Perceived benefits and costs of romantic relationships for women and men: Implications for exchange theory

This investigation examined the perceived benefits and costs of romantic (i.e., reciprocal dating) relationships. In Study 1, subjects provided open-ended reports regarding the benefits and costs associated with romantic involvement. Different groups of subjects ranked (Study 2) and rated (Study 3) these benefits and costs for importance. Companionship, happiness, and feeling loved or loving another were among the most important benefits accompanying romantic involvement. The most serious costs included stress and worry about the relationship, social and nonsocial sacrifices, and increased dependence on the partner. Compared to males, females regarded intimacy, self-growth, self-understanding, and positive self-esteem as more important benefits, and regarded loss of identity and innocence about relationships and love as more important costs. Alternatively, males regarded sexual gratification as a more important benefit, and monetary losses as a more serious cost than did females. Implications for exchange theory are highlighted. Involvement in romantic (i.e., reciprocal dating) relationships is considered one of the most important tasks of early adulthood and is assumed to change the individual in profound ways (Erikson, 1963; Levinson, Darrow, Klein, Levinson, & McKee, 1978). Surprisingly, however, little is known about individuals' subjective perceptions of changes that romantic relationships presumably bestow upon them. The present investigation explored these changes by assessing the perceived benefits and costs that romantic relationships entail for the individual. The investigation extended prior reWe thank three anonymous reviewers for their constructive comments on an earlier draft. We also thank Eric Capaul, Kirstin Carlson, and Jodie Rapkin for their assistance with data collection and coding. Correspondence concerning this article should be addressed to Constantine Sedikides, Department of Psychology, The University of North Carolina at Chapel Hill, Davie Hall, CB 3270, Chapel Hill, NC 27599-3270. search in this area by empirically exploring the content of costs and benefits (rather than their regulation only), by focusing on romantic relationships specifically (rather than on general relationships), and by examining impacts on individuals' perceptions (rather than on general relationship satisfaction). Because of the possibility of differential consequences of romantic involvement for women and men, this investigation additionally examined gender differences) in relationship benefits and costs. 1. A great deal of debate has taken place over the appropriate use of the terms "gender differences" versus "sex differences" (e.g., Deaux, 1993; Gentile, 1993; Unger, 1979; Unger & Crawford, 1992, 1993). Following the suggestion of Unger (1979), we opted for the term "gender differences," given that "the use of the term gender makes it less likely that psychological differences between males and females will be considered explicable mainly in terms of physiological differences between them" (p. 1093; italics in the original). 6 C. Sedikides, M.B. Oliver, and W.K. Campbell Consequences of Romantic Involvement Past theorizing and research Romantic involvement clearly encompasses benefits and costs, as recognized by several theoretical formulations adopting an exchange perspective (Adams, 1965; Aronson, 1969; Blau, 1967; Homans, 1961; Kelley & Thibaut, 1978; Lott & Lott, 1974; Mills & Clark, 1982). Consider a study by Clark (1981), which investigated subjects' beliefs regarding the comparability of benefits and costs that result from exchange versus communal relationships. Clark re-ported that comparability of benefits led to the inference that an exchange relationship existed, whereas noncomparability of bene-fits led to the inference that a communal relationship was present. As is evident from the description of this study, exchange perspectives focus on rules and expectations that regulate the giving and receiving of benefits rather than the subjectively construed content of benefits and costs. Furthermore, exchange perspectives are applicable to social relationships in general rather than being exclusively concerned with romantic relationships. Similar appraisals apply to the role expectations perspective (Parsons & Shils, 1951), the social motives perspective (MacCrimmon & Messick, 1976; Schultz & May, 1989), and the theory of relational models (Fiske, 1991; Fiske, Haslam, & Fiske, 1991). Foa and Foa's (1974, 1980; see also Foa, Converse, Tornblom, & Foa, 1993) resource exchange theory does focus on the content of the exchange by identifying six resources: love, status, information, money, goods, and services. However, this theory was also intended to be a general theory of social rela tionships rather than being concerned with the specifics of romantic relationships. In addition, this theory and relevant research does not generally approach its subject mat-ter from the viewpoint of subjective construals of relationship benefits and costs (for an exception, see Rettig, Danes, & Bauer, 1993). Some investigations have concentrated on the content of exchange resources as applied to romantic (mostly marital) relation-ships. For example, Safilios-Rothschild (1976) reported seven resources that are potentially exchanged between spouses: socioeconomic, affective, expressive, companionship, sex, services, and power in relationships. Sprecher (1985) modified the Safilios-Rothschild list by replacing the resource "power in relationships" with the re-sources "physical appearance" and "intellectual." However, these lists were based on researcher intuition rather than being