefore, our third goal was to investigate the relationship between sexual and relationship satisfaction in a sample of married Chinese men and women.

Although sexual satisfaction per se has rarely been investigated among married Chinese men and women, other measures have been used as indices of sexual satisfaction. As in some North American research, in Chinese research, orgasmic consistency has been used as an index of sexual satisfaction in married Chinese women. Although orgasmic consistency has been shown to be related to sexual satisfaction, the two are conceptually distinct. The relation between orgasmic consistency and sexual satisfaction is complicated by Pan's (1994) findings that 54% of Chinese women experiencing low orgasmic consistency felt this was acceptable and an additional 11% felt their frequency of orgasm was too high.

Sexual satisfaction has also been conceptualized as the absence of dissatisfaction. Researchers in one study asked 1,800 women about their and their husbands' sexual dissatisfaction (Zhou, 1993). Based on reports by the wives, this study showed that 10% of the wives and 5% of the husbands experienced sexual dissatisfaction. The wives reported that their own lack of cooperation and the interference of contraception were the primary causes of their husbands dissatisfaction, whereas their own absence of desire and orgasm were the main reasons for their sexual dissatisfaction. In that study, frequency of coitus, precoital caressing (not defined), frequency of wives' orgasm, and husband's approval of the contraceptive method were positively correlated with sexual satisfaction in both the wives and husbands (as reported by the wives). Sexual satisfaction of husbands was negatively correlated with the husband's age, and sexual satisfaction of the wives was negatively correlated with the duration of marriage. As the data were based on reports by the wives, who may or may not have consulted with their husbands, it is possible that different correlates of sexual dissatisfaction may have been found if the men had been surveyed. In addition, Zhou (1993) did not directly assess sexual satisfaction but likened the absence of dissatisfaction to satisfaction. Although often equated, sexual satisfaction is not the absence of dissatisfaction but rather "an affective response arising from one's subjective evaluation of the positive and negative dimensions associated with one's sexual relationship" (Lawrance & Byers, 1995, p. 268).

These studies have shed some light on the sexual lives of Chinese women. However, the research was not guided by a theoretical model, and sexual satisfaction was poorly conceptualized. Hence, our fourth objective was to explore factors related to the sexual satisfaction of married Chinese men and women. Specifically, we examined the relationships among the frequency of affectionate and sexual behavior, sexual functioning variables, erotophobia-erotophilia, and sexual satisfaction. In addition, because life cycle events may have an impact on a couple's relationship (Hatfield & Rapson, 1993; Morokoff & Gilliland, 1993), we examined the relationship between the impact of recently experienced life events and sexual satisfaction. Similarly, we investigated the relationships between sexual satisfaction and several sexual exchange variables: the level of rewards and costs, relative level of rewards and costs, the perceived equality of rewards and costs, as well as the number of rewards and costs. We also examined the relationship between these variables and overall relationship satisfaction. Finally, we explored the relationships between the types of sexual concerns and problems experienced and sexual satisfaction.

The Interpersonal Exchange Model of Sexual Satisfaction

We investigated the sexual satisfaction of both men and women within the theoretical model proposed by Lawrance and Byers (1995). The IEMSS was developed to overcome some methodological weaknesses found in studies used to investigate sexual satisfaction and to provide a conceptual framework for studying sexual satisfaction (Lawrance, 1994; Lawrance & Byers, in press). As an exchange model, the IEMSS takes into account the levels of rewards and costs, relative levels of rewards and costs, and perceived equality of rewards and costs that the partners exchange in their sexual relationship (Lawrance & Byers, 1995). In predicting sexual satisfaction, the IEMSS also takes into account the relational context in which these sexual exchanges occur.

The IEMSS predicts that sexual satisfaction will be greater to the extent that relationship satisfaction is greater. Sexual satisfaction is also expected to be greater to the extent that the level of rewards (REW) incurred in the sexual relationship exceeds the level of costs (CST) (Lawrance & Byers, 1995). In addition, sexual satisfaction will be greater to the extent that one's relative level of rewards ([CL.sub.REW]) exceeds one's relative level of costs ([CL.sub.CST]), Finally, sexual satisfaction will be greater if equality is perceived to exist between one's levels of rewards ([EQ.sub.REW]) and those of one's partner and between one's levels of costs and one's partner's ([EQ.sub.CST]). Each of the three sexual exchange components of the model (REW-CST, [CL.sub.REW] - [CL.sub.CST], and [EQR.sub.REW] and [EQ.sub.CST]) was expected to add uniquely to the prediction of sexual satisfaction after controlling for the relational component (relationship satisfaction).

The IEMSS has been shown to have good validity for short-term dating relationships (Byers, Demmons, & Lawrance, in press) as well as long-term heterosexual relationships (Lawrance & Byers, 1995). In a Canadian sample of 149 women and 94 men in long-term heterosexual relationships, Lawrance and Byers (1995) found that each component of the model added to the prediction of sexual satisfaction such that, together, the IEMSS components (including relationship satisfaction, which was not in their original model) accounted for 79% of the variance in sexual satisfaction. The model worked equally well for men and women, individuals with and without children, and individuals with higher and lower relationship satisfaction. We validated the model on our sample of married Chinese men and women.

Hypotheses

In the current study we examined three sets of hypotheses. The first series of hypotheses concerned gender differences in the sexual relationships of married Chinese individuals. Women were predicted to characterize their sexual relationships as less,,rewarding, more costly, less equal, and less satisfying than would men. Women were also predicted to report a smaller number of rewards and a larger number of costs than men. In addition, women were expected to be more erotophobic than were men.

The second set of hypotheses pertained to the relationship between both relationship satisfaction and sexual satisfaction and a number of sexual and relationship variables. We hypothesized that the frequency of affectionate and sexual behavior would be positively related to relationship satisfaction and sexual satisfaction and that the presence of sexual concerns and problems would be negatively related to relationship satisfaction and sexual satisfaction. In addition, reward level, relative reward level, and equality of rewards and costs in the sexual relationship were expected to be positively related to relationship satisfaction and sexual satisfaction, whereas cost level and relative cost level in the sexual relationship were expected to be negatively related to relationship satisfaction and sexual satisfaction. We also predicted that relationship satisfaction and sexual satisfaction would be greater to the extent that the number of sexual rewards was greater and the number of sexual costs was fewer.

