Research Support School of Business Administration DISMANTLING A STEREOTYPE: AN EMPIRICAL TEST OF UTILIZATION OF A DATING SERVICE Working Paper #722 Aaron Ahuvia University of Michigan December 1993

I I I I

Dismantling a Stereotype: An Empirical Test of Utilization of a Dating Service* Mara B. Adelman. Ph.D. Northwestern University Department of Communication Studies Aaron Ahuvia University of Michigan School of Business Administration Working Paper Mailing address: Prof. Aaron Ahuvia, Ph.D. University of Michigan School of Business Administration Ann Arbor, MI 48109-1234 (313) 764-6174 *This study was supported by the Office of Research and Sponsored Programs, Northwestern University and the Jewish Community Center of Greater Chicago. Portions of this paper were previously presented at the 1990 annual conference of the Association for Consumer Research, October 4-7 in New York.

a ABSTRACT This paper compares two models of who uses introductory services. The "deficiency model" holds that people join dating services because they have social and psychological deficiencies what prevent them from establishing romantic relationships through conventional channels. The "selectivity model" posits that users of introductory services have high standards for a mate and therefore have difficulty finding people who meet their criteria through traditional social channels. Results supported the selectivity hypothesis and strongly rejected the deficiency model. Key terms: courtship, mate selection, dating, stigma, stereotypes, introductory services.

"Lonely and desperate" (description of people who use dating services, Wall St. Journal, 1989) "I thought it would be a nerd festival" (client of an introductory service) "The first thing everyone wants to know about the service when they call is 'what are the members like?"' (matchmaker) Deeply embedded in the American ethos of finding "Mr. or Ms. Right" lies the romantic model -- an image of a spontaneous, random encounter and instantaneous, mutual attraction. Effortless, this model is imbued with such metaphorical imagery as "magic" and "chemistry" (Ahuvia & Adelman, in press). However, the rise of intermediary channels such as personal ads, video dating, and matchmaking services suggests that the romantic model has not proven sufficient for many singles. As alternatives for meeting potential mates, these channels represent a shift from the spontaneous to the strategic, from implicit intentions to explicit purposes, and from chance encounters to planned meetings. Furthermore, several trends suggest that these services may become an integral part of the American scene in the years ahead. Projected demographics reveal an increasing number of singles in the aging baby boom generation (Bureau of the Census, March 1987; April 1987). This generation often has extensive work commitments, suggesting minimal social time for finding a selective, compatible mate. Finally, this generation engages in increased use of personal services, made possible by high levels of professional and economic attainment. Despite the increased popularity of these formal social intermediaries, users of these channels are often stigmatized. As the above quotes suggest, clients of intermediary channels are portrayed as "losers" -- persons who have a social, psychological, or behavioral deficiency that prevents them from establishing romantic relationships through conventional channels (Darden & Koski, 1988). They may also be seen as too eager to form a relationship (i.e., desperate). Research by Bailey and Garrou (1983) has shown that people believed to be very eager to date are seen as selfish, untrustworthy, and undesirable dating partners. We believe that self-image and status considerations weigh heavily on the choice of whether to use a formal intermediary or to limit oneself to informal channels (Ahuvia & Adelman, 1992). Apart from blind dates, most traditional channels for meeting dating partners are informal and "semi-covert." By this we mean that they serve other purposes in addition to helping singles meet (e.g., parties, bars, health clubs, political organizations, etc.). Singles using these semi-covert channels to meet dating partners can plausibly claim (to themselves and others) that they were there for other reasons. In contrast, "overt channels" such as introductory services offer only one plausible account for their utilization -- an explicit desire to meet a potential partner. Because these channels deny people "causal ambiguity" (Bernstein, Stephenson, Snyder, and Wicklund 1983), fears of appearing "desperate" or beliefs that only "losers" must resort to these explicit efforts may be damaging to clients' self-image. Thus, the overt nature of intermediary channels is often inconsistent with singles' self-image and their projected image to others. This inconsistency results in the stigma that is attached to intermediary channels (see discussion on stigma and self-concept; Jones et al., 1984; pp. 111-154). Underscoring this stigma are sensationalized media portrayals of these introductory services. Efforts to capitalize on their novelty have been grist

