Differences in Romanian men's online personals by sexualities.
Abstract: As Internet usage increases, its ability to provide almost instant connections and to pre-screen potential partners has made it a popular source for meeting people. To contribute to current literature, we examine differences in content between 187 men seeking men and 193 men seeking women in a sample of Romanian (Bucharest) personals in 2007. Results show that men seeking women provide more information about themselves than men seeking men. While the results also show that some Romanian men are more open about expressing their sexualities than others, there are still many obstacles affecting how men choose to share and express their sexualities in Romania.

Keywords: personal ads, sexualities, Romania, Internet
Article Type: Report
Subject: Internet (Social aspects)
Mate selection (Analysis)
Online dating (Forecasts and trends)
Personals (Methods)
Personals (Social aspects)
Bachelors (Demographic aspects)
Single men (Demographic aspects)
Men's studies
Authors: Bartos, Sebastian E.
Phua, Voon Chin
Avery, Erin
Pub Date: 03/22/2009
Publication: Name: The Journal of Men's Studies Publisher: Men's Studies Press Audience: Academic Format: Magazine/Journal Subject: Social sciences; Women's issues/gender studies Copyright: COPYRIGHT 2009 Men's Studies Press ISSN: 1060-8265
Issue: Date: Spring, 2009 Source Volume: 17 Source Issue: 2
Topic: Event Code: 290 Public affairs; 010 Forecasts, trends, outlooks Computer Subject: Internet; Market trend/market analysis
Geographic: Geographic Scope: Romania Geographic Name: Romania Geographic Code: 4EXRO Romania
Accession Number: 204073661
Full Text: With its ability to provide almost instant connection and prescreening of potential partners, the Internet is an increasingly popular source for meeting people (e.g., Phua & Kaufman, 2003). Individuals use online personals to search for friends, romance and/or a committed long-term relationship by advertising their attributes and stating their desires. Not only are individuals seeking different types of relationships online, but online romance is also a global phenomenon. Many countries, poor and rich, such as Russia, Brazil, the U.S., Singapore and Morocco, have their own version of online personals. While online dating services are theoretically available to all, not everyone has access to them. Access is partly dependent on local culture and affordability. Those who have access to the Internet can seek partners in their local cities as well as in foreign cities as the virtual world has no national boundaries.

To date, researchers have examined different issues using personals as data. These issues include gender and sex roles (e.g., Phua, 2002), language use (e.g., Groom & Pennebaker, 2005), age preference (e.g., Jagger, 2005), race preference (e.g., Phua & Kaufman, 2003), health concerns (e.g., Phua, Hopper, & Vazquez, 2002), and construction of identity (e.g., Lester & Goggin, 1999). While researchers are interested in using personals as data in examining mating and dating preferences in different countries, the bulk of the research in English comes from the U.S.

To contribute to the current literature, we focus on a previously under-examined country: Romania. Romania is an interesting example for such research as it has gone through major socio-economic and political changes in the past few decades, and has recently joined the European Union (EU). Specifically, we focus on the differences in the content of personal ads between men seeking men and men seeking women, as alternative sexualities in Romania are still subject to much controversy (e.g., Badescu, Kivu, Popescu, Rughinis, Sandu, & Voicu 2007; Gallup Organization Romania, 2000).

Research on Personals

Common lore often advises that "beauty is in the eye of the beholder." However, studies on dating and mate selection have shown that individuals have strong preferences regarding the desired characteristics in their partners. Overall, physical attractiveness is an important trait in a partner, especially for men. Regan, Medina, and Joshi (2001) have documented the value of being "attractive" or "sexy," regardless of gender and sexual orientation. Physical traits are a good predictor of how many answers a person will receive (Pawlowski & Koziel, 2002). In line with these findings, Romanian men seeking a female partner online describe their ideal match predominantly in terms of physical attributes (Rusu & Bencic, 2007). Height and weight greatly contribute to one's physical appearance, and are easily quantified indicators of attractiveness. Preference for leaner body types has also been documented for U.S. (Lynn & Shurgott, 1984), Polish (Pawlowski & Koziel, 2002), and Romanian men (Rusu & Bencic, 2007). As for height, the last two studies show that both Polish and Romanian men seek shorter women.

