Sociation Today ® 
The Official 
Journal of 
The North 
Association: A 
Refereed Web-Based 
ISSN 1542-6300
Editorial Board:
George H. Conklin,
 Emeritus, North Carolina
 Central University

Robert Wortham,
 Associate Editor,
 North Carolina
 Central University

Rebecca Adams,

Bob Davis,
 North Carolina
 Agricultural and
 Technical State

Catherine Harris,
 Wake Forest

Ella Keller,
 State University

Ken Land,
 Duke University

Steve McNamee,

Miles Simpson,
 North Carolina
 Central University

William Smith,
 N.C. State University

Editorial Assistants

John W.M. Russell,

for Authors

Searchable Index
Sociation Today
from the
Directory of 
Open Access
Journals (DOAJ)

Sociation Today
is abstracted in
Sociological Abstracts
and a member
of the EBSCO
Publishing Group



Volume 10, Number 2

Fall/Winter 2012

Public Presentations of Gendered Bodies: A Look at Gay and Lesbian Online Dating Profiles


Andrew Latinsky

University of Central Florida


    For most people, gender is a taken for granted part of everyday life, a crucial, but rarely examined part of one's identity. However, when examined, the role becomes part of a complex picture, where debate ranges upon the role and impact of the normally invisible notion of male and female. According to Butler (1990), notions about sexuality and gender are social constructions, recreated and reinforced through stylized acts in everyday existence.  Expectations about gender can have such powerful influence as to affect the expectations of physical bodies differently for men and women (Crawley, Foley, and Shehan 2008). For men 'working out' and being healthy means adding muscle, leading to increased size and strength for men. In contrast, exercise and health magazines for women encourage behavior such as dieting and cardio, which reduce the physical size of women (Crawley, Foley, and Shehan 2008). These socially defined differences become part of a larger social structure that influences male dominance over women in social situations (Crawley, Foley, and Shehan 2008). According to West and Zimmermann (1987), this creates and perpetuates the notions surrounding a gender binary, two separate, socially unique genders who 'naturally' act differently from each other via differential social conditioning.

    However, not all persons end up fitting such neat molds, and certain desires can seem to inherently defy certain gendered notions (Butler 1990). Collins' (2000) work Black Feminist Thought outlines how the different social realities of black women's break from notions of what a generic 'woman' is expected to be. Collins (2000) argues that the lived experiences of black women differ from stereotypical ideals. This results in unique modes of thought defined by black historical oppression, their unique need to labor and lead families, all while dealing with the  need to survive within a white power structure. When the intersectionality framework is focused on the notion of gender, the intersection of homosexuality and gender opens the possibility of similarly unique, lived experiences that instead define new gendered roles and resistances outside of the traditional binary of male and female. 

    To this end, this paper focuses on a content analysis of gay and lesbian online daters, looking at their self-described identities and desires for a potential date. Online dating provides a unique opportunity to "examine people's stated preferences in a real-life situation" (Robnett and Feliciano 2011: 807) related to these ideas. Online dating sites also offer the benefit of presenting specific public preferences in areas such as date body type, ethnicity, and education level to outside observers. This is, in some ways, socially unusual to see in the public arena; in almost no other social situation would someone be asked to publicly state that they only prefer to date someone of their own race. In online dating, however, this is fairly common and sometimes even necessary to improve one's chances of finding an ideal partner. In addition, online dating has unique opportunities for misrepresentation and known biases in favor of people presenting themselves in a positive light (Hancock and Toma 2009). Because of such opportunities, public presentations of gendered norms and ideas in interpersonal relationships can be examined in user profiles to answer questions about how gender stereotypes may be impacting or informing people's perceptions and expectations. 

Literature Review

Gendering Bodies by Performance and Surveillance

     In a classic paper on the topic of gender, West and Zimmerman describe their sociological theory of 'doing gender' as "the activity of managing situated conduct in light of normative conceptions of attitudes and activities appropriate for one's sex category"  (1987: 126); describing gender not as something one is, but a performance that is reinforced socially through gendered interaction. They argue that the process of gender is socially constructed and taught to children through a myriad of social institutions and that each has various impacts and possible influences on both children and adults (West and Zimermann 1987). In the United States, this is generally expressed in a gender binary, where all persons are expected to behave either as female or male as their gender based upon presumed biological sex characteristics (Crawley, Foley, and Shehan 2008; Goffman 1977). 

    Processes that create the behaviors expected of a person's gender are not one sided or restricted to familial and institutional levels: "to 'do' gender is not always to live up to normative conceptions of femininity or masculinity; it is to engage in behavior at the risk of gender assessment" (West and Zimermann 1987: 136). In describing gender assessment, this can be thought of as a form of performance, where the actor presents information about themselves in both verbal and nonverbal ways by behaving in certain manners to the social 'audience' (Goffman 1959). In Symbolic Interaction theory, as these meanings pass back and forth between the observer and the observed, facilitated by social environment, a form of social reality is created between people (Blumer 1969).