2 for recent films (e.g., Sea of Love, Crossing Delancy), newspaper articles (Brotman, 1993; Freedman, 1989; Ryan 1993), magazines stories (Bennet, 1989; and Brand, 1988; Jannot, 1993; Howard, 1992), and publicized TV talk shows such as "20/20" and "Phil Donahue." In some cases, portrayals by the popular media have perpetuated the negative stereotype of client utilization. For example, in the film Crossing Delancy, while the leading couple was introduced through an "old world" style Jewish matchmaker, their relationship only became viable after they learned that neither one of them had actually joined the matchmaking service. On the other hand, Austrom, de Sousa, and Hanel (1980) have shown that the number of singles ads placed in a Canadian singles publication increased following a positive article in the Toronto Star. We have found that media portrayals tend to make reference (implicitly or explicitly) to the negative stereotype of the people who use these services, even if it is in the context of denying the stereotype's validity. At the same time, they may increase utilization of these services because (a) some of the press attempts to refute these stereotypes, and (b) "any publicity is good publicity." Despite this widely noted stigma, the past decade has seen a meteoric rise in both the number of introductory services for mate-seeking (Adelman and Ahuvia, 199 a) and the level of academic interest in this phenomenon (Ahuvia & Adelman, 1992; and Woll and Cozby, 1987 for reviews). The growth of these innovative channels calls for an investigation of the accuracy of client stereotypes. Furthermore, understanding the client base of these services is particularly important to family practitioners who may wish to refer clients to these services. Two previous pieces of research have investigated who uses formal overt intermediaries. Darden and Koski (1988) used ethnographic methods to explore the perceptions of singles ad users that meeting people in this way was a deviant activity. This research attests to the strength of the stigma as evidenced by the concern of singles ad users with justifying their participation and their fears that the other users would be "losers" (p. 388). While their methodology did not allow for quantification, the researchers descriptions tended to contradict to the common negative stereotypes. Furthermore, Darden and Koski found that "even those (singles ad users) who had not established any serious relationships through the ads... became aware that there were many good, intelligent, attractive older single people around" (p. 395). In contrast to these findings, a survey conducted by Goodwin (1990) found that members of a British dating service were more educated and held higher status jobs than the average British citizen, yet the dating service members scored lower on a dating skills and assertiveness inventory than 3 student comparison groups. Goodwin (1990) and the popular stereotype could be seen as proposing a "deficiency model" to explain the use of social introduction services. This deficiency model holds that one significant reason why people use these services is because they lack the assertiveness and social skills needed to form relationships through informal social channels. The current research began with a related study on the matchmaker/client relationship which included a series of 27 in-depth interviews with current and former users of a matchmaking service. Throughout these interviews our impressions supported those of Darden and Koski (1988) and conflicted with the deficiency model. In contrast to the deficiency model, we came to believe that a "selectivity model" might be more accurate. The selectivity model holds that people use formal social intermediaries because their standards for a partner are relatively high and have not been met through traditional social channels. Rather than losers, these people could be characterized as "choosers" -- people who are highly selective in their criteria for a potential lifelong partner and who seek a

3 discerning and efficient channel for meeting the right match. While the present study views the selectivity and deficiency models as competing alternatives, they are not logically incompatible. It is possible that people who use formal social intermediaries are both deficient in social skills and relatively selective in their criteria for a romantic partner. Furthermore, the population of the typical dating service may be a mix of people; some who fit the deficiency model, some who fit the selectivity model, others who fit both models, and still others who fit neither. Nonetheless, our subjective impression from the interviews was that as a generalization about the typical member of the service we were studying, the selectivity model was correct and the deficiency model was not. The present research was designed as a test this impression using more objective survey methodology. In addition to performing a general investigation of these two models, the possibility of gender differences was tested for all hypotheses. Because our research question in this paper is limited to who joins matchmaking services, our investigation of gender differences was restricted to looking for interaction effects between gender and membership in the matchmaking service. introduction service. Included in this stereotype is the image that typical members are shy, communicatively incompetent, socially isolated, lonely, and suffer from low self-esteem. Hla: Shyness; Members will report higher levels of shyness than will non-members. Hlb: Eliciting openness; Members will report less success at eliciting openness from others in conversation than will nonmembers. Eliciting openness consists mainly of putting others at ease, and eliciting selfdisclosure from them. Hlc: Social network satisfaction; Members will report less satisfaction with the ability of their social network to meet their needs than will non-members. Hld: Loneliness; Members will report higher levels of loneliness than will nonmembers. Hle: Self-esteem; Members will report lower self-esteem then will non-members. HYPOTHESES Deficiency hypothesis The deficiency model was based on the popular stigma associated with clients of these services. HI: General Deficiency Hypothesis. Users of introductory services (herein referred to as "members") are more likely to be deficient in regards to social and psychological characteristics than nonusers of these services. In order to facilitate analysis, this general hypothesis was broken down into five sub-hypotheses. These hypotheses specify more clearly the components of what we consider to be the general stigma associated with members of a Selectivity Hypothesis The selectivity model is our rival to the deficiency hypothesis and is based on the data gathered in the exploratory interviews. H2: General Selectivity Hypothesis: Users of introductory services are more likely to be selective in regards to their criteria for a mate (i.e., economic, physical, personhood, and sociability attributes) than non-users of these services. Based on a factor analysis (described below), this general selectivity hypothesis was broken down into four sub-hypotheses. In each case it is hypothesized that matchmaking

