This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on http://publichealth.jmir.org, as well as this copyright and license information must be included.
Gay, bisexual, and other men who have sex with men (MSM) regularly experience homophobic discrimination and stigma. While previous research has examined homophobic and HIV-related intergroup stigma originating from non-MSM directed at MSM, less is known about intragroup stigma originating from within MSM communities. While some research has examined intragroup stigma, this research has focused mostly on HIV-related stigma. Intragroup stigma may have a unique influence on sexual risk-taking behaviors as it occurs between sexual partners. Online sexual networking venues provide a unique opportunity to examine this type of stigma.
The purpose of this study is to examine the presence and patterns of various types of intragroup stigma represented in Men Seeking Men Craigslist sex ads.
Data were collected from ads on Craigslist sites from 11 of the 12 US metropolitan statistical areas with the highest HIV/AIDS prevalence. Two categories of data were collected: self-reported characteristics of the authors and reported biases in the ads. Chi-square tests were used to examine patterns of biases across cities and author characteristics.
Biases were rarely reported in the ads. The most commonly reported biases were against men who were not “disease and drug free (DDF),” representing stigma against men living with HIV or a sexually transmitted infection. Patterns in bias reporting occurred across cities and author characteristics. There were no variations based on race, but ageism (mostly against older men) varied based on the ad author’s age and self-reported DDF status; bias against feminine gender expression varied based on self-reported sexual orientation; bias against “fat” men varied by self-reported DDF status; bias against “ugly” men varied by a self-report of being good-looking; and bias against people who do not have a DDF status varied based on self-reported HIV status and self-reported DDF status.
Despite an overall low reporting of biases in ads, these findings suggest that there is a need to address intragroup stigma within MSM communities. The representation of biases and intragroup stigma on Craigslist may result from internalized stigma among MSM while also perpetuating further internalization of stigma for men who read the sex ads. Understanding patterns in the perpetuation of intragroup stigma can help to better target messages aimed at making cultural and behavioral shifts in the perpetration of intragroup stigma within MSM communities.
Experiencing discrimination and stigma may have negative physical and mental health consequences for gay, bisexual, and other men who have sex with men (MSM). Research examining how discrimination among MSM is experienced has addressed two pervasive and often interacting forms of stigma: discrimination based on actual or perceived sexual orientation [
Another form of stigma that MSM may experience is intragroup stigma, defined as stigma within communities of MSM. Intragroup stigma may result from either the internalization of homophobic stigma among MSM or the heterogeneity of the MSM community. Communities of MSM contain a range of individual characteristics, such as race, age, HIV serostatus, etc, which could act as the basis for the generation of stigma within the MSM community. Research has examined how HIV stigma has been perpetuated among MSM; this type of intragroup stigma can lead to a rift or fracture within MSM communities with divisions based on HIV serostatus [
A useful source for examining intragroup stigma among MSM and between sexual partners is through Internet-based sex-seeking websites and apps. Online sex-seeking has become increasingly popular [
Data were collected from 11 of the 12 metropolitan statistical areas (MSAs) with the highest HIV/AIDS prevalence in the United States, ranked by the Enhanced Comprehensive HIV Prevention Planning (ECHPP) project [
Data collection was performed consecutively over 11 days (October 8, 2013 through October 18, 2013) with data collected from 1 city per day. After 2 data analysts developed a codebook with a list of variables for data extraction, they coded the first 50 ads for testing. Once the codebook was tested and finalized, a data analyst used the codebook to extract the data from the remaining ads. To minimize bias, data were collected from the first 200 ads listed before 2:30 PM (a randomly selected time) in each city’s time zone, standardizing the time of day for which data were collected across cities. Ads that were not looking for sex (eg, ads selling sex toys) or where couples created an ad together were excluded. This allowed for the correct identification of author characteristics. The total sample size included 2200 sex ads (200 per city). No identifying information was collected, and there was no interaction between the data collector and the subjects. Data were extracted from ads, entered into an Excel (Microsoft Corp) spreadsheet, and imported into STATA version 13.1 (StataCorp LP) for analysis.
We collected two types of variables: self-reported characteristics of the ad authors and reported biases in the ads. Domains not mentioned were coded as such in the data set. Self-reported characteristics included race/ethnicity (white, black, Latino, Asian, other); age (entered as a continuous variable and later categorized into age groups 18-25, 26-35, 36-45, 46 and above); sexual orientation (homosexual, heterosexual, bisexual); HIV status (negative, positive); self-reported “disease and drug free” status (“DDF,” “clean,” “healthy”); and self-reported physical appearance (“good looking,” “not good looking”). We use the terms “DDF,” “clean,” and “healthy” throughout this paper because those are the terms used by the sex ad authors. Since some characteristics were present in very few ads, categories were combined when analyzing the data. For characteristics, the race categories were combined to include only white, black, and other, and “DDF,” “clean,” and “healthy” were combined in one DDF status category.