The final hypotheses concerned the IEMSS. Each of the four components of the model was expected to be positively related to sexual satisfaction and to add uniquely to the prediction of sexual satisfaction. The IEMSS was expected to predict sexual satisfaction equally well for men and women. The IEMSS was also expected to predict sexual satisfaction over and above other factors associated with sexual satisfaction.

Method

Participants

Questionnaires were mailed to 198 women and 213 men in Beijing and 257 women and 256 men in Shanghai who were married. Completed questionnaires were returned by 193 (41%) men and 231 (51%) women. The returned questionnaires had a high frequency of missing data. Twenty-four of the 424 participants were omitted from the sample because they were missing an excess of 10% of data. In addition, 39 participants who had missing data for more than 1 item on the Global Measure of Sexual Satisfaction and the Global Measure of Relationship Satisfaction, the 2 primary criteria, were excluded from all analyses. To minimize the loss of participants because of missing data, if only one item was missing for these scales, the participant's mean score on the remaining four items was substituted for that missing value. Fifty-seven participants who were missing one of the five items on one of the scales were retained. Most of these respondents did not complete the positive-negative dimension of these scales. This may have been due to ambiguity of the meaning of the Chinese characters used to translate these concepts. All other missing data were replaced with the group (gender) mean for that item.

The sample consisted of 170 men and 191 women for 361 participants. Approximately half of the sample resided in Shanghai (52.4%) and half in Beijing (47.6%). They ranged in age from 21 to 77 (M = 39.6, SD = 8.9), and their partners ranged in age from 21 to 69 (M = 39.5, SD = 9.1). Participants had been married between 1 and 42 years (M = 12.6, SD = 8.7), and the majority (87.8%) had at least 1 child. The majority of participants (60.4%) had had one or more other serious romantic relationships prior to marriage.

Measures

All measures were translated from English to Chinese by an individual whose native tongue is Chinese. Translations were then verified by two other individuals who were born and raised in China but who had lived in Canada and studied in English for a number of years. All measures were incorporated into one questionnaire, the Interpersonal Exchange Model of Sexual Satisfaction Questionnaire (Lawrance & Byers, in press).

A background questionnaire was used to collect demographic information such as gender, age of respondent, age of respondent's partner, duration of marriage, previous serious romantic involvement, and child status. Other background questions, including current involvement in therapy, frequency of physically affectionate behavior (kissing, hugging, cuddling), and frequency of sexual activities, were embedded in the appropriate sections of the overall questionnaire.

Impact of life , events was measured using the Impact of Life Events Scale-Revised. Participants indicated whether, in the previous three months, they had experienced any major life events such as the death of a family member/close friend, a change in employment or finances, moving to a new residence, or gaining a new family member. Ten items were selected from Sarason, Johnson, and Seigel's (1978) 44-item life events scale. Items were selected to represent typical life cycle events that may have an impact on a couple's sexual relationship (Hatfield & Rapson, 1993; Morokoff & Gilliland, 1993). On a five-point scale ranging from very negative (-2) to very positive (+2), participants indicated the impact of each event they had experienced. Impact ratings were summed across items, yielding final scores ranging from -20 (all events were experienced and all had a "very negative" impact) to +20 (all events were experienced and all had a "very positive" impact).

The Global Measure of Sexual Satisfaction (GMSEX) (Lawrance & Byers, 1995, in press) was used to assess satisfacton with the sexual relationship. Respondents rated their sexual relationship on five seven-point bipolar scales: good-bad, pleasant-unpleasant, positive-negative, satisfying-unsatisfying, and valuable-worthless. Possible scores on the GMSEX range from 5 to 35, with higher scores indicating greater sexual satisfaction. Evidence for the validity of GMSEX was provided in a sample of 133 individuals in long-term heterosexual relationships as well as in a sample of 90 college students who had been dating/cohabitating for more than 1 year (Lawrance & Byers, 1995, in press). Good test-retest reliability and internal consistency of GMSEX were also provided by Lawrance and Byers (1995, in press). Internal consistency was high for the current sample ([Alpha] = .97 for men and [Alpha] = .98 for women).

The Global Measure of Relationship Satisfaction (GMREL) is similar to the GMSEX, except that participants rate their overall relationship. Lawrance and Byers (1995, in press) provided evidence for the validity and reliability (test-retest and internal consistency) of GMREL. Internal consistency was high for the current sample ([Alpha] = .98 for men and [Alpha] = .97 for women).

Level of sexual functioning was assessed using two questions based on items employed by Frank, Anderson, and Rubenstein (1978) in their study of the prevalence of sexual dysfunction in a non-clinical population. Respondents were asked to indicate which of the nine presented sexual concerns they had experienced in the previous three months. Seven items were taken directly from Frank et al. (1978) (e.g., inability to relax, too little foreplay, attracted to someone else). Two additional items were included: I like to do things my partner does not like to do and My partner asks me to do things that I do not like to do. In a separate question, respondents were asked to indicate whether they and/or their partner had experienced any of the seven listed sexual problems in the previous three months. Items taken from Frank et al. (1978) addressed inhibited sexual desire, inhibited sexual excitement, premature orgasm, inhibited orgasm, and anorgasmia. Two additional items were included: medication interferes with the sexual response and illness interferes with the sexual response.

The Exchanges Questionnaire is a six-item measure. Three items require respondents to think of their sexual relationship over the previous three months and indicate (a) how rewarding their sexual relationship is, (b) how their level of rewards compares to their expectations about how rewarding their sexual relationship "should" be (relative level of reward), and (c) how their level of rewards compares with the level of rewards they think their partner receives in the sexual relationship (equality of reward). Three other items assess costs using the same format. Level of rewards (REW) and level of costs (CST) were rated on nine-point scales with endpoints, not at all rewarding [costly] (1) and extremely rewarding [costly] (9). Relative reward level ([CL.sub.REW]) and relative cost level ([CL.sub.CST]) were also rated on nine-point scales with endpoints, much less rewarding [costly] in comparison (1) and much more rewarding [costly] in comparison (9). Perceived equality of rewards ([EQ.sub.REW]) and perceived equality of costs ([EQ.sub.CST]) were rated on nine-point scales with anchors, my rewards [costs] are much higher and my partner's rewards [costs] are much higher. The latter two scales were recoded such that the mid-point, representing perfect equality, was assigned a score of 4, and both endpoints were assigned scores of 0. The REW, CST, [CL.sub.REW]- and [CL.sub.CST] items were used to calculate two components of the IEMSS. REW-CST and [CL.sub.REW]) [CL.sub.CST] scores were calculated by subtracting the cost score from the reward score.