One can argue that attractiveness and youthfulness are related. In a recent study, Teuscher and Teuscher (2007) report that both gay and straight men rate younger people as more attractive. Studying Polish personals, Pawlowski and Koziel (2002) find that youth is the strongest predictor of the number of responses a woman receives. Other studies using personals also document men's preference for younger partners, regardless of their sexual orientation (e.g., Kenrick, Keefe, Brian, Barr, & Brown, 1995). Rusu and Bencic's (2007) study confirms the same trend for straight Romanian men who seek partners on the Internet. It is also worth noting that men's preference for younger partners has been documented in a diverse range of cultures (Buss & Schmitt, 1993). Jagger (2005) calls the preference for youth "an index of the 'tyranny of youth' in a consumer culture" (p. 103). However, the preference for a particular age may reflect the availability of partners rather than solely reflecting a preference for youth. For example, Kaufman and Phua (2003) posit that adults in their early twenties prefer people around their age or older simply because there are fewer options as the legal age for sexual activities is around 18 years old, while adults in older ages (50+) may ask for younger partners simply because they are more available. Interestingly, they also find that Asian Americans are more open to older partners, suggesting age preference may be influenced by how older people are perceived within a culture.

Research has supported the assertion that there is race preference in mate selection. For example, Phua and Kaufman (2003) show that Asian Americans tend to prefer either Whites or people of similar ethnic group for dates. Preference in hair color, eye color and skin tone are sometimes used to suggest race preference (Phua & Caras, 2008). Blond women, usually associated with being white, are promoted by the media, but not actually favored by all men (Rich & Cash, 1993). Lynn and Shurgot (1984) report that red hair may actually be on top of hair color preferences.

Research has also found that advertisers highlight and request certain non-physical characteristics. Income and education are important in a man, but not in a woman--at least that is what evolutionary-psychological studies usually maintain (e.g., Townsend & Wasserman, 1998). There is some recent evidence that straight men might actually prefer women with less education and less income (e.g., Greitemeyer, 2007). Others (e.g., Regan et al., 2001) maintain that people seek wealthy and educated partners, regardless of their gender and sexual identity. Other characteristics that are important include gender roles, sexual roles, and health status. More importantly, preferences vary by advertisers' characteristics, such as age, gender and sexualities (Bailey, Kim, Hills, & Linsenmeier, 1997).

Most studies on personals have focused on the U.S., with fewer on other countries, such as Brazil, and Poland. To the best of our knowledge, Rusu and Bencic's (2007) study is the only study on Romanian personals. Their study focuses on heterosexual men and women. To contribute to the current literature, we examine the content differences in personal advertisements between men seeking men and men seeking women. Sexuality is an increasing important human rights issue, particularly for Romania who has recently joined the European Union. This is a baseline study and our specific goal is to examine the main characteristics that advertisers highlight and what they specify in a desired partner.

Data and Method

We collected the data from a Romanian Internet website in spring 2007. In this study, we focused on personals posted by men living in Bucharest, the capital of Romania, because these personals made up 45% of all personals posted by men. At the same time, no other single city had a large enough sample for us to conduct reliable sampling.

We first stratified the sample into men seeking men and men seeking women. Within each group, we systematically sampled every third case until we reached 200 cases. We reviewed the 400 cases and deleted cases that were duplicates (e.g., two personals having the same photographs) and those who self-identified as foreigners. The final sample size is 380, with 187 men seeking men and 193 men seeking women.

The personals consist of two sections: a grid-like part, where advertisers could select their option from a pull-down menu, and a space for write-in. The first part has an option for an English version that we copied verbatim. In this brief report, we will focus mainly on data from the first part. For the website to pre-code these variables indicates that these are probably the most common characteristics used in personals. In addition, the first section provides sufficient information for analysis. However, we used the data from the second part to check for consistent interpretation of the current data.

We caution readers not to simply interpret men seeking men to be gay or bisexual, or men seeking women to be straight for three reasons (e.g., Phua & Kaufman 1999). First, sexual behaviors and sexual identities are closely related but conceptually distinct. For example, 11% of men who identified as heterosexual in an online survey admitted they had had cybersex with other men (Ross Mansson, Daneback, & Tikkanen, 2005). Second, we do not know the sexual identity of these people, except for a few who specifically identified themselves. Third, the term "gay" is a Western construct even though it is commonly used around the world. In addition, different countries may have pre-existing terminologies that describe individuals' sexual identities that reflect[begin strikethrough]s[end strikethrough] local cultures. However, we are not suggesting that interpreting these people as "gay" or "straight" is necessarily incorrect but rather we are reminding readers that commonly used terminologies may vary in their meanings depending on the cultural context (e.g., McLelland, 2000).