     In a gender framework, humans rely upon "mutual monitoring" (Goffman 1977: 5) of gender stereotypical behavior as part of our social training. Starting early on, young children are taught at institutional and familial levels about their gender and the behaviors expected of being either male or female (Martin 1988; West and Fenstermaker 1995). Further compounding this issue are the narratives of masculinity and femininity, and an increasing perception of gender-based differences as children age (Messner 2000). In Kane's (2006) interviews of parents of preschoolers, mothers and fathers describe efforts to allow both sons and daughters to at times break normative gender roles. However, masculine behavior would inevitably be associated as being counter to being female, and boys in particular were strongly supposed to engage in hegemonic/stereotypical masculine behavior (Kane 2006). Particularly when children begin to age, gender is further reinforced and watched upon hegemonic lines within peer groups. In a study of reactions to females growing out bodily hair, Fahs noted that participants created gender role-based conflicts between family, partners, and coworkers in response to the change to the point of describing emotions such as feeling like a "circus freak" (2011: 463).

    A person's socially defined gender in the gender binary becomes an inevitable part of one's identity, with avenues of resistance limited. Even attempts to break away from gender stereotypes can instead place people in situations that are also stereotypically gendered, but with an illusion of rebellion (Wilkins 2004). With such cultural and eventual self-policing, 'natural' stereotypes about men and women are taught and eventually become the expectation of how to behave for children. These stereotypical behaviors create the common gender roles we see, like men needing to be large, muscular, and dominant breadwinners (Kimmel 2008)  and females needing to be small, dainty, and slim through avenues such as dieting (Crawler, Foley, and Shehan 2008, Hall et al. 2010). 

Intersectionality and Gender/Sexuality Framework

    It is in these contexts that the relation of gender and other concepts, such as sexuality and race, becomes something of interest. If all people of the same gender are socially conditioned in similar ways and are expected to behave 'naturally' as either male or female, we might expect uniform ideas of how to behave across the spectrum of humanity. Yet, historical, racial, and other factors can lead to entirely different expectations of what it means to 'be' female or male (Collins 2000). Focusing on race, Collins' (2000) conceptualization of intersectionality argues that the lived experiences of black women have always defied the stereotypical gender norms of whites. Collins (2000) cites those such as Sojourner Truth and her lived experience working as a slave and the years of hard labor she endured in spite of women's gender roles to the contrary in during slavery times. She argues there is a connection between such social systems such as race, gender, class, sexuality, and nationality in how systems of oppression function and that they are not independent of one another (2000). 

    In these ideas about gender, two ideas of how gender influences the individual are revealed. One is expressed in the gender binary of male and female and the surveillance system that educates and reinforces people to act as either male or female (Crawley, Foley, and Shehan 2008). In this view, all bodies are gendered male and female from birth and childhood experience, carried into eventual self-policing of what it means to be female or male and, although some resistance may occur later in life, each body is inevitably trapped in the gender it was given, and the gendered messages that come with it usually cannot be escaped within hegemonic social life. Intersectionality-based ideas, in contrast, would predict that minorities such as gays and lesbians could potentially escape such direct hegemonic gendering, their lived experiences acting as buffers and creating unique resistances to stereotypical expectations (Collins 2000). Indeed, findings suggest that gays and lesbians deal with greater gender role conflict, and may be more tolerant views in regards to gender roles. Gender role conflict in gays and lesbians are well known (ex: Blashill and Hughes 2009 & Szymanski and Carr 2008), though often versed in terms of a potential issue or psychological stressor. When examining that stereotypical gender role-playing in gay and lesbian relationships does in fact occur, Marecek, Finn, and Cardell (1982) found nonetheless that such stereotypical gender conformity was less common than in heterosexual relationships. In general terms, Shechory and Ziv (2007) found that same-sex gay and lesbian couples held more liberal and open views on gender roles than heterosexual couples. More specifically, we have cases such as Moulton III and Adams-Price's (1997) findings that "homosexual males are more tolerant of individuals displaying gender discordant behavior than are heterosexual males" (448) in regards to cross-dressing.  However, do these more liberal views continue to hold when the pressures of finding a potential partner, particularly via online dating, come into play?

Queer Bodies and the Internet

     In a subculture where terminology mixes lacy (feminine) gay men and Bears (large, hairy, masculine gay men), and the butch (masculine lesbian woman) seeks a lipstick lesbian (feminine) (Thorne and Coupland 1998), it can seem like the gender binary begins to break down in regards to what is 'normal' in gay and lesbian gender relations. However, media portrayals that range from advertising concepts to the online paint a different picture, with hegemonic influences clouding the spaces that gay and lesbian persons attempt to manipulate. 
In the advertising world, an idea of a "Dream Market" (Oakenfull 2007: 50) surrounds the gay and lesbian communities, due in part to a perceived prosperity in gay and lesbian households. However, Oakenfull (2007) notes that the advertising realm's focus on a "one size fits all" (51) approach has led to a series of biases that focus the media on gay men as the target default population with a focus particularly on gay male sexuality being used to sell (Sender 2003). One result of this is a greater importance of pornographic media and bodies on gay men compared to heterosexual men; this intense scrutiny of the physical tends to lead to a higher level of bodily dissatisfaction for gay men (McArdle and Hill 2009).
The focus on sexuality for gay men continues into the arena of relationships, with print dating profiles for gay men having a more varied sexual lexicon and a greater focus on masculine physical appearances (Bartholome, Tewksbury, Bruzzone 2000; Thorne and Coupland 1998). In contrast, Thorne and Copeland (1998) found that, for lesbian women, there was much greater focus on politics and interpersonal rapport as desirable traits in a partner. Similarly, Groom and Pennebaker describe a "report" versus "rapport" (2005: 458) style that differentiates gay men from lesbian women in posted profiles about themselves. Concurrent with the broad issues one finds in the public presentations of gay and lesbian bodies, particularly in media and cyberspace, that exploration and acknowledgement can generally be accepted, but only if relegated into a hidden or less visible status (Brookly and Cannon 2009; Sender 2003). It is this pressure from the heteronormative culture to hide deviant and queer norms that highlights media based portrayals that focus on gay men, to the exclusion of lesbian women, and views all homosexuals as generalized, heteronormative male body (Brookly and Cannon 2009; Oakenfull 2007; Sender 2003). This is part of the social structure that leads to a stereotypically masculine ideal body for gay men (Oakenfull 2007; Sanchez 2012), while the comparatively invisible lesbian woman can be found to have adopted more diverse, although feminine-biased, ideas of gendered bodies and roles (Sender 2003; Thorne and Copeland 1998).