Lawrance and Byers (in press) provided evidence for the validity of the measures of level of rewards and level of costs in their student sample. They found that level of rewards was significantly correlated with the Index of Sexual Satisfaction (ISS; Hudson, Harrison, & Crosscup, 1981) as well as with a single-item measure of sexual satisfaction, rs = -.66 and .64, p [is less than] 001, respectively. Level of costs was significantly correlated with the ISS but not with the single item measure, r = .30, p [is less than] . 0 1 and r = -. 15, ns, respectively. Further support for the measures of level of reward and level of costs was provided in a study of 133 individuals involved in long-term heterosexual relationships (Lawrance & Byers, 1995). In this sample, rewards and costs were significantly correlated with GMSEX, rs = .83 and -.73, p [is less than] .001, respectively. Similarly, good three-month test-retest reliability was demonstrated for rewards, r .64, p [is less than] .001 and costs, r = .56, p [is less than] .001.

The Rewards/Costs Checklist was developed based on responses to open-ended questions (Lawrance & Byers, in press). It includes 46 items. Examples of these items are level of affection expressed during sexual activities, amount of spontaneity in your sex life, degree of privacy you and your partner have for sexual activities, and engaging in sexual acts that you dislike but your partner enjoys. For each item, respondents indicate whether it is generally a reward in their sexual relationship, generally a cost, both a reward and a cost, or neither a reward nor a cost. An item was considered a reward [cost] if a respondent indicated that it was a reward [cost] or both a reward and a cost.

The Sexual Opinion Survey (SOS; Fisher, Byrne, White, & Kelley, 1988) is a 21-item scale designed to measure erotophobia-erotophilia--i.e., the disposition to respond to sexual cues with negative to positive affect. We used the five-item short form of the SOS, which consists of five items that were most predictive of total SOS scores for men and women on the basis of regression analysis. These items assess attitudes toward masturbation, erotic material, public exposure, unusual sex practices, and having more than one long-term sexual relationship. According to Fisher et al. (1988), the short form of the SOS is highly correlated with the original version (rs = .71 and .82 for female and male samples, respectively). Respondents indicate on a seven-point rating scale how strongly they endorse each statement describing a positive or negative evaluative response to a sexual situation. Higher scores indicate more positive evaluation/affective response to sexual cues or greater erotophilia. The original version of the SOS has been demonstrated to show good validity and reliability (Fisher et al., 1988; Lawrance & Byers, 1995, in press). Chronbach's [Alpha] in the current sample was .70, which is comparable to internal consistency estimates for the short form with a Canadian student sample (Demmons, Lawrance, & Byers, 1993).

Procedure

Questionnaire packages were mailed to randomly selected married individuals between the ages of 22 and 70 living in Beijing or Shanghai. Because of the complexity of the questionnaire and most Chinese individuals' unfamiliarity with this type of questionnaire, only individuals who had completed high school were included in the sample. In China, the Public Security Bureau in each city maintains computerized personal information on each citizen. Four residential communities in Shanghai and five residential communities in Beijing were randomly selected. Within these communities, individuals who met the selection criteria were randomly selected according to the numbers of their identity cards. All correspondence with potential participants was from the researchers, not from the Public Security Bureau. All questionnaire packages included a cover letter detailing the nature of the study, a questionnaire, and a pre-addressed, stamped return envelope. Individuals who were married and wished to participate were asked to complete and return the survey. Participants were given an opportunity to make comments at the end of the questionnaire.

Results

A MANOVA was conducted to determine if participants who were excluded from the analyses because of the missing data criterion differed from those who were retained, on the following demographic variables: gender, age, age of partner, duration of marriage, previous serious romantic involvement, child status, and impact of life events. One participant was excluded from this analysis because of additional missing data. Results of the MANOVA were significant, F(7, 415) = 4.01, p [is less than] .001. Follow-up ANOVAs indicated that those who were excluded from the sample were older, F(1, 421) = 9.71, p [is less than] .01; had older partners, F(1, 421) = 16.47, p [is less than] .001; had been married longer, F(1, 421) = 15.80, p [is less than] .001; and reported that the impact of life events they had experienced was less positive, F(1, 421) = 5.94, p [is less than] .05, than participants who were retained in the sample.

A MANOVA was conducted to determine if the men and the women who were retained in the sample differed on any of the following demographic variables: city, age, age of partner, duration of marriage, previous serious romantic involvement, child status, and impact of life events. Results of the MANOVA were significant, F(7, 353) = 30.43, p [is less than] .001. Follow-up ANOVAs indicated that significantly more men (70%) than women (52%) reported one or more previous serious romantic involvements, F(1, 359) = 12.78, p [is less than] .001. Also, women's partners were reported to be significantly older (M = 40.5) than were men's partners (M = 38.3), R 1, 359) = 5.40, p [is less than] .05. Finally, although both men and women reported a neutral impact of life events, the impact reported by women (M = 0.2) was significantly more positive than that reported by men (M = -0. 1), ]?(l, 359) = 7.07, p [is less than] .0 1.

Sexual Attitudes

SOS scores indicated that, on average, participants had neither an erotophobic nor an erotophilic orientation toward sex (see Table 1). As there is little research on specific sexual attitudes in China, responses to the individual items on the SOS were also examined as an indication of sexual attitudes toward these five separate areas of sexuality, On average, attitudes toward masturbation (M = 4.5, SD = 2.0), sexually explicit material (M = 5.0, SD = 1.8), and unusual sex practices (M = 4.7, SD = 1.9) were somewhat positive. Attitudes toward a person exposing himself or herself in public (M = 3.6, SD = 2.2), and having more than one long-term sexual partner (M = 3.4, SD = 2.1), were somewhat negative.