de-rived from subjects' own accounts, and the perceived importance (and, most notably, the perceived relative importance) of the exchange resources for the individual was not tested empirically. Van Yperen and Buunk (1990) created a list of 144 "exchange elements," later reduced to 24 through factor analysis. Subjects were then asked to "indicate whether each of these elements was a positive, negative, or no contribution to an intimate relationship" (p. 293), and these ratings of the exchange elements were then compared to a global assessment of equity. These researchers found that the most important positive contributions to a relationship were being commit-ted to it, being sociable and pleasant to be with, leading an interesting and varied life, and taking care of the children. The most negative contributions to a relationship were being suspicious and jealous, being addicted to tobacco and/or alcohol, and being unfaithful. Although Van Yperen and Buunk's study provided useful information concerning contributions to relationships, their research differs from the present investigation in two primary ways. First, Van Yperen and Buunk were interested in contributions of exchange elements to relation-ship satisfaction rather than the benefits and costs that relationships incur upon the individual. Second, the list of exchange elements employed in their study was researcher-derived rather than subject-derived. Sternberg and Wright (described in Sternberg, 1987) divided their sample into Perceived benefits and costs 7 three groups depending on the length of their current intimate relationship (short-term, medium-term, and long-term) and asked subjects (among other questions) about "what attributes of a relation-ship . . . males and females view as important for the success of an intimate relation-ship at different points in the relationship." Sternberg and Wright's results are rather difficult to evaluate, however, because the description of their study and findings was unusually brief. For example, no explanation was provided as to how the attributes were derived from subjects' reports, what scales subjects used to rate these attributes, and how the data were analyzed. In addition to the aforementioned studies, several theoretical statements have also attempted to delineate the content of benefits and costs that accompany romantic relationships. Hypothesized benefits of romantic involvement include assistance in everyday tasks, coping with stressful life events, material support, suggestions for effective behavioral regulation, and satisfaction of crucial psychological needs such as intimacy, power, social integration and alliance, being nurturant, and reassurance of one's own worth (Duck, 1991; McAdams, 1984, 1988; Rook & Pietromonaco, 1987; Weiss, 1969,1974). Hypothesized costs of romantic relationships include ineffective or excessive help, unwanted or unpleasant in teractions such as invasions of privacy and criticisms (Rook & Pietromonaco, 1987), fear of exposure, fear of abandonment, angry attacks, and loss of individuality (Hat-field, 1984). However, these theoretical statements (which were based on researcher intuition, clinical observation, or both) were not followed up by empirical tests. Contributions of the present investigation The present investigation extends past theorizing and research in several noteworthy ways. First, this investigation focuses on subjective construals of relationship benefits and costs. Subjective construals are operationalized as subjects' retrospective (and, unavoidably, reconstructive) reports of the benefits and costs associated with their past and present relationships. Second, the investigation is concerned exclusively with romantic relationships. Third, the investigation is interested in the content of benefits and costs. Although past research incorporated a subset of the above three contributions, not a single study has subsumed all three contributions simultaneously. Finally, this investigation adds uniquely to the literature in going beyond a simple list of benefits and costs and instead focusing on the relative importance of these benefits and costs for the individual. Research has yet to identify what benefits and costs are most and least important to romantically involved individuals. As mentioned above, the current investigation is concerned with the content and relative importance of perceived benefits and costs. Why is this endeavor worthwhile? We think it is worthwhile for several reasons. One reason is that this concern addresses a serious analytical shortcoming of exchange theory—namely the failure to specify the conceptual dimensions of benefits and costs. In much of the exchange literature, benefits and costs are specified on an ad hoc or studyby-study basis. Arguably, then, exchange theory remains tautological and thus incapable of empirical falsification. Advances in operationalizing (and interpreting others' operationalizations of) benefits and costs are clearly needed. A second reason is that knowledge of benefits and costs of romantic involvement may increase understanding of the expectancies that individuals bring to relationships, which may in turn affect relationship satisfaction. A third reason is that this endeavor may contribute to a better un2. Berg and his colleagues (Berg & McQuinn, 1986; Berg,Piner,& Frank,1993) developed a model composed of four hypotheses that predict resource ex-change in close and nonclose relationships. Hypothesis 2 states that the resources exchanged in close relationships will be more particularistic than the resources exchanged in nonclose relationships. The present investigation can be thought of as an at -tempt to find out what exactly those particularistic resources are. 8 