4 clients will score higher on scales measuring the importance of the following qualities required in a potential mate than will the comparison group. H2a: Selectivity regarding financial resources H2b: Selectivity physical attractiveness H2c: Selectivity personality characteristics regarding regarding H2d: Selectivity regarding sociability METHODOLOGY Subject Pool All respondents in this study were associated with a large Jewish agency that sponsors a wide variety of programming and activities for singles. The agency is located in an upscale neighborhood community center in a major metropolitan area. The targeted audience for this service are professional, welleducated Jewish singles ranging from their mid-twenties to mid-forties who are looking for a lifelong partner. Due to the small pool of available older males, age limitations for women are more restrictive than for males (approximately age 35 for women). The users group was drawn from the current clients of a non-profit, matchmaking service sponsored by this agency. The member group included current clients (N=53 ) and former clients (N=39). Initial phone contact with the matchmaking service by prospective clients serves as pre-screening prior to an invitation for an in-depth interview by the matchmaker. Upon acceptance by the service and payment of a nominal annual fee ($250), the client is then notified of prospective dates and, upon agreement of the dates acceptability, can make direct contact for follow-up. On average, approximately 150 members use this service throughout the year. Prior analysis of written applications for the total membership reveals a balance of sex ratio, a predominantly well-educated, high-income, never-married membership (see table 1). (Insert table 1 about here) A comparable group of non-users (herein referred to as "comparison group") was drawn from the single members of the larger Jewish community organization that sponsors the introductory service (see table 1). Descriptive statistics showed no significant differences between the comparison and member groups on distribution by sex, age, and years lived in the metropolitan area. However, a chisquare test revealed that the members were somewhat more likely to have completed graduate school then were respondents from the comparison group (p <.05). Data Collection Data collection occurred over a twoyear period beginning December 1987 and ending August 1989. Initial interviews were conducted with 27 current and former members of the service with a proportionate number of males and females. Requests for the interviews were initiated by the matchmaker. The subsequent follow-up interviews were conducted by a trained graduate student. Interviews took place in a site selected by the informant (e.g., his or her home, or a restaurant). The interview format began with an open-ended question regarding the client's life situation at the time of joining the introductory service. Open- and close-ended questions concerning this event included the ways work, family, and social networks facilitate means for meeting others; reasons for using introductory services and specifically the service in question; their expectations for the service; personal feelings about utilization; initial and ongoing contact with the matchmaker; and dating experiences with other members.