Dating Preferences and Online Dating

    Theories about the logic and types of people we become attracted to range from 'logical,' economic assumptions, such as dating as a means of seeking to maximize genetic gains in children (Davis 1941; Merton 1941) to concepts that focus on culture-based explanations, such as assimilation theory (which suggests there are deterministic reasons for in-group dating and marriage) (Gordon 1964; Kinder and Sanders 1996). Logically, if a person never spends time around another person for racial, cultural, economic, or other reasons, it is impossible for a relationship to form. In one example of racial isolation, Kinder and Sanders' (1996) survey data of demographics within cities indicates a physical level of racial exclusion, showing neighborhoods divided strongly on racial lines. However, in the rising age of the internet, a new style of dating has rapidly begun to grow in the United States: one that trades neighborhoods for the information superhighway.

     A sample of Americans surveyed by Pew in 2005 documents the rise of online daters and their common characteristics. Sautter, Tippett, and Morgan's (2010) analysis found that 5.6% of all respondents have used online dating sites, with males who are educated beyond high school and living in non-rural areas having the highest probability of being online dating website users., the largest online dating site, claims that one in five relationships today start online.

    Two of the most unique principles to the online profile are the selective representation of a profile picture of a person, and the ability of a user to tailor her or his presentation of self to other online daters. Hancock and Toma (2009) reviewed fifty-four online profile pictures of heterosexuals, comparing users' self-reported accuracy versus independent judges and coders identifying areas of misinformation. Users rate the accuracy of their photographs very highly, but that independent judges find them to be significantly less accurate: Additionally, women's photos were significantly less accurate than men, and patterns of deception were found to promote youthfulness in both genders (Hancock and Toma 2009). This suggests that social desirability is a major factor when users choose which photo to use as their profile picture. Similarly, in a study of misrepresentations in online profiles, Hall et al. (2010) found that while women were more likely to misrepresent their weight on their online profiles, men were more likely than women to misrepresent personal attributes such as politeness and personal interests. Suggesting that for heterosexuals, positive self-representation includes a stereotypically gendered component.

    As shown, the examination of online dating is beginning to grow with the relatively recent explosion of growth of the industry, at least for heterosexuals. A problem though appears when one wants to examine the intersection of online dating and the gay and lesbian community: study of the phenomenon is nearly nonexistent. Many of these studies reflects the views, preferences, photographs, etc. exclusively of heterosexuals (ex: Robnett and Feliciano 2011; Levin, Taylor, and Caudle 2007). Almost all other research in a review of the literature leads to a different methodological issue towards this effect, as seen in Sauter, Tippett, and Morgan's (2009) work. A representative study would have to include a certain number of gay and lesbian daters in their 'Internet dater' category, but their survey does not contain a question on sexuality, and so no secondary analysis can be performed in order to examine that variable (Hall et al. 2010). In the case of Hall et al.'s online survey, nowhere in the paper is any indication that sexuality was a variable under examination or controlled for, although factors such as gender, age, weight, and education were covered. Of all the online dating literature examined in this review, only two papers seem to have knowingly gathered data on gay and lesbian online daters: Lawson and Leck (2006) due to their use of a snowballing interview methodology, and Groom and Penebaker (2005) due to homosexual dating profiles as a focus. However, in the case of both studies, the online dating field has evolved greatly in the last seven years. It is in this context, I find the need for an examination of a population that although it can be assumed in a random sample to be gathered, does not seem to be recognized greatly in the literature.

    Overall, online dating as a modern phenomenon breaks the immediate boundaries of traditional dating. For example, the default search of only asks for the sex/gender of the user, the desired date sex/gender, a request for the user's zip code, and defaults to only showing profiles with pictures. This creates a situation where one no longer carries some of these limitations of only geographic and social boundaries. Instead, a person must specifically define their preferences in an ideal date to have any improved chance of finding someone suited to them. In contrast to the more subtle form of exclusion by lack of interaction, a user is now in a position where they must decide to explicitly state what their racial, education, etc. preferences are in this public arena. In addition, it leads to a unique form of public presentation of self, where there are opportunities and incentives to give misleading, overly positive presentations that can attract greater attention, leading to a possibly increased chance of dating success. Because this is a rapidly growing public venue, where deception in favor of perceived social desirability is possible, aspects of how people present themselves in gendered manners can be seen in a rapidly growing experience in the United States. Because of the broad appeal of online dating in the modern day, an opportunity is presented to look at a normally unexamined population of gay men and lesbian women.