C. Sedikides, M.B. Oliver, and W.K. Campbell derstanding of why individuals enter and re-main in romantic relationships. Stated otherwise, this endeavor will address empirically the link between motivation and romantic relationships. Finally, this endeavor is likely to have applied implications; that is, the results may provide useful information such for professionals as marital therapists and policy planners in terms of the benefits and costs that currently matter in our culture. Gender Differences in Perceived Benefits and Costs Undoubtedly, accounts of perceived bene-fits and costs of romantic relationships will vary among individuals. Predicting the rela tive importance of benefits or seriousness of costs, therefore, is a complex task involving numerous variables, some of which may be difficult to identify. Nevertheless, research on romantic and sexual attitudes and beliefs suggests that gender may be one such predictor. Incorporating gender as an attribute variable can easily raise controversy: Gender is often confounded with other variables, it is difficult to determine the causes behind any apparent gender differences, and it is statistically impossible to prove a null hypothesis that asserts gender similarities (for a review, see Jacklin, 1981). Despite these difficulties, though, an interest in the question of gender differences on the part of scholars has led to a substantial body of past research in the area of romantic beliefs and experiences. Males score higher than do females on measures of romanticism (e.g., love at first sight) and females approach relationships with a more pragmatic orientation than do males (Fengler, 1974; Peplau & Gordon, 1985; Rubin, Peplau, & Hill, 1981; Sprecher & Metts, 1989). Conversely, males score higher on measures that indicate "game-playing" or "uncommitted" orientations to romantic relationships (e.g., preferring freedom to interact with multiple partners), whereas females report higher levels of "passion" once a relationship has developed and experience more intense positive and negative symptoms or emotional responses (Dion & Dion, 1985; Hatfield, 1983; Hendrick & Hendrick, 1987; Wood, Rhodes, & Whelan, 1989). These gender differences in romantic be liefs and experiences may be associated with differences in the importance that females and males place on intimacy and sexuality. If women place greater emphasis on intimacy than do men, one would expect women to experience greater feelings of euphoria when intimacy is achieved, and greater feelings of disappointment when intimacy is endangered. Alternatively, if men place greater emphasis on sexuality than do women, men should experience greater satisfaction and disappointment in reaction to sexual gratification and frustration, respectively. Consistent with these predictions, Levinger (1964; see also Sprecher & Sedikides, 1993) reported that communication was a stronger predictor of global marital satisfaction for wives than for husbands, whereas sexual satisfaction was a stronger predictor for husbands than for wives. Gender differences in the importance of intimacy and sexuality have been illustrated in studies concerning emotional contexts in which sexual activity occurs. In general, it appears that whereas men experience and enjoy sexual intercourse in relationships at various levels of emotional commitment, women's sexual activities tend to be experienced and enjoyed most often in relation-ships involving commitment or, at least, emotional attachment (McCabe,1987; Sack, Keller, & Hinkle, 1981). Based on their survey of college students regarding sexual attitudes and behaviors, Carroll, Volk, and Hyde (1985) concluded that "Male motives more often include pleasure, fun, and physical reasons, whereas females' motives include love, commitment, and emotion" (p. 136). Gender differences in orientations to-ward romantic relationships can also be found in research examining the attributes of romantic partners that males and females rate as important. Overall, this literature Perceived benefits and costs 9 supports the idea that females tend to place a greater emphasis on intimacy than do males (e.g., rating qualities such as "warmth" as more important) and to be more pragmatic in their orientations (e.g., rating qualities such as "financial security" as more important), and that males tend to place a greater emphasis on sexuality than do females (e.g., rating "physically attractive" or "sexy" as more important; Buss & Barnes,1986;Dindia & Allen,1992;Howard, Blumstein, & Schwartz, 1987; Laner, 1977; Nevid, 1984; Simpson & Gangestad, 1992). Based on the above literature, several predictions can be generated. First, gender differences in intimacy suggest that females may be more likely than males to consider love, communication, affection, and in-creases in love, sharing, or understanding as important benefits of involvement. Alternatively, gender differences in sexuality and "game-playing" orientations imply that males may be more likely than females to regard sexual activity or sexual gratification as more important benefits, and monogamy or lack of freedom as more serious costs. Second, given females' higher scores on measures of passionate experiences, one might also expect that females would perceive feelings of passion or joy as more important benefits than males, with these benefits, in turn, likely to be associated with perceived positive changes in selfevaluation (Long,1989). On the other hand, these gender differences on symptoms of love may exacerbate gender differences in anxiety in response to relational difficulties, leading females to perceive the cost of worry about the relationship as more serious, possibly associated with perceived negative changes in self-evaluation.