5 Interviews ranged from 45 minutes to two hours and were recorded and transcribed for subsequent analysis of emerging themes. Based on these interviews two extensive questionnaires were developed for member and comparison groups covering the experience of single life. The questionnaire sent to members included items on their experience with the matchmaking service, whereas the questionnaire sent to the comparison group sought their views of introductory services. All items used in this study were drawn from the section of the questionnaire common to both groups. In order to assure confidentiality no names were included on the questionnaires. A postcard was mailed separately by the informants for a summary of the survey results. Three hundred and fifty-seven questionnaires were mailed to the members group, of which 25.8% (N=92) were returned. A total of 157 questionnaires were also mailed to the comparison group, of which 40% (N=64) were returned. Eight subjects from the comparison group who were users of other matchmaking services were deleted from the sample, leaving N= 56 in the comparison group. A modest return rate in both mailings may be due to the highly sensitive questions and length of the questionnaires (44 pages for member, 33 pages for comparison group). Analysis of Ouantitative Data The first stage in the data analysis was to validate scales included in the questionnaires relevant to the stated hypotheses. The items used to measure the deficiency model (H1) came either from pre-existing scales or from scales developed specifically for this study. Because the number of items used in the scales for the deficiency hypothesis was too large to be included in a single factor analysis, a series of three overlapping analyses were performed.1 These analyses resulted in five factors, each related to one of the five sub-hypotheses of the deficiency model, Hla through Hle (see table 2). (Table 2 about here) Nineteen items were used to test the selectivity hypothesis (H2). Each item described a quality of a potential spouse (e.g. "Self-confident") that respondents in the interviews had indicated was important to them. In the survey, respondents rated each quality on a fivepoint Likert scale in terms of how much they required that quality in a person they would be willing to marry (1= not important; 5= absolutely essential/must have). The factor analysis of these items resulted in a four-factor solution. Each of these four factors formed the basis for one of the four sub-hypotheses H2a through H2d (see table 3). (Table 3 about here) In all cases, extraction of factors was based on Maximum Likelihood Estimation. The proper number of factors was estimated by using a scree test and also by taking into account the interpretability of the resulting factors. Factor rotation was conducted using both an orthogonal procedure (Varimax) and an oblique rotation (Oblimin). The resulting factor structures were stable over both types of rotation. The function of these factor analyses was to (a) determine the basic factor structure, (b) assure that all items loaded on the appropriate scales, (c) eliminate items that did not load highly (structure matrix loading > 0.3) on any factor, and (d) eliminate items that detracted from discriminant validity (i.e., loaded highly on more than one factor). As a result of this analysis, several items were dropped from previously published scales or in some cases re-coded as part of a different scale. The alpha reliability coefficients were then calculated for each of the resulting scales. All of the alpha levels were easily acceptable, except for the loneliness scale (alpha =.66). As a result, interpretation of the respondents scores on this scale should be treated with caution. See tables 2 and 3 for reliability coefficients. Stage two of the analysis consisted testing of the two main hypotheses.

6 Analysis of Variance (ANOVA ) was used to test for significant differences between the member and comparison groups as well as check interaction effects between gender and membership. The results of these analyses are presented in table 4. (Table 4 about here) RESULTS & DISCUSSION Hypothesis 1, which represented the popular stereotype of matchmaking clients as more deficient in regards to social and psychological characteristics than non-users of these services, was strongly disconfirmed by the data. None of the five sub-hypotheses even approached significance in the direction indicated by the deficiency model. On the contrary, two of the sub-hypotheses reached or approached significance in the opposite direction of that predicted by the stereotype. Specifically, members were found to be less shy than the comparison group (p<.10), and they were found to have higher self-esteem than the comparison group (p<.01). The first of these findings, that members are more outgoing then the comparison group, is in direct opposition to the findings of Goodwin (1990) but makes sense in light of what using a matchmaker is really like. Joining a matchmaking service is setting oneself up for a long series of blind dates. While shy people may need these introductions more than others, they are also likely to find the prospect of all those blind dates somewhat threatening. In short, a sad paradox prevails -- those people who could potentially benefit the most from using a matchmaker may be too socially uncomfortable to take advantage of the potential benefits of this service. The second finding, that members have higher self-esteem then the comparison group, is quite interesting in light of the stigma attached to these services. In Sirgy's (1982) review of the role of self-concept in consumer behavior, he discusses two general theories of how consumers react to product images. In the "self-esteem motivation" theory, consumers are attracted to products that they see as having positive images, regardless of the consumer's own self-image. This theory does not predict that anyone would be attracted to a stigmatized product or service. However, people with high selfesteem might cope better with product stigma than people for whom self-esteem is low and therefore more of an active concern. This suggests that a high selfesteem person might be more likely to use a stigmatized product or service than a person with low self-esteem (assuming there were strong perceived benefits). In the "self-consistency motivation" theory, consumers are attracted to products that match their own self-image. This theory predicts that a person who saw himself or herself as socially undesirable would be attracted, or at least not averse to, a service that has the public image of being for social failures. The data presented here directly contradict the selfconsistency motivation theory by showing that high self-esteem is associated with a willingness to use a stigmatized service. These findings suggest that clients' high self-esteem may constitute an emotional defense mechanism in overcoming the stigma associated with using introductory services. The investigation of possible gender differences revealed that the differences (or lack of differences) between the members and the comparison group were not dependent on the gender of respondents. No significant interaction effects between gender and group membership were found. Our rival hypothesis points to the greater selectivity of members in regards to their criteria for a mate (i.e., economic, physical, and personhood attributes) than non-users of these services. This hypothesis was supported in two of the four tests. No support was found for the sub-hypotheses H2a: selectivity regarding financial resources, or H2d: selectivity regarding a highly sociable mate. However, support was found in the analysis for the main effect of membership in that H2b: selectivity