Comparisons were performed on men seeking men and men seeking women, in order to determine whether one of the two groups was more likely to specify their own characteristics and those of their desired partners. Pearson's chi-squared test was performed to establish statistical significance. However, we are cognizant that these stated preferences may change in other circumstances, such as in face-to-face interactions (e.g., Kurzban & Weeden, 2007).

Characteristics of Romanian Men in Personals

Overall, more than half of both samples (i.e., men seeking women and men seeking men) mentioned most of the coded characteristics (see Table 1). The exceptions are the mentions of income for men seeking men (17.6%) and for men seeking women (37.3%), and the inclusion of at least one photograph by men seeking men (26.2%). Men seeking women are more likely than men seeking men to mention individual characteristics in their personals. The differences are statistically significant in most cases, except the mentions of height, weight, and field of work. While the mentions of these three characteristics are not statistically different between the two groups, they all constitute the highest percentages of both groups, suggesting that height, weight, and field of work are important pieces of information to be included in personals for both groups.

The differences that are statistically significant range from 5.5% for mentioning eye color and 6.5% for mentioning children to about 20% for mentioning income and 35.5% for including at least one photograph. The most significant difference, 35.5%, is whether advertisers include a photograph. Sixty-two percent of men seeking women provide one or more photographs of themselves when compared to only 26% of men seeking men. Displaying a photograph clearly showing one's face in personals may be a form of "coming out" for individuals who embrace alternative sexualities. This lower percentage among the men seeking men sample is not surprising as Romania is still a rather conservative and traditional country where homosexuality is frowned upon (e.g., Badescu et al., 2007). Throughout the 1990s, Romanian mass-media has treated gay issues superficially. Newspapers systematically promoted such stereotypes as gays being pedophiles, HIV-infected, mentally ill, and anti-Christian; a notable theme is the idea that gay rights are forcefully imposed by the E.U. and thus they are a blow to Romania's sovereignty (Creteanu & Coman, 1998; Spineanu-Dobrota, 2005). The language used in some major newspapers was remarkably trivial as they often referred to anal sex in order to mock gay rights (Creteanu & Coman, 1998). This unfriendly environment affects everyone who is not heterosexual, even though not all men seeking men self-identified as gay or bisexual.

While comparatively lower, this 26% of personals signifies that some people with alternative sexualities are no longer hiding in Romania. These individuals are seeking a partner via online personals; some of them are even willing to reveal who they are by showing a clear photograph of themselves. Nonetheless, the percentage of people in Romania who are out or are comfortable with their alternative sexualities is unknown. While the results show that some Romanian men are more open about expressing their sexualities than others, there are still many obstacles affecting how people choose to share and express their sexualities in Romania. We also have to keep in mind that these advertisers live in the capital city, perhaps the most modern city in Romania.

Another difference worth noting is whether advertisers mentioned their income levels. This point is important in part because it constitutes the lowest percentage in both groups. At the same rime, it has the second largest discrepancy--20% difference. Consistent with this finding, men seeking women are also more likely to mention their occupation when compared to men seeking men (94.8% and 86.6%, respectively). These findings support earlier studies that posit men seeking women are more likely to offer financial security in exchange for beauty and attractiveness (e.g., Weiderman, 1993), whereas financial stability is not necessarily the most significant information offered in men seeking men's personals. We are not suggesting that financial status is less important in partnering among men seeking men. Instead, we are suggesting that the mentioning of income may not be of high priority in personals for men seeking men when compared to men seeking women.

What Romanian Men Want

Overall, less than one-third of both men seeking women and men seeking men mentioned most of the coded characteristics for potential matches (see Table 2). The exceptions are the mentions of hobbies for men seeking men (67.4%) and for men seeking women (79.3%), and the mention of "field of work" by men seeking women (34.2%). Hobbies seem to be the most important of an ideal partner's characteristics. Although the percentages for mentioning a match's characteristics are much lower than those for self characteristics, men seeking women again tend to be more specific than men seeking men. However, of the fourteen characteristics coded, only seven are statistically different between the two groups. The differences that are statistically significant range from 9.2% for mentioning match's education to 13.6% for mentioning match's maximum height.