Table 1
Descriptive Statistics and Gender Differences for Relationship
Characteristics

                                                 Men        Women
Variable                                      M       SD     M
Erotophobia-erotophilia                     20.7(*)   4.5   19.7(**)
Relationship [satisfaction.sup.a]           24.3      6.9   25.5
Sexual [satisfaction.sup.a]                 23.1(*)   6.5   24.6
Frequency of physical [affection.sup.a]      3.3      1.4    3.4
Frequency of sexual [activity.sup.a]         3.2      1.2    3.4
Number of sexual [concerns.sup.a]           1.3(*)    1.4    1.7(*)
Number of sexual problems--[self.sup.a]      1.2      1.1    1.4
Number of sexual problems--[partner.sup.a]   1.5(**)  1.4    0.7(**)
Level of [rewards.sup.b]                     5.2      1.9    5.2
Level of [costs.sup.b]                       4.9      1.8    4.3
Relative level of [rewards.sup.b]            4.5      1.9    4.7
Relative level of [costs.sup.b]              4.9      1.8    4.6
Equality of [rewards.sup.b]                  2.8      1.1    2.9
Equality of [costs.sup.b]                    2.6      1.2    2.7
Number of [rewards.sup.c]                   24.8      8.9   24.0
Number of [costs.sup.c]                     21.7      9.3   18.4(*)

                                             Women      Total
Variable                                     SD        M      SD
Erotophobia-erotophilia                      5.2     20.2    4.9
Relationship [satisfaction.sup.a]            6.6     24.9    6.8
Sexual [satisfaction.sup.a]                  6.8     23.9    6.8
Frequency of physical [affection.sup.a]      1.6      3.3    1.5
Frequency of sexual [activity.sup.a]         1.3      3.3    1.2
Number of sexual [concerns.sup.a]            1.6      1.5    1.6
Number of sexual problems--[self.sup.a]      1.2      1.3    1.2
Number of sexual problems--[partner.sup.a]   1.0      1.1    1.2
Level of [rewards.sup.b]                     2.2      5.2    2.1
Level of [costs.sup.b]                       1.8      4.6    1.8
Relative level of [rewards.sup.b]            2.0      4.6    1.9
Relative level of [costs.sup.b]              1.7      4.7    1.8
Equality of [rewards.sup.b]                  1.3      2.8    1.2
Equality of [costs.sup.b]                    1.2      2.7    1.2
Number of [rewards.sup.c]                    9.9     24.4    9.4
Number of [costs.sup.c]                     10.0     20.0    9.8

Note: N = 361 (170 men, 191 women). Higher scores indicate a greater level of that variable. Variables with the same superscript were entered in the same MANOVA. Gender differences in erotophobia-erotophilia were examined in an ANOVA. Asterisks reflect a significant gender difference for that variable.

(**) p <.001

(*) p <.05

Characteristics of the Relationship

Descriptive statistics for relationship characteristics can be found in Table 1. Overall, respondents reported moderate levels of. relationship satisfaction and sexual satisfaction. On a six-point scale ranging from "rarely or never" to "once a day or more," respondents reported moderate frequencies of physical affectionate and sexual behavior. However, a substantial percentage of respondents reported that they rarely or never engaged in physical affection (16%) and/or sex (13%). In contrast, a smaller percentage reported that they engaged in physically affectionate (9%) or sexual activity (1.4%) once a day or more. The average frequency of affectionate behavior (M = 3.3, SD = 1.5) and sexual behavior (M = 3.3, SD = 1.2) corresponded to a rating between "2 or 3 times a month" and "once or twice a week."

Respondents were also asked to report any recent concerns regarding their sexual functioning. Almost three quarters of both men and women reported experiencing at least one of the nine concerns listed. Too little foreplay before intercourse (24.9%), too little tenderness after intercourse (24.4%), and feeling turned off (24.1%) were the most commonly experienced concerns. When asked to indicate whether they and/or their partner had experienced any sexual difficulties within the last three months, more than two thirds of participants indicated at least one problem for themselves (78.7%), and slightly more than half (56.6%) reported at least one problem for their partner. The most frequent sexual problems for women, both by their and their partner's reports, were inhibited orgasm (38.7% and 40.6%, respectively) and difficulty getting excited (35.1% and 41.8%, respectively). The most frequent sexual problems for men, both by their and their partner's reports' were premature orgasm (37.6% and 31.4%, respectively) and maintaining sexual excitement (37.6% and 19.4%, respectively). Only five (1.4% of the sample) individuals reported having sought counseling for sexual problems.

Descriptive statistics for the sexual exchange variables are listed in Table 1. On average, participants indicated that their sexual relationship was moderately rewarding and moderately costly. Overall, participants perceived their level of rewards to be higher than their level of costs, t(360) = 3.94, p [is less than] .00 1. Participants indicated that the rewards and costs in their sexual relationship were comparable to their expectations. Their relative rewards and relative costs did not differ from each other, t(360) = -.81, ns. Furthermore, respondents evaluated their level of rewards and costs to be moderately equal to the levels of rewards and costs of their partner. Reward equality and cost equality did not differ significantly, t(360) = 2.04, ns. Overall, respondents reported a greater number of rewards than costs, 0360) = 6.67, p [is less than] .001.

Respondents were also asked to indicate the types of rewards and costs incurred in their sexual relationship. For men, the items most often identified as rewards were physical sensations from touching (90.6%), who initiates sexual activities (83.5%), and conceiving a child (81.2%). Women were most likely to indicate physical sensations from touching (90.6%), amount/type of foreplay (before intercourse or orgasm) (87.4%), and how often you experience orgasm (79.6%) as rewards. The most common costs for men were pleasing/trying to please your partner sexually (72.4%), degree of consideration your partner shows for your feelings (67.6%), and amount of time spent engaging in sexual activities (65.9%). Women's most common costs were conceiving a child (65.4%), degree of consideration your partner shows for your feelings (54.5%), and how easily you reach orgasm (52.9%).

Gender Differences

An ANOVA was used to compare men and women on their SOS scores. Results indicated that women were more erotophobic than men, F(1, 343) = 45.35, p [is less than] .001 (see Table 1).

A one-way MANOVA was conducted to examine gender differences in seven aspects of the sexual relationship: frequency of affectionate behavior, frequency of sexual behavior, sexual concerns, sexual problems for self, sexual problems for partner, relationship satisfaction, and sexual satisfaction. Results of the MANOVA were significant, F(7, 353) = 8.82, p [is less than] .001. Follow-up ANOVAs revealed that women reported significantly greater sexual satisfaction, F(1, 359) = 4.37, p [is less than] .05, and more sexual concerns, F(1, 359) = 4.60, p [is less than] .05, than did men. Men reported significantly more sexual problems for their partner than did women, F(1, 359) = 35.17, p [is less than] .001 (see Table 1).

To determine whether the men and the women differed on any of their sexual exchanges, a MANOVA examining gender differences in the level of rewards, level of costs, relative rewards, relative costs, reward equality, and cost equality was conducted. Results of this MANOVA were not significant, F(6, 354) = 1.70, p [is greater than] .10, indicating that there were no significant gender differences among these variables.