7 regarding physical attractiveness, approached significance (p<.10), and H2c: selectivity regarding personality characteristics and orientation towards others reached statistical significance (p<.01). Once again, the investigation of possible gender differences revealed that these findings were not dependent on the gender of respondents. No significant interaction effects between gender and group membership were found. Up to this point we have been assuming the differences between the member and the comparison group are predictive of membership in a matchmaking service. It is possible, however, that our interpretation of these findings is confuses cause with effect. Rather than antecedent-motivating factors, these differences may really represent the consequences of using this type of service. Perhaps, for example, it is not the case that people with high selfesteem are more likely to join a matchmaking service but, rather, that use of a matchmaker increases ones self esteem. This possibility is consistent with previous research which as found that some clients do receive extensive emotional and social support benefits for interacting with the matchmaker (Adelman & Ahuvia, 1991b). To investigate this possibility, correlations were computed between the number of dates members received through the matchmaker and their scores on the scales used in this paper. Because men received significantly more dates on average than did women (Mmen =11, Mnmn = 8, p <.05), it was possible for the effects of the number of dates a member received to be confounded with gender. To prevent this type of confounding, the effect of gender, age, and education were partialed out of the correlations. In all cases, the correlations were not significant, thus showing that the findings of this study do not become more pronounced with greater use of the service. This correlational analysis leads us to conclude that the differences between members and the comparison group found in this study reflect personal characteristics of the respondents that predate their involvement with the service. CONCLUSION The purpose of this paper was to investigate the accuracy of stereotypes that surround the users of social introductory services. Our findings contradict the general image that the users of matchmakers and similar services must have "something wrong with them," or in the common colloquial phrase be "losers." In fact, we found that users of the matchmaking service in this study were less shy and had higher self-esteem than a demographically similar comparison group. Furthermore, no differences were found between the two groups on measures of satisfaction with their social network, ability to elicit selfdisclosure from others, or loneliness. These findings confirm the subjective impressions of some researchers (Darden & Koski, 1988) but contradict other past findings (Goodwin, 1990) and the general reasoning that people use introductory services do so because they are not assertive or socially skilled enough succeed in traditional channels. As an alternative account to the popular stereotype embedded in the social deficiency hypothesis, we pose that clients of introductory services are more selective in their preferences for a potential mate than non-users of these services. This selectivity hypothesis holds that the primary motivation for using an introduction service is frustration with more traditional social channels that are not perceived as allowing the singles to meet sufficiently attractive others. Evidence for this conclusion comes from the findings that members of the matchmaking service were higher in self-esteem, more socially outgoing, and tended to have higher standards in their choice of mate than the comparison group. Based on these findings, we speculate the overriding motivation for joining the service was a desire to find "Mr. or Ms. Right" and an unwillingness to compromise in the