Research Question

    As Goffman described, "It is not, then, the social consequences of innate sex differences that must be explained, but the way in which these differences were (and are) put forward as a warrant for our social arrangements…" (1977: 302). Although gays and lesbians, as an overall minority status, evade some of the Foucaultian discipline about gender norms (Foucault 1977), the question still stands: What gendered identity influences are being favored in presentations of self? In a culture where Toys R Us and FAO Schwartz attempted to sell toys in a "Boy's World" and "Girl's World" (Jackson II 2002: 131), where parents, peers, and culture all assume boys must act masculine and girls must act feminine (Crawley, Foley, and Shehan 2008) from early childhood, do these notions take precedence? Or does the lived experiences of gay men and lesbian women, with identities and concepts that flaunt conventional gendered norms such as cross-dressing (Moulton III and Adams-Price 1997), create openings that allow new spaces and expressions for gendered presentations of self? Online dating's unique ability to promote certain, idealized images as the presentation of self (Hancock and Toma 2009), combined with the public  nature of the profiles presented, grants an opportunity to examine how gendered ideas influence how gay and lesbian actors present themselves to an audience of those they seek an ideal relationship with.


    In examining gendered elements, this research examines the self-reported body types of online dating profiles of gay men and lesbian women from as well as the body types these people state they want to date. This focuses on potential public presentations of bodies in traditionally gendered ways, such as males being large and athletic and females being smaller and lithe (Crawley, Foley, and Shehan 2008).

    This research presents a content analysis of 410 public dating profiles selected from the online dating website who were seeking same-sex dating partners, collected between January and April of 2012. At the time of data collection, posting an online profile is free; however, full functionality-such as the ability to send and receive messages with other members-requires a paid subscription (a common model in modern online dating).'s unique selling point is its size, having acquired many other fairly large online dating competitors such as and Yahoo Personals. This transformed the different websites into one large standardized pool of daters, which as of 2009 consisted of over 1.3 million paying subscribers and 20 million users ( form 10-K 2010). As such, claims to be responsible for starting more dates, relationships, and marriages than any other online dating website, representing nearly a fifth of the total market (form 10-K 2010).

    On each profile, posters selected demographic information from checklists about themselves, such as age, education level, ethnicity, religion, etc. In addition, each profile listed allows a poster to choose multiple options from the same checklist for the same demographic information, but instead focused on what they desire in a potential date. All of this information is posted on the public profile, allowing other people to see information about the poster and their preferences in a potential date or contact. Choosing no option from the checklist is often a choice (although some, such as gender and age, are required to create the account), and results in default answers of either 'I'll tell you later' if about the poster of the profile, or 'No Preference' if about a potential date.

    For example, in the category of Education daters selected from these six choices to describe their academic achievement: High School, Some College, Associates Degree, Bachelor's Degree, Graduate Degree, and PhD/Post-Doctoral. Using this, posters could either designate a personal education level or refuse to answer (I'll tell you later"), both of which show up under the "ME" section of the profile. For date preferences, posters could select any number of options (including all six at once) or none at all ("No Preference"), which show up under the "MY DATE" section for education. Education variables for this analysis are grouped into five major categories: No Answer, High School Education, Some College, College Graduate, and Post-College education for analysis. 

    Height was measured in centimeters, converted from the imperial foot and inch height given by the poster, for ease of recording. Age is determined by giving a birthdate to the website upon account creation, listed in years, with a minimum required age of 18 to use the website. Ethnicity in this study was restricted to only White/Caucasian and Black/African American. Although a poster could list multiple ethnicities, only those who listed a single ethnicity category were collected in order to eliminate the chance of a poster being listed in both the White/Caucasian and Black/African American sample.

    Body type options taken from the site for posters are broken into three major categories for chi-squared analysis, with the options taken directly from the site used given next to the categorization. 

  • 'Athletic': Athletic and Toned
  • 'Average': Slender, About Average
  • 'Overweight': A few extra pounds, Heavyset, Stocky, Curvy (Female only choice), Full-Figured (Female only choice), Big-and-beautiful (Female only choice)
Date body types are given similar categorization, accounting for the fact a poster can choose multiple options. Because each person could select from only one category, or could choose from multiple categories, three additional categories of analysis were created.
  • 'Conventionally Attractive': Selections from only the 'Athletic' and "Average" categories
  • 'Diverse': Selections from either 'Athletic' and 'Overweight', or 'Average' and 'Overweight' categories.
The last category is reserved for those who selected from all three categories without using the default choice of no preference.

    410 total profiles were selected among those who identified as seeking a same-sex date between the ages of 18 and 50 and living within 50 miles of four major US cities: New York, Los Angeles, Chicago, and Atlanta. New York and Chicago have 103 profiles each; Los Angeles and Atlanta have 102. These four cities were chosen to represent geographic diversity in the United States (Robnett and Feliciano 2011) and to ensure a reasonable sample size for collection. Numbers were split so that comparison groups of White/Black and Male/Female have equal sizes in the total sample, giving an overrepresentation of Black/African descent ethnicity. Sorting of profiles by default on the website is done by a method called "match picks;" in order to remove the unknown meaning of this categorization, a different selection was used. Profiles were instead sorted by "activity date," placing profiles of those who had logged onto their profile most recently at the top of the search. This had the added benefit of ensuring that a user must be active on the website to be in the sample. In addition, to remove the risk of a profile being incomplete due to being new to the website, no profile listed as "NEW" at the time of collection was included in this sample.

Variables collected from each profile can be split into two broad categories for analysis: demographics/appearance of the poster, and desired traits in a potential date. 