The MANOVA of gender differences in the number of rewards and costs incurred in the sexual relationship was significant, F(2, 358) = 5.07, p [is less than] .01. The mean number of rewards and costs are reported in Table 1. Men reported a significantly greater number of costs than did women, F(1,359) = 10. 05, p [is less than] .01. No gender differences for the number of rewards were found, F(1,359) = .60, ns.

Factors Related to Relationship and Sexual Satisfaction

Zero-order correlations were used to investigate the relationships between both relationship satisfaction and sexual satisfaction and several aspects of the sexual relationship (see Table 2). Results indicated that for both men and women, greater sexual satisfaction was associated with greater frequency of affectionate and sexual behavior, fewer sexual concerns, fewer sexual problems for self and partner, and greater relationship satisfaction. Erotophobia-erotophilia was not associated with sexual satisfaction for either men or women. Greater relationship satisfaction also was associated with greater frequency of affectionate and sexual behavior and fewer sexual concerns. However, only for men was greater relationship satisfaction associated with fewer sexual problems for partner and lower erotophilia. In contrast, only for women was greater relationship satisfaction associated with fewer sexual problems for self. Overall impact of life events was not associated with sexual satisfaction nor relationship satisfaction for men or women.

Table 2
Zero-order Correlations Between Relationship
Characteristics and Sexual Satisfaction and
Relationship Satisfaction for Men and Women

                                  Relationship Satisfaction
                                          Men     Women

Erotophobia-erotophilia                -.25(*)    -.07
Relationship satisfaction                --         --
Sexual satisfaction                     .70(**)    .57(**)
Frequency of physical affection         .37(**)    .50(**)
Frequency of sexual activity            .23(*)     .39(**)
Sexual concerns                        -.42(**)   -.46(**)
Sexual problems--self                  -.14       -.19
Sexual problems--partner               -.29(**)   -.09
Level of rewards                        .49(**)    .48(**)
Level of costs                         -.27(**)   -.02
Relative level of rewards               .50(**)    .51(**)
Relative level of costs                -.17(*)    -.06
Equality of rewards                     .15        .33(**)
Equality of costs                       .44(**)    .34(**)
Number of rewards                       .38(**)    .44(**)
Number of costs                        -.28       -.09
Impact of life events                   .09        .06

                                     Sexual Satisfaction
                                       Men        Women

Erotophobia-erotophilia               -.08         .14
Relationship satisfaction              .70(**)     .57(**)
Sexual satisfaction                     --          --
Frequency of physical affection        .50(**)     .53(**)
Frequency of sexual activity           .47(**)     .57(**)
Sexual concerns                       -.41(**)    -.50(**)
Sexual problems-self                  -.19        -.24
Sexual problems-partner               -.43(**)    -.17(*)
Level of rewards                       .66(**)     .73(**)
Level of costs                        -.26(**)     .03
Relative level of rewards              .66(**)     .69(**)
Relative level of costs               -.19(*)     -.13
Equality of rewards                    .22(*)      .34(**)
Equality of costs                      .48(**)     .32(**)
Number of rewards                      .55(**)     .65(**)
Number of costs                       -.32(**)    -.18(*)
Impact of life events                  .09         .14

Note: N = 361 (170 men and 191 women)

(*) p <.001

p (*) <.05

Zero-order correlations also were examined between the sexual exchange variables and both sexual satisfaction and relationship satisfaction (see Table 2). For both men and women, greater sexual satisfaction was associated with higher levels of rewards, higher levels of relative reward, and greater equality of rewards and costs. For men only, greater sexual satisfaction was associated with lower levels of costs and lower relative cost levels. An identical pattern of results was found for relationship satisfaction, with the exception that for men equality of rewards was not related to relationship satisfaction.

Zero-order correlations between the number of rewards and costs reported and sexual and relationship satisfaction indicated that for both men and women, greater sexual satisfaction was associated with a higher number of rewards and a lower number of costs. Similarly, greater relationship satisfaction was associated with a higher number of rewards for both men and women. However, only for men was greater relationship satisfaction associated with a lower number of costs. These results are reported in Table 2.

To explore which sexual concerns and problems were associated with sexual satisfaction, correlational analysis was used (see Table 3). To avoid inflating the Type 1 error rate, a Bonferroni correction was used, and alpha was set at p [is less than] .002. For both men and women, the following concerns and problems correlated negatively with sexual satisfaction: I feel "turned off," there is too little foreplay before intercourse, I have anorgasmia, and my partner has trouble maintaining sexual excitement. Only for women were the following concerns and problems negatively correlated with sexual satisfaction: I am unable to relax, I am not interested in sex, there is too little tenderness after intercourse, I like to do things my partner does not like, I have trouble getting excited, and my partner has an illness that interferes with the sexual response. Only for men were partner's trouble getting excited, partner's inhibited orgasm, and partner's anorgasmia negatively correlated with sexual satisfaction.

Table 3
Zero-order Correlations Between Specific Sexual Concerns and
Problems and Sexual Satisfaction for Men and Women

                                                 Sexual Satisfaction
Sexual Concerns                                      Men    Women

My partner chooses inconvenient times for sex       -.10     -.01
I am unable to relax                                -.081    -.22(*)
I am not interested in sex                          -.17     -.41(*)
I like to do things my partner does not like to do  -.16     -.23(*)
My partner asks me to do things I don't like to do  -.09      .13
I feel "turned off"                                 -.25(*)  -.40(*)
There is too little foreplay before intercourse      .27(*)  -.31(*)
There is too little tenderness after intercourse    -.22     -.24(*)
I am attracted to someone other than my partner     -.23     -.03

Sexual Problems                                       Self
                                                 Men     Women

Trouble getting sexually excited                -.22      -.31(*)
Trouble maintaining sexual excitement           -.12      -.01
Premature orgasm                                 .01       .18
Inhibited orgasm                                 .05      -.03
Anorgasmia                                      -.23(*)   -.31(*)
Medication(s) that interfere with sex           -.06       .03
An illness that interferes with sex              .02       .02

Sexual Problems                                     Partner
                                                 Men      Women

Trouble getting sexually excited               -.35(*)    -.21
Trouble maintaining sexual excitement          -.25(*)    -.22(*)
Premature orgasm                                .04        .12
Inhibited orgasm                               -.32(*)     .07
Anorgasmia                                     -.33(*)    -.10
Medication(s) that interfere with sex          -.002      -.05
An illness that interferes with sex             -.05      -.33(*)

Note: N = 361 (170 men and 191 women)

(*) p<.002

Age and Sexual Satisfaction

A number of respondents indicated that they felt that age was a crucial variable in sexual responding and sexual satisfaction. Zero-order correlations showed that older age was associated with less frequent physical affection (r = -.41, p [is less than] .001), less frequent sexual activity (r = -.41, p [is less than] .001), and lower sexual satisfaction (r = -.19, p [is less than] .001). Age did not correlate significantly with relationship satisfaction, number of sexual concerns, number of sexual problems for self, number of sexual problems for partner, or erotophobia-erotophilia.