8 choice of a marital partner. In contemporary vernacular, rather than being "losers," members of this matchmaking service were "choosers." Members joined this matchmaking service not because they were unacceptable to others but because others were unacceptable to them. What might account for the sharp discrepancies between the current findings and those of Goodwin (1990)? Several key differences stand out as potentially explaining the divergent findings. First, Goodwin studied a large, for profit, computer dating service located in Britain, whereas the current research looked at a small, Jewish, non-profit matchmaker located in an upscale Chicago neighborhood. Although these differences may be important, it is difficult to determine which if any of them are the key without further research. Another possible difference comes from the way the two studies were conducted. The current research compares the clients of the matchmaking services to a closely matched group of singles of the same age, ethnicity, gender balance, and place of residence (although the members were slightly more educated than the comparison group). Goodwin however, compared the assertiveness and dating skills of the members to three student samples. It is therefore possible that the differences in assertiveness and dating skills he found reflect some discrepancy between the two groups (e.g. age, education, ethnicity, social class, social environment, etc.) other than the fact that one group were members of a dating service. In fairness to Goodwin it should be noted that the members of the dating service were generally highly educated, professional, and ethnically mainstream. Since this is also a fairly good description of the typical college student it is difficult to attribute the differences in assertiveness and dating skills to these particular demographic factors. However, it is still possible that age and immediate social environment (e.g. being in a college setting) as well as other unknown factors may have effected the outcome of these measures. The current study is also not without its weaknesses, many of which prompt useful areas for future research. Of principal concern is the issue of generalizability. We chose to do an indepth study of a single matchmaking service rather than try to provide a representative sample of similar services or of alternative intermediaries such as video dating, computer dating, singles ads, singles clubs, etc. This leaves many unanswered questions regarding to what degree our findings could be generalized across other services. For example, this service costs a moderate fee ($250 per year), does not use hard-sell sales techniques, and is sponsored by a reputable non-profit organization. In contrast, most introductory services charge much higher fees (often $500 - $3,000), rely extensively on high pressure sales strategies, and are strictly for-profit businesses. This might affect the character of the typical clientele, but it would be difficult to predict in exactly what way. Furthermore, the matchmaker in this study catered exclusively to Jewish clients. The acceptance of the matchmaker as a historically sanctioned position within Jewish culture may have given a sense of legitimacy to this service not enjoyed by secular agencies. This in turn may have affected the type of clients drawn to this service. Despite these objections, we have good reasons for believing that the clients studied here are not atypical of clients in other agencies. Woll and Cozby (1987), in their review of video dating and other alternatives to meeting mates, write that "the image that video dating, newspapers, and clubs serve only a small group of 'lonely hearts' lacking in basic social and/or dating skills is simply not accurate" (p. 71). Their impressionistic judgment based on extensive contact with users of these social services is similar to the conclusions drawn in this study. Furthermore, we speculate that the basic educational and career demographics of the respondent group are similar to the target market for most new social intermediaries, primarily because these

9 clients can afford the high fees demanded by these services. Lastly, the traditional image of a Jewish matchmaker did not seem to insulate the respondents from the stigma associated with the use of the service. Interviews with the matchmaker indicate that the overwhelming concern of most potential clients was that the person s/he would like to meet could not possibly be a member of the agency -- because everyone knew that "only losers would use a matchmaker" (themselves excepted of course). Another limitation of this study comes from the type of data (i.e., self-report) collected for this study. Because the stereotype of clients of these services is that they are not attractive to others, it would have been useful to have some systematic form for both member and comparison group social skills. Perhaps more importantly, no attempt was made to judge the physical attractiveness of the respondents. Although judgments made by the paid interviewer indicated that the clients were generally presentable and often quite attractive, it would have been advantageous to have better measures of this variable given its demonstrated importance in predicting interpersonal attraction (Curran 1973a, Curran and Lippold 1975, Mathes 1975). Despite these limitations, we believe that this study contributes to the development of our understanding of formal social intermediaries. As we indicated at the beginning, it is possible that utilization of these social channels could expand over the coming years. If this is the case, these conduits to romance represent a major change in the social ethos that has governed the dating and courtship process since the 1920s. To better grasp this trend and the way service intermediaries are permeating social relations and culture, we need to more fully understand the people who are experimenting with these innovative social channels. Furthermore, we believe that the stigma associated with this channels may in itself be detrimental to the development of romantic relationships. Apart from these services being perceived as anti-romantic, the people who clients meet through the service may suffer negative attributes by simply being a member. In stripping away the stigma, we begin to see that introductory services are not the last resort of social rejects, but rather a professional service for a self-assured and very demanding clientele. 1Because factor loadings are only meaningful in the context of the other items included in the analysis, this process of performing three overlapping factor analyses prevents us from presenting a table of factor loadings for the individual scale items.