What is interesting is that weight and height are statistically more important for men seeking women than for men seeking men. Consistent with Rusu and Bencic's (2007) study on heterosexuals, of men seeking women in our sample who have a height preference, seventy-nine percent stated that their potential mate's maximum height be shorter or the same as them; the remaining 21% would consider if the partners are within a couple of centimeters taller than themselves. On the other hand, all men seeking men who have a height preference stated the maximum height of their partners to be taller than them. With regards to maximum weight, about 88% of men seeking men have no specific stipulation when compared to 76% of men seeking women. By contrast, there is very little interest, if any, in such issues as eye color, hair color and hair length.

About 30% of both men seeking men and men seeking women stated a maximum age for their partners. If we consider the age difference between advertisers and their stated maximum age, about 86% of men seeking men and 69% of men seeking women would consider an older partner, even though the first group is more willing than the latter group to consider a wider age range. The difference between whether they prefer an older partner is statistically different between the two groups. Still, both groups have a higher percentage of advertisers willing to consider an older partner. For men seeking women, this finding suggests an interesting nuance that is consistent with Pawlowski and Koziel's (2002) results that show younger women receive a greater number of responses than older women. Together, these results suggest that men may be more likely to answer personals posted by younger women but are more willing to cast a wider net to solicit more responses of which they could pick and choose to answer. The extent to which this strategy is common among men deserves further research.

Both men seeking men and men seeking women are moderately interested in their future partner's field of work, while they are equally uninterested in his or her income. This finding is not surprising as Romania is still a traditional country where most consider the husband the primary breadwinner. More than half of the adults in Romania believe that men cannot take care of young children (Gallup Organization Romania, 2000), and there should not be any househusbands (Magyari-Vincze, 2005). The same idea is transparent in the disapproval of paternal leaves (Baban, 2000) as one man has put it, "men are not maternal." As such, the income from potential spouses may be considered irrelevant. A slight deviation from this macho ideal is that men seeking women are more likely to specify a level of education than men seeking men. However, the majority in both groups do not specify any preference.

Marital status is more important to men seeking women than to men seeking men. Again, this primarily reflects the fact that only marriages between men and women are possible in Romania. For men seeking men, the lack of interest in marital status could have two meanings. First, this question may be interpreted as essentially irrelevant as same-sex marriage is not legally possible in Romania. Second, men seeking men may be more understanding of the fact that under the past and current gender and sexual systems in Romania, many men who desire another man may end up marrying a woman (e.g., see McLelland {-(2000)-} for similar explanations of Japanese gay men). As such, marital status may be less important than whether the other person is willing to engage in sexual activities. This is consistent with the fact that most of them do not specifically self-identify as gay or bisexual in their personals. However, this issue warrants further exploration that is beyond the scope of this brief report.


In this brief report, we provide a baseline analysis of Romanian men's personals. The results indicate that a greater proportion of men included information about themselves versus what they want in a partner. The results also show an interesting distinction between men seeking men and men seeking women. Our results show that men seeking women provided more information on themselves than men seeking men. For example, a higher percentage of men seeking women specify the ideal range of height and weight of their partners. This appears to be inconsistent with studies from the U.S. that reported men seeking men being more open and expressive about what they want in a partner than men seeking women. Specifically, Phua (2002) argues that "[men seeking] men's personals are the real analytic lens here because they are the ones faced with the ambiguities that need to be negotiated" (p. 188). However, in this report, we find that the opposite is true: men seeking women provided more information in their online personals. This inconsistency could be reflecting that these two groups emphasized different characteristics. Looking at the mention of sexual acts and discretion (results not presented in this brief report), a higher percentage of men seeking men mentioned sexual acts and discretion than men seeking women. Another possible explanation could be that models of communication among people with alternative sexualities in Romania are less established and are still evolving.

A point worth repeating is that 26% of the men-seeking-men personals included a photograph. Even though the percentage is much higher for men seeking women and that the difference is statistically significant, this 26% suggests that some people are willing to reveal their alternative sexualities and resist the relatively homophobic culture in Romania. These "brave" individuals may be a selective group and the obstacles they continue to face are not captured by the data. We caution readers not to interpret the personals market as an exact mirror of real life.