Testing the IEMSS

To test the IEMSS, participants' scores for each component were entered into a hierarchical regression analysis predicting sexual satisfaction. Relationship satisfaction was entered on the first step and the sexual exchange components (REW-CST, [CL.sub.REW] - [CL.sub.CST], and [EQ.sub.REW] and [EQ.sub.CST]) were entered simultaneously on the second step. Relationship satisfaction (GMREL) accounted for 41% of the variance in sexual satisfaction, F(1, 359) = 244.65, p [is less than] .001. The addition of the sexual exchange components accounted for an additional 17% of the variance in sexual satisfaction, F-change 35.97,p [is less than] .001. Semi-partial correlations indicated that all sexual exchange components, REW-CST (sr = .49), [CL.sub.REW] - [CL.sub.CST] (sr = -.27), and EQREW (sr =.14) and [EQ.sub.CST] (sr = . 10) contributed uniquely to the prediction of sexual satisfaction. The final regression equation accounted for 58% of the variance in sexual satisfaction, F(5, 355) = 96.77, p [is less than] .001. As predicted, all components of the model added uniquely to the prediction of sexual satisfaction.

It was proposed that in predicting sexual satisfaction, the model would be robust to the effects of gender. In other words, gender was not expected to add to the model, nor to interm act with the components of the model. As in the previous analysis, the IEMSS components were entered as the first two steps. Gender was entered on the third step, and the interaction terms between gender and the IEMSS components (gender by GMREL; gender by REW-CST; etc.) were entered simultaneously on the fourth step. Gender added significantly but not meaningfully to the amount of variance accounted for in sexual satisfaction, [R.sup.2]-change = .005, F-change = 4.05, p [is less than] .05. Women reported greater sexual satisfaction than men did (sr = .11). As predicted, the interaction terms did not significantly improve the prediction equation, [R.sup.2]-change .01, F-change = 1.65, ns, indicating that the IEMSS worked equally well for women and men.

A final regression analysis was used to determine whether components of the model continued to predict sexual satisfaction after other variables found to be related to sexual satisfaction were controlled. Age, frequency of affectionate behavior, frequency of sexual behavior, number of sexual concerns, and number of sexual problems for self and partner were entered simultaneously prior to entry of the IEMSS components. The IEMSS components were entered hierarchically on the next two steps. The first step accounted for 47% of the variance in sexual satisfaction, F(6, 354) = 51.37, p [is less than] .001. As predicted, after taking the control variables into account, relationship satisfaction (GMREL) continued to add significantly to the prediction of sexual satisfaction, [R.sup.2]-change = .11, F-change = 90.82, p [is less than] .001. The sexual exchange components also added uniquely to the prediction of sexual satisfaction, [R.sup.2]-change = .06, F(11,349) = 13.98, p [is less than] .001. The final regression equation accounted for 63% of the variance in sexual satisfaction, F(11, 349) = 54.85, p [is less than] .001. As predicted, the components of the model added uniquely to the prediction of sexual satisfaction over and above other factors related to sexual satisfaction.

Discussion

Results of this study provide very good support for the validity of IEMSS for predicting the sexual satisfaction of married Chinese men and women. As predicted, all components of the model (GMREL; REW-CST, [CL.sub.REW]-[CL.sub.CST], and [EQ.sub.REW] and [EQ.sub.CST]) were significantly related to sexual satisfaction. The more rewards exceeded costs, relative rewards exceeded relative costs, one's levels of rewards and costs were perceived to be equal to one's partner's, and the more relationally satisfied respondents were, the greater the level of sexual satisfaction. Together, the components of the IEMSS accounted for 58% of the variance in sexual satisfaction. In addition, as expected, the IEMSS was robust to the effects of gender in that it can be used to account for the sexual satisfaction of both men and women. The validity and utility of the IEMSS were further demonstrated in that each component of the model continued to add to the prediction of sexual satisfaction when a number of other characteristics related to sexual satisfaction were controlled.

Although we provided support for the utility of the methods and measures used to assess factors that affect sexual satisfaction in a sample of married Chinese men and women, some cautions regarding the generalizability of our findings are warranted. For instance, 54% of individuals who were invited to participate in our study did not return their questionnaires. Individuals who participate in a sexual survey may differ in significant ways from individuals who refuse to participate in such a survey. Nonetheless, our response rate was better than many similar mailed surveys. As all respondents were high school graduates living in Beijing or Shanghai, the results cannot be generalized to individuals with lesser education or those living in rural areas. In particular, our respondents are likely to be less traditional and have greater sex knowledge than Chinese individuals with less education and living in non-urban settings. In addition, respondents completed a translation of a questionnaire that has been used successfully only within a Western culture. Its validity for studying sexuality within this Eastern culture is not known. In fact, written comments by some respondents suggest that some concepts (e.g., "rewards" and "costs") did not translate well into Chinese or required a great deal of sophistication from the respondent to be fully understood. Older individuals were more likely to have been dropped from the sample because of missing data resulting from difficulty understanding some items. Although this may have an impact on the generalizability of our findings, our measures were nonetheless found to have high reliability, and 11.6% of the sample were older than 50.

Characteristics of the Sexual Relationship

The current study also provided descriptive data regarding the overall and sexual relationships of married Chinese men and women. The results indicate that married Chinese men and women are generally moderately satisfied with their overall relationship and their sexual relationship with their spouse. They reported that they engage in physically affectionate behavior and sexual activity on average between once and twice a week and two or three times a month. They find their sexual relationships to be more rewarding than costly, although both their levels of rewards and costs are moderate. The level of rewards and costs that they incur in their sexual relationship are comparable to their expectations and are considered moderately equal to those of their partners.