10 REFERENCES Adelman, M. B. & Ahuvia, A. C. (1991a), "Mediated Channels for Mate Seeking: A Solution to Involuntary Singlehood?" Critical Studies in Mass Communication, 8, 273-289. Adelman, M. B. & Ahuvia, A. C. (1991b). Matchmakers as urban agents: A multimethod study. Paper presented at the Annual Conference of the International Communications Association, Chicago, IL. Ahuvia, A. C. & Adelman, M. B. (in press), "Marketing Metaphors for Meeting Mates," In R. Belk (Ed.). Research in Consumer Behavior. Greenwich, CT: JAI Press. Ahuvia, A. C. & Adelman, M. B. (1992), "Formal Intermediaries in the Marriage Market: A Typology and Review," Journal of Marriage and the Family, 54, 452-463. Austrom, D., M. de Sousa, and K. Hanel (1980), The Interpersonal Marketplace: Companions Ads, Presented at the Annual Meetings of the Canadian Psychological Association, Calgary, Alberta. Bailey, R. C. and D. G. Garrou (1983), "Dating Availability and Religious Involvement as Influences on Interpersonal Attraction," The Journal of Psychology, 113, 95-100. Bennet, J. (1989), The Data Game. The New Republic, Feb. 13, 20-22. Bernstein, W. M.; B. O. Stephenson, M. L. Snyder, and R. A. Wicklund (1983), "Causal Ambiguity and Heterosexual Affiliation," Journal of Experimental Social Psychology, 19(1), 78-92. Brand, D. (1988), "Make Me a Perfect Match," Time, Nov. 28,14-15. Brotman, B. (1993, February 7), "Classified Adventures," Chicago Tribune, Section 5, pp. 1 & 7. Bureau of the Census (March 1987), "Marital Status & Living Arrangements," Current Population Reports, Series P-20, #423. Bureau of the Census (April 1987), "Population Profile of the United States 1984/85," Current Population Reports, Series P-23, #150. Cheek, J. M. and A. H. Buss (1981), "Shyness and Sociability," Journal of Personality and Social Psychology (41), 537-540. Darden, D. K. and P. R. Koski (1988), "Using the Personals Ads: A Deviant Activity?" Deviant Behavior (9), 383-400. Freedman, A. M. (1989), "Beautiful Women Don't Go to Bars, But Helena Does," Wall Street Journal, September 22, p.l. Goodwin, R. (1990), "Dating Agency Members: Are They 'Different'?," Journal of Social and Personal Relationships, 7,423-430. Jannot, M. (1993, Feb.), "Love Brokers," Chicago Magazine, 60-67 & 86. Jones, E.F., A. Farina, A. H. Hastorf, H. Markus, D. T. Miller, and R. A. Scott (1984), Social Stigma: The Psychology of Marked Relationships. New York: W.H. Freeman and Company. Miller, L. C., J. H. Berg, and R. L. Archer (1983), "Openness: Individuals Who Elicit Intimate SelfDisclosure," Journal of Personality and Social Psychology, 44 (6), 1234 -1244.

11 Riggio, R. E. (1986), "Assessment of Basic Social Skills," Journal of Personality and Social Psychology, 51, 649-660. Rosenberg, M. (1965), Society and the Adolescent Self-Image, Princeton, NJ: Princeton University Press. Ryan, N. (1993, Feb. 11), "Love at 1st Sound Bite: Video Dating on Rise," Chicago Tribune, D1. Sirgy, M. J. (1982), "Self-Concept in Consumer Behavior: A Critical Review," Journal of Consumer Research (9), 287-300. Woll, S. BR and C. P. Cozby (1987), "Videodating and Other Alternatives to Traditional Methods of Relationship Initiation," Advances in Personal Relationships (1), 69-108.

12 TABLE 1 SAMPLE DEMOGRAPHICS MEMBERS COMPARISON (N=92) (N=56) M S.D. M S.D. AGE 32 4.0 33 5.8 YEARS LIVED IN AREA 18 12.6 20 14.1 Percent Percent MALES 48.9 53.6 W/COLLEGE ED. 26.1 40.0 W/GRADUATE ED. 70.7 56.4

13 TABLE 2: ITEMS FOR SCALES IN HYPOTHESES 1. Hla: Shyness Alpha =.83 Items rated on a four-point Likert scale from "strongly agree" to "strongly disagree." Responses are coded so that high scores indicate a greater degree of shyness. All items on this scale are taken from Riggio (1986), except where noted. I am often uncomfortable at parties and other social functions. I am more shy with members of the opposite sex. I don't find it hard to talk to strangers. I feel tense when I'm with people I don't know well. People feel relaxed around me.* I am socially awkward. I feel inhibited in social situations. Hlb: Eliciting Openness Alpha =.80 Items rated on a four-point Likert scale from "strongly agree" to "strongly disagree." Responses are coded so that high scores indicate the respondent is better at eliciting openness. All items on this scale are taken from Miller, Berg, and Archer (1983). People frequently tell me about themselves. I've been told that I'm a good listener. People trust me with their secrets. I easily get people to "open up." I enjoy listening to others. I encourage people to tell me how they feel. I'm very accepting of others. I'm sympathetic to people's problems. I can keep people talking about themselves.