  • Poster: Age, City, Ethnicity, Body Type, Height, Political Views, Education, Is there a photograph of the poster?
  • Desired Date: Body Type, Education
    Table 1a and 1b present the demographic data of the sample. Overall, the sample's average age is between 37 and 38 years of age, regardless of gender. Self-described political views of posters appear to be heavily biased in favor of self-describing as a liberal (39.3%) compared to a conservative (2.7%), although the majority refuses to label themselves on such lines, choosing other options such as non-conformist, some-other-viewpoint, etc. (54.4%) as their description. Overall, the sample is contains a majority of both males and females having at least a bachelor's degree, with about 22% of each gender having education beyond a bachelor's degree.

Table 1a
Demographic Characteristics of Sample

Political Views
Total Sample
No Answer
Religious Views
No Answer
Spiritual but Not Religious

Table 1b
Demographic Characteristics of Sample by Gender

  Male Female
Age 38.2 37.51
Height (cm) 179.64 166.48
Height (feet, inches) 5' 11" 5' 5.5"
No Answer 6.9% 9.8%
High School 1.5% 2.9%
Some College 20,8% 27.8%
College Graduate 49.0% 37.1%
Post College 21.8% 22.4%

    In addition, a logistic regression analysis examines effects of gender, controlling for ethnicity, education, poster body type, and age on willingness to not post a profile picture on the online dating profile. This particular notion is examined because the's website makes a particular focus of profiles pictures. Among its within-website importance is the proclamation to all users that people with profiles containing a photo have 15 times more views than non-photo-containing profiles, the search engine defaulting to only profiles with pictures, and requests to ask a person to place a picture on their profile if they do not presently have one. As such, a tradeoff is presented; does one reveal an image of their body, making the process of dating more efficient, or does one hide their body based upon the perceived disadvantage of a socially deemed unattractive body type?

    Gender will test females against males, ethnicity will test blacks contrasted to whites, and poster body type will test the Overweight category against the 'conventionally attractive' categories of 'Athletic' and 'Average'.

    Following the method outlined by Frankfort-Nachmias and Leon-Guerrero (2011), all 'no answer's' for poster education are regressed to the sample mean. This allowed the 8.6 percent of the sample that did not answer the education variable to be used in the regression analysis. With no reasonable method to regress the body type, all seven (1.7 percent) of the posters who gave no answer for their body type are excluded from the analysis.


    Since gay men have a greater focus in popular media presentations, and a more strict masculine, sexualized, presentation as the ideal (Brookly and Cannon 2009; Oakenfull 2007; Sanchez 2012). Public presentations about gay male masculinity in online dating profiles will focus on heteronormative, large, athletic norms of masculinity. In contrast, the lesbian woman's less powerful media focus (Sender 2003; Thorne and Copeland 1998) will cause a mixed reaction, that will mostly likely default to heteronormative standards of femininity.

Hypothesis 1 (Body)

  • 1a: Gay men will be more likely to self-describe as "athletic and toned" (Athletic Category). 
  • 1b: Lesbian females will self-describe as "average" or "slender" (Average Category) due to stereotypical female norms, and be less likely to self-describe in the 'Overweight' category than men. 
  • 1c: Males and females will seek similar characteristics (Athletic men, Average women) in possible dates.
Hypothesis 2 (Has Photo?)
  • Females will be no less likely to have a photograph of themselves on display than men, due to stricter standards and stereotypes related to attractiveness. 
Hypothesis 3 (Has Photo Regression): 
Lesbian females will be no less likely to have a photograph of themselves on display than men, even when controlling for age, race, education, and the body type of the user.


Tables 2 through 4 present the results of chi-squared analysis on dependent variables. Appendix A shows crosstabs broken down by race of tables two and three, confirming the trends are consistent when controlling for race. Table 2 presents the results of chi-squared analysis on the posters self-described body type by gender.

Table 2
Chi-squared Results on Self-Described Body Type Based on Gender

No Answer
1 (-2.5)
6 (2.5)
7 (1.7%)
87 (16.5)
54 (-16.5)
141 (34.4%)
98 (-1.5)
101 (1.5)
199 (48.5%)
19 (-15.5)
44 (12.5)
63 (15.4%)
Note: Chi-squared results in each column are given as number observed, with the unstandardized residuals in parenthesis.
*p<.05  **p<.01

    Statistically significant differences in poster self-described body type due to gender are found at the .01 alpha level. Self-described 'Average' body type was the most common (48.5%) followed by 'Athletic' (34.4%) and 'Overweight' (15.4%), with only 1.7% refusing to answer. Males are only more likely than females to describe their bodies as athletic and toned, consistent with hypothesis 1a, with 87 males compared to 54 females giving such a response. Females are instead more likely to describe their body in the 'Overweight' category, with 44 females compared to 19 males. Smaller differences were found in the 'Average' and no answer categories, with 98 males compared to 101 females in the 'Average' category, and 1 male compared to 6 females not giving a response. This lack of difference on the 'Average' category combined with female descriptions of self in the 'Overweight' category leads to a full rejection of hypothesis 1b.

    Table 3 presents the chi-squared analysis on the body types of ideal dates based on gender.