While comparisons of personals between men seeking men and men seeking women provide interesting results, we recommend that future research focus on differences between female sexualities. At the same time, in our study we have limited our sample to men living in Bucharest. Future studies should also examine the phenomena in rural areas and smaller cities where the lingering influences of communism through re-traditionalization and the "ruralization" (referring to the penetration of rural values and beliefs in larger cities during the communist era) may be more prominent (e.g., Cirstocea, 2003). Previous research has identified race preference as important (e.g., Phua & Kaufman, 2003) but it is conspicuously absent in the personals in this study. While the majority of the population in Romania are Romanians, how the racial tension between Roma and Romanians affect their racial preference is worth investigating.


Baban, A. (2000). Men's perceptions of sexuality, reproductive behavior and partner relations. Prague: Open Society Foundation.

Badescu, G., Kivu, M., Popescu, R., Rughinis, C., Sandu, D., & Voicu, O. (2007). Viata in cuplu [Couple life]. Bucharest: Open Society Foundation. Retrieved September 30, 2007, from http://www.osf.ro/ro/publicatii.php

Bailey, J. M., Kim, P. Y., Hills, A., & Linsenmeier, J. A. W. (1997). Butch, femme, or straight acting? Partner preferences of gay men and lesbians. Journal of Personality and Social Psychology, 73, 960-973.

Buss, D. M., & Schmitt, D. P. (1993). Sexual strategies theory: An evolutionary perspective on human mating. Psychological Review, 100, 204-232.

Cirstocea, I. (2003). Conceptul de gen o categorie utila pentru analiza socio-istorica a regimului comunist din Romania [The concept of gender, a useful category for the socio-historical analysis of the Communist regime in Romania]. In I. Baluta & I. Cirstocea (Eds.), Directii si teme de cercetare in studiile de gen din Romania [Research directions and themes in Romanian gender studies] (pp. 95-143). Bucharest: Colegiul Noua Europa.

Creteanu, A., & Coman, A. (1998). Homosexuality in the written media in Romania. Retrieved 30 September, 2007, from http://accept.ong.ro/presaeng.html#presa

Gallup Organization Romania. (2000). Barometrul de gen (August, 2000). [Gender barometer (August 2000)] Bucharest: Open Society Foundation. Retrieved September 30, 2007, from http://www.gallup.ro/romana/poll_ro/releases_ro/pr030411_ro/ pr03041l_ro.htm

Greitemeyer, T. (2007). What do men and women want in a partner? Are educated partners always more desirable? Journal of Experimental Social Psychology, 43, 180-194.

Groom, C. J., & Pennebaker, J. W. (2005). The language of love: Sex, sexual orientation, and language use in online personal advertisements. Sex Roles, 52, 447-461.

Jagger, E. (2005). Is thirty the new sixty? Dating, age and gender in a postmodern, consumer society. Sociology, 39, 89-106.

Kaufman, G., & Phua, V. C. (2003). Is ageism alive in date selection among men? Age requests among gay and straight men in internet personal ads. The Journal of Men's Studies, 11, 225-235.

Kenrick, D. T., Keefe, R. C., Brian, A., Barr, A., & Brown, S. (1995). Age preferences and mate choice among homosexuals and heterosexuals: A case for modular psychological mechanisms. Journal of Personality and Social Psychology, 69, 1166-1172.

Kurzban, R., & Weeden, J. (2007). Do advertised preferences predict the behavior of speed daters? Personal Relationships, 14, 623-632.

Lester, N. A., & Goggin, M. D. (1999). "Extra! Extra! Read all about it!" Constructions of heterosexual black male identities in the personals. Social Identities, 5, 441-468.

Lynn, M., & Shurgot, B. A. (1984). Responses to lonely hearts advertisements. Personality and Social Psychology Bulletin, 10, 349-357.

Magyari-Vincze, E. (2005). Gender, ethnicity and the construction of the social order: A view from below in the context of Romania. Anthropological Journal on European Cultures, 14, 197-225.

McLelland, M. (2000). Is there a Japanese "gay identity"? Culture, Health & Sexuality, 2, 459-472.

Pawlowsi, B., & Koziel, S. (2002). The impact of traits offered in personal advertisements on response rates. Evolution and Human Behavior, 23, 139-149.

Phua, V. C. (2002). Sex and sexuality in men's personal advertisements. Men and Masculinities, 5, 178-191.

Phua, V. C., & Caras, A. (2008). Personal brand in online advertisements: Comparing White and Brazilian male sex workers. Sociological Focus, 41, 238-255.