Although the frequency of physical affection in the current sample is less than that of a similar Canadian sample of married men and women (Lawrance & Byers, 1995), the frequency of sexual behaviors is comparable to that found in the Canadian sample, as well as the Zhou (1993) sample. Respondents indicated that they engaged in both sexual and affectionate behavior with a similar frequency. Hence it appears that physical affection is mostly expressed within the context of sexual exchange. This may be particularly true of older individuals. In the past, expression of affectionate behavior in public was officially frowned upon and rarely observed. In fact, prior to the early 1980s even holding hands in public was considered to be deviant behavior or a social transgression (Pan, 1989b). Individuals were taught that love should be kept within an individual's heart, rather than expressed overtly through affectionate behavior such as hugging, kissing, and cuddling. This does not mean that Chinese couples do not care about each other. Rather, they may express their caring through looks and body language instead of physical affection. The finding that older individuals reported less frequent affectionate and sexual behavior than did younger individuals suggests that this cultural norm has changed somewhat. In addition, at least in our sample, more frequent physical affection was related to higher relationship satisfaction and higher sexual satisfaction, and younger individuals reported greater sexual satisfaction.

Descriptive data on the frequency of sexual concerns and problems among married Chinese individuals were also provided in the current study. The majority of the sample indicated that they had experienced at least one sexual concern within the last three months. The most frequent sexual concerns were too little foreplay before intercourse, too little tenderness after intercourse, and feeling turned off. In addition, two thirds of the sample indicated that they had experienced at least one sexual problem within the last three months. The most frequently reported concerns and problems by this Chinese sample were similar to those reported by a Canadian sample (Lawrance, 1994). By both the reports of men and women, the most frequently experienced sexual problems for women were trouble getting sexually excited and inhibited orgasm, whereas the most frequently experienced sexual problems for men were premature orgasm and maintaining sexual excitement.

These data suggest that the sex lives of Chinese men and women could be enhanced. Yet, in keeping with cultural norms, although the majority of respondents reported having a sexual concern or problem, fewer than 2% of respondents had sought counseling for sexual problems. The findings of only moderate sexual satisfaction and sexual rewards as well as moderate sexual costs and prevalent sexual concerns and problems are not unexpected, given the absence of comprehensive sex education, available sex counseling, and public discussion of sexual behavior and sexual satisfaction in the past. As the rewards and costs as well as sexual concerns and problems were found to be related to sexual satisfaction, it is likely that the increasing openness to sex education and public discourse on sexuality issues in China will have a positive impact on individuals' sexual satisfaction and sexual functioning. The following comments made by participants in the current study suggest that at least this fairly well-educated sample of Chinese individuals are aware of the negative impact that the policy of limiting public discourse about sex has had on them personally and on the society in general:

For a long time, Chinese people

have said nothing about the topic

of sex, but in practice there are

problems everywhere. I think such a

survey is very significant.

It is very important to raise the

quality of our Chinese people's

marriage, family, and sexual life.

Following the improvement of our

Chinese people's everyday life, it is

necessary to do research on

marriage and sexual problems.

Gender Differences

We predicted that Chinese women would report lower sexual satisfaction and would describe their relationship as less rewarding, more costly, and less equal than would Chinese men. In general, these predictions were not supported. Contrary to predictions, women in this study reported greater levels of sexual satisfaction than did men and equal levels of rewards and costs, relative rewards, relative costs reward equality, cost equality and number of rewards.

These results may reflect men's and women's different expectations from their sexual relationship. Because women, but not men, in China have been socialized not to expect pleasure from sex, it is possible that the exchanges they experience in their sexual relationship are mediated by such expectations. For example, if a woman does not expect her sexual relationship to be at all satisfying/rewarding and yet does receive some sexual satisfaction, she may rate her relationship as more satisfying/rewarding than a man experiencing equal satisfaction but expecting his sexual relationship to be very satisfying. In the current study, respondents were not asked the level of rewards and costs or the level of sexual satisfaction that they expected in their sexual relationship but rather how their rewards and costs compared to their expectations. Therefore, it was not possible to ascertain how rewarding and costly respondents expected their sexual relationships to be. Further research is needed to investigate the impact of expectations on perceived reward and cost level, relative reward and cost level, perceived equality of rewards and costs, and sexual satisfaction.

Also contrary to predictions, men reported a significantly greater number of costs than did women. Elsewhere, we examined gender differences in the types of rewards and costs reported by this sample of Chinese men and women (Renaud, Byers, & Pan, 1996a). Although for the most part men and women reported similar types of rewards, generally men were more likely than women to report as costs the items that reflected their partners' participation in the sexual exchange. These results suggest that a lack of partner response and the effort required to get a female partner to respond sexually may be important sexual costs for men, or at least for these educated, less traditional men who expect a mutually pleasing sexual relationship. These findings may reflect the cultural message to women within China that women should not respond to their husband's sexual stimulation of them. This interpretation is also supported by the findings in the current study that women were more erotophobic and reported more sexual concerns related to women's sexual functioning than did men. For example, women were more likely than men to report that they feel "turned off" and are not interested in sex. In line with this, men were more likely than women to report that their partner had a sexual problem.

Factors Related to Sexual and Relationship Satisfaction

We found that greater sexual satisfaction was associated with greater relationship satisfaction. A similar finding, using different measures, was reported by Pan (1989a) in a survey of 977 married individuals living in Beijing that was published in the Chinese language only. It is also consistent with previous North American research (Henderson-King & Veroff, 1994; Lawrance & Byers, 1995; Oggins et al., 1993). In addition, as predicted, greater frequency of both affectionate and sexual behavior was associated with greater sexual and relationship satisfaction. Consistent with previous researchers (Frank et al., 1978), we found a weak but significant relationship between both one's and one's partner's sexual problems and sexual satisfaction. Although fewer sexual problems for self was associated with greater sexual satisfaction for both men and women, only for women was it associated with greater relationship satisfaction. In contrast, only for men was greater relationship satisfaction associated with fewer sexual problems for partner. Therefore, it appears that the sexual difficulties of women play a greater role in the relationship satisfaction of men and women than do the sexual difficulties of men. Further research investigating the differential impact of men's and women's sexual functioning on sexual and relationship satisfaction is needed.

The sexual exchanges of married Chinese men and women also were found to be associated with sexual satisfaction and relationship satisfaction. As predicted, sexual satisfaction was associated with higher levels of rewards, higher levels of relative rewards, and greater interpersonal equality of rewards and costs. In addition, greater sexual satisfaction was associated with a higher number of rewards and a lower number of costs for both men and women. However, only for men was sexual satisfaction associated with lower levels of costs and lower levels of relative cost. With the exception of number of costs and equality of rewards, an identical pattern of results was found for relationship satisfaction. Greater relationship satisfaction was associated with lower numbers of costs only for men. This pattern of results appears to suggest that costs incurred in the sexual relationship are more detrimental to the relationship satisfaction and sexual satisfaction of men than of women. Elsewhere, we examined which specific rewards and costs correlated with sexual satisfaction in this sample of married Chinese men and women (Renaud, Byers, & Pan, 1996b). Not only did we find that a greater number of specific costs correlated with sexual satisfaction for men than for women, but that a number of factors have an impact on the sexual satisfaction of men if they are perceived as a cost but have little impact if they are perceived as a reward.