Table 3
Chi Squared Results on Ideal Date Body Type(s) 
Based on Gender

No Answer/
No Preference
53 (0.5)
52 (-0.5)
105 (25.6%)
Athletic and Toned (1)
30 (13)
4 (-13)
34 (8.3%)
Average (2)
16 (1)
14 (-1)
30 (7.3%)
Overweight (3)
2 (-2)
6 (2)
8 (2%)
Attractive (1 & 2)
90 (-1)
92 (1)
182 (44.4%)
(1&3) or (2&3)
2 (-.5)
3 (.5)
5 (1.2%)
All Categories
(1, 2 & 3)
11 (-11.5)
34 (11.5)
45 (11%)
Note: Chi squared results in each column are given as number observed, with the unstandardized residual in parenthesis. 
*p <.05 **P <.01

Statistically significant differences in stated ideal date body types are found at the .01 alpha level. Overall, the most popular ideal date body type category was selected from the 'Conventionally Attractive' category (44.4%), selecting from both 'Athletic' and 'Average' categories. This was followed by having no selections, and thus the default of no preference (25.6%), selections from all categories (11%), selecting only the 'Athletic' category (8.3%), 'Average' (7.3%), 'Overweight' (2%), and the 'Diverse' category with the least (1.2%). Females are more likely than males to describe ideal partner types in all three major categories than men, with 34 females compared to 11 males. Males are instead more likely to state their ideal date's body type is 'Athletic' and toned, with 30 males compared to 4 females. All other categories had no major gender based difference, with residuals of two or less. Similar to table two and consistent with the hypothesis 1c, males were more likely to desire an athletic body; however, females similarly did not state a desire for the expected body type. Instead, females were more likely to choose an inclusive categorization for an ideal date's body, selecting from all three groups.

Table 4 presents the results of chi squared analysis examining the likelihood of having a picture on the online profile of a gay man compared to a lesbian woman. 

Table 4
Chi Squared Results on if Profile Has Photograph or Not

Has Photo
182 (7)
168 (-7)
350 (85.3%)
No Photo
23 (-7)
37 (7)
60 (14.7)
Note: Chi-squared results in each column are given as number of observed with the unstandardized residual in parenthesis.
* p<.05 **p<.01.

     85.3% of all 410 users profiles had a picture, compared to 14.7% who did not. Overall, men were significantly more likely than women to have a profile picture at the .05 alpha level.  Table 5 presents the results of a logistic regression analysis, examining the effects of gender on likelihood to not have a photograph of oneself on the online dating profile.  Race, age, self-description as a conventionally attractive body type category, and education level are controlled for in the model.

Table 5
Effect of Gender Controlling for Ethnicity, Age, Education and attractiveness on Likelihood to Not Have a Photograph

Increasing Age
Increasing Education
Chi Square
C&S Rsquare=
Note: Cell entries are given as logistic regression coefficient/odds ratio with the standard error given in parenthesis.
* p <.1  **p<.05  ***p<.01

 Overall, the regression model was found to be a significant predictor of the status of a photograph in a profile at an alpha of .01. Of the variables under examination, statistical significance at an alpha of .01 when controlling for all other variables in the model were found for two variables. Difference was found in comparison of the racial categories of black and white, where being black increased the odds of not having a photo by 3.524 times. An increase in the scaled variable age predicted a lower likelihood of having a photograph in the profile by 1.069 times per year. Females contrasted to males and rising education were found to have a statistical significance at an alpha of .1. Being female instead of male predicts a 1.698 times greater likelihood of not having a profile picture, consistent with the hypothesis expectations. Each category of education level increase predicting a 1.415 greater likelihood of not having a profile picture. Being in either the 'Athletic' or 'Average' category, contrasted to the 'Overweight' category, had no impact. 


    Overall, results suggest that there is influence of both stereotypical gender norms and media portrayals in willingness to present an image publically, how one presents their body type and the types of bodies that lesbian women and gay men present in their online dating profiles. Gay men are both more likely to describe themselves as 'Athletic' and to state their only ideal date is someone of an 'Athletic' body. However, the full picture becomes a bit more complicated in that feminine stereotyping suggests dieting and slim figures as the ideal for women (Crawley, Foley, and Shehan 2008). Women were found instead to have acceptance of more diverse forms of bodies as shown in the higher likelihood of female posters self-describing in the 'Overweight' category and having a greater willingness to date from all three bodily categories. Although this did not agree with the hypothesis of the paper as given, this is in concordance with some of the literature that suggested lesbian women care more about rapport, and are more open to multiple body types (Groom and Pennebaker 2005). In addition, although only at the .1 alpha level for the regression, the chi squared data evidence suggests women were more likely to not have a profile picture than men, with a level of significance at .1 even when race, age, education, and body type are accounted for, and in spite of's efforts to get profile pictures on all profiles.

    As a further layer, to discuss either gender's presentation to their audience in any real term of 'diversity' is misleading. Although 'Overweight' choices consisted of 3 of 6 possible body types for men, and 6 of 9 for women, few men or women dared to describe themselves as such, even if some of the terms such as stocky, big-and-beautiful, and curvy are ones that carry a positive connotation. In regards to an ideal date, both genders show a notable bias towards dating body types that meet stereotypical standards of attractiveness. In this online dating environment, public presentations appear to consist of stereotypically attractive body-typed persons, seeking only those who are similarly conventionally attractive. Suggesting that Goffman's (1977) view of people as actors, when placed on such a public arena, will give performances that are structured on very hegemonic scripts of who is supposed to be attractive.

    On a structural level, it is also interesting that three terms, curvy, big-and-beautiful, and full-figured, are exclusively allowed to be used by women or those seeking women. Reflecting a level of gendered expectations, that consequently limits gay men to only six types of bodies as possible descriptions on the website. In this case, the stage itself in Goffman's analogy has been restricted, and only certain props within a heteronormative setting can be used for the initial public presentation towards the desired audience.