14 TABLE 2 CONTINUED: ITEMS FOR SCALES IN HYPOTHESES 1. Hlc: Network Satisfaction Alpha =.84 The following items were rated on a four-point Likert scale from "very satisfied" to "very dissatisfied." Responses are coded so that high scores indicate a greater degree of satisfaction. In the past few months or so, how satisfied have you been with your leisure time? In the past few months or so, how satisfied have you been with your time at home? In the past few months or so, how satisfied have you been with your friendships? The following items were rated on a five-point Likert scale from "not at all" to "completely." To what extent does the group of people with whom you spend your free time adequately meet your needs for companionship? To what extent does the group of people with whom you spend your free time adequately meet your needs for sharing personal concerns? To what extent does the group of people with whom you spend your free time adequately meet your needs forfeeling as if you belong to a personal community? To what extent does the group of people with whom you spend your free time adequately meet your needs for caring for/giving to others? The following items were rated on a four-point Likert scale from "never" to "very often." How often are you without anyone to talk to about yourself? How often are you without anyone with whom you can share experiences and have fun? Hid: Loneliness Alpha =.66 The following item was rated on a four-point Likert scale from "strongly agree" to "strongly disagree." Responses are coded so that high scores indicate a greater degree of loneliness. I am regarded as lonely by others. The following item was rated on a five-point Likert scale from "not at all" to "completely." When you think of your life as a single person... all the daily pleasures and daily problems... do you feel lonely?

15 TABLE 2 CONTINUED: ITEMS FOR SCALES IN HYPOTHESES 1. Hle: Self-Esteem Alpha =.82 The following items were rated on a four-point Likert scale from "strongly agree" to "strongly disagree." Responses are coded so that high scores indicate high self-esteem. All items on this scale are taken from Rosenberg (1965), except where noted. I feel that I'm a person of worth, at least on an equal basis with others. I am able to do things as well as most other people. I feel I do not have much to be proud of. I wish I could have more respect for myself. I certainly feel useless at times. At times I think I am no good at all. I feel nervous when speaking to someone in authority.** I have trouble looking someone right in the eye.** ** Taken from Cheek and Buss (1981) ** Taken from Cheek and Buss (1981)

t 16 TABLE 3: ITEMS FOR SCALES IN HYPOTHESIS 2. H2a: Selectivity regarding financial resources Alpha =.83 Good (economic) provider Ambitious/career oriented H2b: Selectivity regarding physical attractiveness Alpha =.80 Good sexual partner Handsome/beautiful Well-built, good body H2c: Selectivity regarding personality characteristics Alpha =.79 Well educated Flexible, open to differences Self-confident Open and communicative Demonstrably affectionate Intellectually stimulating Good sense of humor Enjoys home/family life Concern for social issues Decent/good/kind H2d: Selectivity regarding sociability Alpha =.79 Highly sociable, extroverted Spontaneous, exciting

17 TABLE 4: DIFFERENCES BETWEEN MEMBER AND COMPARISON GROUPS SUBJECTS INTERACTION SCALE: ALL MEMBERS CONTROL P-VALUE P-VALUE N=148 N=92 N=56 SELF ESTEEM 3.16 3.26 3.00.005 SHYNESS 2.33 2.26 2.45.072 ELICITING OPENNESS 3.20 3.20 3.19 - NETWORK SAT. 3.11 3.12 3.09 - LONELINESS 2.58 2.52 2.69 - CHOOSER RE: MONEY 2.74 2.80 2.65 CHOOSER RE: SEXY 3.03 3.11 2.91.056 CHOOSER RE: PERSON 3.86 3.95 3.71.009 CHOOSER RE: FUN 2.79 2.76 2.83

I i i I I I