The Future of Sexual Satisfaction Research in China

This study provided support for the utility of the IEMSS in a sample of married Chinese men and women. However, the amount of variance in sexual satisfaction accounted for by the IEMSS components in the current sample (58%) was less than that found in a similar study with a Canadian sample (79%) (Lawrance & Byers, 1995). Therefore, actors in addition to the IEMSS appear to contribute to sexual satisfaction in the relationships of married Chinese individuals. Some of these may be environmental/situational. Although we attempted to assess those life events that might have affected sexual satisfaction, the impact of life events was not found to be related to sexual or relationship satisfaction. Future researchers need to assess life situations within Chinese culture that are most likely to affect sexual satisfaction, rather than use a more general measure of life events. Future researchers also should examine the relationships between nonsexual exchanges within couples and sexual satisfaction. Specifically, any contribution of the addition of relationship exchange variables to the IEMSS needs to be explored. Finally, as many as half of Chinese couples marry, not for love, but to get married (Pan, 1993a). It is not known whether the interactions of these individuals differ from those individuals who experience the more affective or passionate aspects of a romantic relationship. The role that love and romantic passion plays in the sexual relationships and sexual satisfaction of Chinese individuals needs to be explored.

References

Byers, E. S., Demmons, S., & Lawrance, K. (in press). Sexual satisfaction within dating relationships: A test of the Interpersonal Exchange Model of Sexual Satisfaction. Journal of Social and Personal Relationships.

Demmons, S., Lawrance, K., & Byers, E. S. (1993, June). The relationship between sexual communication and sexual satisfaction. In E. S. Byers (Chair), Testing common assumptions about human sexuality. Symposium presented at the meeting of the Canadian Psychological Association, Montreal.

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(2), 357-394.

Fisher, W. A., Byrne, D., White, L., & Kelley, K. (1988). Erotophobia-erotophilia as a dimension of personality. The Journal of Sex Research, 25, 123-151.

Frank, E., Anderson, C., & Rubenstein, D. (1978). Frequency of sexual dysfunctions in "normal" couples. New England Journal of Medicine, 299, 111-115.

Gil, V. E. (1993). Sexology in contemporary China: Some reflections on an emergent discourse. Sexology delegation to the People's Republic of China citizen ambassador program-PTP, International/ Chinese Medical Association, 1-11.

Hatfield, E., & Rapson, R. L. (1993). Love, sex and intimacy: Their psychology, biology, and history. New York: Harper-Collins.

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

Herold, E. S., & Byers, E. S. (1994). Sexology in China. The Canadian Journal of Human Sexuality, 3, 1263-269.

Hudson, W., Harrison, D., & Crosscup, P. (1981). A short-form scale to measure sexual discord in dyadic relationships. The Journal of Sex Research, 17, 157-174.

Lawrance, K. (1994). Development and validation of the interpersonal exchange model of sexual satisfaction in long-term heterosexual relationships. Doctoral dissertation submitted to the University of New Brunswick, Fredericton, NB.

Lawrance, K., & Byers, E. S. (1995). Sexual satisfaction in long-term heterosexual relationships: The Interpersonal Exchange Model of Sexual Satisfaction. Personal Relationships, 2, 267-285.

Lawrance, K., & Byers, E. S. (in press). Interpersonal Exchange Model of Sexual Satisfaction Questionnaire. In C. Davis, W, L. Yarber, R. Baurerman, G. Schreer, & S. Davis (Eds.), Sexuality related measures: A compendium (2nd ed.). Thousand Oaks, CA: Sage.

Morokoff, P., & Gilliland, R. (1993). Stress, sexual functioning, and marital satisfaction. The Journal of Sex Research, 30, 43-53.

Oggins, J., Leber, D., & Veroff, J. (1993). Race and gender differences in black and white newlyweds' perception of sexual and marital relations. The Journal of Sex Research, 30, 152-160.

Pan, S. (1989a). Marriage and sexuality in current Beijing City. In Beijing Marriage in the Late 1980's. Beijing: Beijing City Government.

Pan, S. (1989b). Quantitative behavioral analysis of public heterosexual petting in Chinese public parks. Sexuality in Asia, 173-184.

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

Pan, S. (1993b). China: Acceptability and effect of three kinds of sexual publication. Archives of Sexual Behavior, 22, 59-71.

Pan, S. (1994). Chinese wives: Psychological and behavioral factors underlying their orgasm frequency. Unpublished manuscript, Renmin University, Beijing.

Poulin, C. (1992). Towards a multidimensional and multidirectional model of female sexual arousal. The Canadian Journal of Human Sexuality, 1, 129-132.

Renaud, C., Byers, E. S., & Pan, S. (1996a). Factors related to sexual satisfaction in Mainland China. Canadian Journal of Human Sexuality, 5, 243-252.

Renaud, C., Byers, E. S., & Pan, S. (1996b, November). Sexual satisfaction and relationship satisfaction in Mainland China. Paper presented at the meeting of the Society for the Scientific Study of Sexuality, Houston, TX.

Sarason, I., Johnson, J., & Seigel, J. (1978). The Life Experiences Survey. Journal of Consulting and Clinical Psychology, 46, 932-946.

Tiefer, L. (1988). A feminist critique of the sexual dysfunction nomenclature. New York: Haworth.

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

Address correspondence to E. Sandra Byers, Ph.D., Department of Psychology, University of New Brunswick, Fredericton, NB, Canada E3B 6E4. Telephone: [phone]. Fax: [phone].E-mail:[e-mail].

This research was conducted in partial completion of the requirements for the doctorate degree of the first author with the supervision of the second author. The authors thank Kelli-an Lawrance for her work in designing the study and Kang Li and Min Wang for their help with the translation of the questionnaire. We also appreciate the helpful comments of Guy Grenier, Lisa Price, Joan Wright, and Pamela Yates on an earlier version of this manuscript. Requests for reprints should be sent to E. Sandra Byers, Ph.D., Department of Psychology, University of New Brunswick, Fredericton, NB, Canada E3B 6E4.

Manuscript accepted July 28, 1997