    Although it is outside of the primary focus of this paper, it is worth mentioning from an intersectional dimension the surprise that, even accounting for gender, age, and standards of attractiveness, a black gay or lesbian individual is 3.5 times more likely not to have a profile picture than their white counterparts. It is hard to state why the racial component is such a strong and significant predictor; such high percentages of profile pictures overall undermine explanations such as lack of camera or pictures. This suggests a potential avenue of race based factors that may be playing a role. In addition, controlling for the same factors, age was found to be strongly significant if less directly powerful per year. This is in line with ideas of prior research that social desirability and ideals of attractiveness focused around youth do play a role in public presentations of online daters (Hancock and Toma 2009). By itself, a profile picture may seem like a small loss; however, taken in a broader context, those who are lesbian females, black, or older, are forgoing something directly and strongly advertised by the system as effective in order to hide their public presentation from scrutiny. Self-disempowerment in this case could be a reflection of a larger, harder to see, trend. 

    Taken together, a mix of elements from the gender binary and intersectional stance can be seen. Gender, the media, and ideal notions about masculine bodies appears to influence the public presentation of one's body, with gay men in particular showcasing themselves as having and desiring stereotypically male athletic bodies. This agrees with the body of literature showing the prevalence and influence of media images showcasing hegemonic norms of masculinity as the ideal gay male body type (Bartholome, Tewksbury, Bruzzone 2000; Thorne and Coupland 1998). This becomes particularly notable to my personal experience because, during collection, multiple persons familiar with the concept of this paper stated they were worried that lesbian women might conflate with men in regards to presenting a desire for athletic bodies in regards to body type. In part due to the physical stereotype of a 'butch' lesbian woman. Despite this popular concern, lesbian women appear to be rebelling against the norms that the only ideal body type is slender and thin, and do not subscribe to only the 'butch' lesbian as an alternative. More lesbian women than expected using only the gender binary's model describe themselves in terms that imply extra weight over the 'Average' category, and women were more willing to accept multiple body types as ideal dates compared to gay men. Yet, with such high numbers of both gay men and lesbian women who describe themselves in only conventionally attractive ways and seeking only conventionally attractive people, it appears that, in public presentations, the desire to present oneself publically as an attractive person in an attempt to gain the socially desirable goal of an attractive partner is what most likely informs how online daters view their bodies.

Limitations and Suggestions for Future Research 

    Future research could minimize the limitations of this analysis by expanding the dataset and allowing direct comparisons on more complex terms. Collection of a heterosexual sample would allow more precise comparisons of the effect being gay or lesbian has on public presentation. A comparison of an exclusively gay and lesbian population in this analysis may be hiding that stereotypical standards of attractiveness are far more extreme in the heterosexual population. In addition, the population size is relatively small at only 410 profiles, expansion of the gay and lesbian dataset would allow a greater level of confidence in the interactions of gender and sexuality in statistical analysis. Furthermore, although research indicates that persons tend to mislead in favor of greater social desirability (Hancock and Toma 2009), the study itself focused exclusively on heterosexuals. Gay men and lesbian women may not behave in the same presumed manner, despite the likelihood presented in this data analysis, and ensuring that the trend holds in such cases would be crucial to the validity of online dating research in general.

    In further avenues of research, more complex intersectionality-focused analysis could be done by incorporating variables such as race through racial preferences of potential dates and class through stated income levels. Expansion into analysis of other online dating websites may yield unique effects due to the website or allow greater variety of options in user choice of such variables as body type for analysis. Due to the strong effects of race and age in the regression analysis, more focus on race or age based effects could be performed.


Align Your Brand with the Leading US Relationship site. via IACA advertising.  Retrieved on October 15th, 2011.
percent20Overview percent20One percent20Sheet.pdf

Bartholome, Adreanna, Richard Tewksbury, and Alex Bruzzone. 2000. ""I Want a Man:" Patterns  of Attraction in All Male Personal Ads." The Journal of Men's Studies. 8(3):309-21.

Blashill, Aaron J. and Honore M. Hughes. 2009. "Gender Role and Gender Role Conflict:  Preliminary Considerations for Psychotherapy with Gay Men." Journal of Gay & Lesbian  Mental Health 13(3):170-86

Blumer, Herbert. 1969. Symbolic Interactionism: Perspective and Method. Berkeley:  University of California Press

Brookey, Robert A. and Kristopher L. Cannon. 2009. "Sex Lives in Second Life." Critical Studies  in Media Communication 26(2):145-64.

Butler, Judith. 1990. Gender Trouble: Feminism and the Subversion of Identity. London:  Routledge

Collins, Patricia. H. 2000. "Gender, Black Feminism, and Black Political Economy." Annals of  the American Academy of Political and Social Science 568:41–53.

Collins, Patricia H. 2000. Black Feminist Thought: Knowledge, Consciousness, and the Politics  of Empowerment. London: Routledge.

Crawley, Sara L., Foley, Lara J., and Constance L. Shehan. 2008. Gendering Bodies. Lanham:   Rowman and Littlefield Publishing Company.

Davis, Kingsley. 1941. "Intermarriage in Caste Societies." American Anthropologist 43(3): 376- 95.

Fahs, Breanne. 2011. "Dreaded 'otherness': Heteronormative Patrolling in Women's Body Hair  Rebellions." Gender & Society 25:451-72.