Phua, V. C., Hopper, J., & Vazquez, O. (2002). Men's concerns with sex and health in personal advertisements. Culture, Health and Sexuality, 4, 355-363.

Phua, V. C., & Kaufman, G. (1999). Using the census to profile same-sex cohabitation: A research note. Population Research and Policy Review, 18, 373-386.

Phua, V. C., & Kaufman, G. (2003). The crossroads of race and sexuality: Date selection among men. Journal of Family Issues, 24, 981-994.

Regan, P. C., Medina, R., & Joshi, A. (2001). Partner preferences among homosexual men and women: What is desirable in a sex partner is not necessarily desirable in a romantic partner. Social Behavior and Personality, 29, 625-633.

Rich, M. K., & Cash, T. F. (1993). The American image of beauty: Media representations of hair color for four decades. Sex Roles, 29, 113-125.

Ross, M. W., Mansson, S-A., Daneback, K., & Tikkanen, R. (2005). Characteristics of men who have sex with men on the Internet but identify as heterosexual, compared with heterosexually identified men who have sex with women. CyberPsychology & Behavior, 8, 131-139.

Rusu, A., & Bencic, A. (2007). Choosing a mate in Romania: A cognitive evolutionary psychological investigation of personal advertisements market. Journal of Cognitive and Behavioral Psychotherapies, 7, 27-43.

Spineanu-Dobrota, S. (2005). Homosexualitatea intre normal si patologic [Homosexuality between normal and pathological]. Bucharest: Tritonic.

Teuscher, U., & Teuscher, C. (2007). Reconsidering the double standard of aging: Effects of gender and sexual orientation on facial attractiveness ratings. Personality and Individual Differences, 42, 631-639.

Townsend, J. M., & Wasserman, T. (1998). Sexual attractiveness: Sex differences in assessment and criteria. Evolution and Human Behavior, 19, 171-191.

Wiederman, M. W. (1993). Evolved gender differences in mate preferences: Evidence from personal advertisements. Ethology and Sociobiology, 14, 331-335.


Babes-Bolyai University, Romania


Gettysburg College, PA


Gettysburg College, PA

Sebastian E. Bartos, Department of Clinical Psychology and Psychotherapy, Babes-Bolyai University, Romania; Voon Chin Phua, Department of Sociology and Anthropology, Gettysburg College; Erin Avery, Department of Sociology and Anthropology, Gettysburg College.

This project was partially funded by the Professional Paper funding through the Provost's Office at Gettysburg College. We would also like to Ambika Kirkland, Melissa Gagermeier, and Desiree Ciambrone for their constructive comments.

Correspondences concerning this article should be address to Voon Chin Phua, Department of Sociology and Anthropology, Gettysburg College, 300 North Washington Street, Campus Box 412, Gettysburg, PA 17325. Electronic mail: vphua@gettysburg.edu
Table 1
Percentage Mentioning of Respondent's Characteristics

                           Men seeking men   Men seeking women
Variable                       n = 187            n = 193

Income                          17.6              37.3 ***
Photos (showed face
  except for two)               26.2              61.7 ***
Personality descriptors         73.3              86.0 **
Height                          96.8                99.0
Weight                          95.2                97.4
Eye color                       89.8              95.3 **
Hair style                      87.7              94.8 **
Hair color                      86.1              93.8 **
Occupation                      86.6              94.8 **
Field of work                  100.00               97.9
Hobbies                         84.0              90.7 **
Marital status                  78.6               86.5 *
Mentioned children              88.8                95.3

* p <0.05. ** p <0.01. *** p <0.001.

Table 2
Percentage Mmtinning of Prcferences in Match's Characteristics

Variable                Men seeking men   Men seeking women
                            n = 197            n = 193

Minimum match age            28.3               32.6
Maximum match age            29.4               32.1
Minimum match weight         10.2             22.8 ***
Maximum match weight         12.3              24.4 **
Minimum match height         13.4             26.4 ***
Maximum match height         13.9             27.5 ***
Eye color                     0.0                0.0
Hair color                    0.5                1.0
Hair length                   2.1                3.1
Education                     5.3              14.5 **
Income                        1.1                2.6
Field of work                30.5               34.2
Marital status                5.9              15.5 **
Match's hobbies              67.4              79.3 **

** p <0.01. *** p <0.001.
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.