Foucault, Michel. 1977. Discipline & Punish: The Birth of the Prison. New York: Random  House.

Frankfort-Nachmias, Chava and Anna Leon-Guerrero. 2011. Social Statistics for a Diverse  Society 6th ed. Thousand Oaks, CA: Pine Forge Press.

Goffman, Erving. 1959. Presentations of Self in Everyday Life. Carden City: Doubleday and  Company Inc.

Goffman, Erving. 1979. Gender Advertisements. New York: Harper and Row Publishers

Goffman, Erving. 1977. "The Arrangement Between the Sexes." Theory and Society 4(3):301- 31.

Gordon, Milton Myron. 1964. Assimilation in American Life: The Role of Race, Religion, and National Origins. Oxford University Press.

Groom, Carla J. and James W. Pennebaker. 2005. "The Language of Love: Sex, Sexual  Orientation, and Language Use in Online Personal Advertisements." Sex Roles 52(7):  447-61.

Hall, J. A. et al. 2010. "Strategic Misrepresentation in Online Dating: The Effects of Gender, Self- monitoring, and Personality Traits." Journal of Social and Personal Relationships 27(1):  117-135.

Hancock, J. T., and Toma, C. L. (2009). "Putting Your Best Face Forward: The Accuracy of Online  Dating Photographs." Journal of Communication 59(2):367-386.

IAC/INTERACTIVECORP Form 10-K. United States Security and Exchange Commission.  Fiscal year ending in December 31, 2009. Retrieved 2012-04-08.

Jackson, Ronald L. II. 2006. Scripting the Black Masculine Body. Albany, NY: State University  of New York Press.

Kane, Emily. 2006. "No Way My Boys are Going to be Like That! Parents' Responses to Gender  Nonconformity" Gender and Society 22: 455-76.

Kinder, Donald R. and Lynn M. Sanders. 1996. Divided by Color : Racial Politics and  Democratic Ideals. Chicago, Ill: University of Chicago Press

Kimmel, Michael. 2008. Guyland: The Perilous World Where Boys Become Men. New York:  HarperCollins Publishers

Lawson, Helene M. and Kira Leck. 2006. "Dynamics of Internet Dating." Social Science  Computer Review 24(2):189-208

Levin, Shana, Pamela L. Taylor and Elena Caudle. 2007. "Interethnic and Interracial Dating in  College: A Longitudinal Study." Journal of Social and Personal Relationships 24(3): 323-41

Marecek, J. J., S. E. S. Finn and M. M. Cardell. 1982. "Gender Roles in the Relationships of  Lesbians and Gay Men." Journal of Homosexuality 8(2):45-49.

Martin, Karin. 1998. "Becoming a Gendered Body: Practices of Preschools" American  Sociological Review 63:494-511.

McArdle, Keri A. and Melanie S. Hill.  2009. "Understanding Body Dissatisfaction in Gay and  Heterosexual Men: The Roles of Self-Esteem, Media, and Peer Influence" Men and  Masculinities 11(5):511-32.

Merton, Robert K. 1941. "Intermarriage and the Social Structure: Fact and Theory."  Psychiatry 4:361-74.

Messner, Michael. 2000.  "Barbie Girls Versus Sea Monsters: Children Constructing Gender."  Gender & Society 16:765-84.

Moulton, John L., III. and Carolyn Adams-Price. 1997. "Homosexuality, Heterosexuality, and  Cross-Dressing: Perceptions of Gender Discordant Behavior." Sex Roles 37(5-6):441-50.

Oakenfull, Gilian. 2007. "Effects of Gay Identity, Gender and Explicitness of Advertising 
Imagery on Gay Responses to Advertising." Journal of Homosexuality 53(4):49-69

Robnett, Belinda and Cynthia Feliciano. 2011. "Patterns of Racial-Ethnic Exclusion by Internet  Daters." Social Forces 89(3):807-28.

Sautter, Jessica M., Rebecca M. Tippett and S. P. Morgan. 2010. "The Social Demography of  Internet Dating in the United States." Social Science Quarterly 91(2):554-75.

Sender, Katherine. 2003. "Sex Sells: Sex, Class, and Taste in Commercial Gay and Lesbian  Media." GLQ 3(9):331-65.

Shechory, Mally and Riva Ziv. 2007. "Relationships between Gender Role Attitudes, Role  Division, and Perception of Equity among Heterosexual, Gay and Lesbian Couples." Sex  Roles: A Journal of Research 56(9-10): 29-38

Szymanski, Dawn M. and Erika R. Carr. 2008. "The Roles of Gender Role Conflict and  Internalized Heterosexism in Gay and Bisexual Mens Psychological Distress: Testing  Two Mediation Models." Psychology of Men & Masculinity 9(1):40-54. 

Thorne, Adrian and Justine Coupland. 1998. "Articulations of same-sex desire: Lesbian and gay  male dating advertisements." Journal of Sociolinguistics 2(2):233-57.

West, Candace and Don Zimmerman. 1987. "Doing Gender". Gender & Society 1:125-51.

West, Candace and Sarah Fenstermaker. 1995. "Doing Difference." Gender & Society 9(1):8-37

Wilkins, Amy C. 2004. "So Full of Myself as a Chick": Goth Women, Sexual Independence,  and Gender Egalitarianism." Gender & Society 18:328-48.


A .pdf file of the cross tabulations of body-type analysis controlling for race.   Click here.

 Return to Sociation Today, Fall/Winter 2010 

©2012 by Sociation Today