Biases were defined as an ad in which the author specifically reported not wanting a characteristic in a sex partner or an ad that used stigmatizing language. The ad had to contain language stating “no X” or “X only,” with X representing a specific characteristic. For example, an ad was coded as including a bias if it included language such as “no HIV positive guys” or “white men only.” The following biases were collected: racism (saying no to black/Latino/Asian/other partners), ageism (saying no to a particular age group/range), weightism (saying no to “fat” or “underweight” men), heightism (saying no to tall or short men), transphobia (saying no to transgender people), physical appearance (saying no “ugly” men), gender expression (saying no “feminine” men or “no femmes”), HIV stigma (saying no “positive” men), and DDF status (saying must be “DDF,” “clean,” or “healthy”). DDF status was included as a bias because the terms “DDF,” “clean,” and “healthy” were considered stigmatizing language [
The self-reported characteristics of the ad authors are described in
Overall, there were very few explicit reports of biases. Bias against men who are not “DDF” was the mostly commonly reported, with 24.55% (540/2200) of ads mentioning the need for a “DDF,” “clean,” or “healthy” partner. There were also more biases against physical appearance than most other biases with 4.36% (96/2200) of ads containing bias against “ugly” men. Weightism, which almost exclusively comprised bias against “fat” men, was reported in 2.32% (51/2200) of ads. Bias against gender expression, comprising bias against “feminine” men, was reported in 1.9% of the ads. Among ads with ageist biases (34/2200, 1.55%), most reports were against older men (32/34, 94.1%). There was very little racial bias reported (7/2200, 0.32%); these biases were reported against white men (n=4) and black men (n=3). Homophobia was the lowest reported bias with only 1 ad (1/2200) expressing bias against homosexual men (0.05%). No ads contained reports of bias against height, transgender people, or HIV status.
Variations in reported biases by city are presented in
Biases stratified by Northeastern cities (N=800).a
Biases ( |
MSAs with the highest HIV prevalence in the United States—Northeast | ||||
Baltimore |
New York |
Philadelphia |
Washington |
Total |
|
Racism ( |
2 (1.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) | 3 (0.4) |
Ageism ( |
3 (1.5) | 0 (0.0) | 3 (1.5) | 0 (0.0) | 6 (0.8) |
Weightism ( |
7 (3.5) | 2 (1.0) | 2 (1.0) | 2 (4.0) | 13 (1.6) |
Physical appearance ( |
4 (2.0) | 10 (5.0) | 9 (4.5) | 6 (3.0) | 29 (3.6) |
Gender expression ( |
5 (2.5) | 3 (1.5) | 2 (4.0) | 3 (1.5) | 13 (1.6) |
Homophobia ( |
0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 1 (0.1) |
DDF status ( |
55 (27.5) | 48 (24.0) | 33 (16.5) | 55 (27.5) | 191 (23.9) |
an=200 in each city.
b
Biases stratified by Southern cities (N=800).a
Biases ( |
MSAs with the highest HIV prevalence in the United States—South | ||||
Atlanta |
Dallas |
Houston |
Miami |
Total |
|
Racism ( |
1 (0.5) | 1 (0.5) | 1 (0.5) | 1 (0.5) | 4 (0.5) |
Ageism ( |
1 (0.5) | 1 (0.5) | 3 (1.5) | 4 (2.0) | 9 (1.1) |
Weightism ( |
0 (0.0) | 5 (2.5) | 6 (3.0) | 7 (3.5) | 18 (2.3) |
Physical appearance ( |
6 (3.0) | 11 (5.5) | 7 (3.5) | 5 (2.5) | 29 (3.6) |
Gender expression ( |
2 (1.0) | 5 (2.5) | 3 (1.5) | 4 (2.0) | 14 (1.8) |
Homophobia ( |
0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
DDF status ( |
40 (20.0) | 61 (30.5) | 52 (26.0) | 46 (23.0) | 199 (24.9) |
an=200 in each city.
b
Biases stratified by Midwestern and Western cities (N=600).a
Biases ( |
MSAs with the highest HIV prevalence in the United States—Midwest and West | |||
|
Chicago |
Los Angeles |
San Francisco |
Total |
Racism ( |
0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
Ageism ( |
4 (2.0) | 10 (5.0) | 5 (2.5) | 19 (3.2) |
Weightism ( |
9 (4.5) | 3 (1.5) | 6 (3.0) | 18 (3.0) |
Physical appearance ( |
8 (4.0) | 24 (12.0) | 6 (3.0) | 38 (6.3) |
Gender expression ( |
2 (1.0) | 4 (2.0) | 7 (3.5) | 13 (2.2) |
Homophobia ( |
0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
DDF status ( |
45 (22.5) | 49 (24.5) | 56 (28.0) | 150 (25.0) |
an=200 in each city.
b
Variations in biases by author characteristics are presented in
Bias against feminine men varied significantly by the sexual orientation of the ad author (
Bias against men who are not “DDF” varied by HIV status (
Weightism varied significantly by the DDF status of the author (
Variations in biases by author characteristics for all cities (N=2200).
Characteristics of ad |
Biases | ||||||||
|
|
Racism | Ageism | Weightism | Physical appearance | Gender expression | Homophobia | DDF | Total |
|
.11 | .87 | .06 | .11 | .09 | .99 | .79 |
|
|
|
White, n (%) | 2 (0.4) | 4 (0.7) | 10 (1.9) | 23 (4.3) | 7 (1.3) | 0 (0.0) | 125 (23.0) | 541 |
|
Black, n (%) | 0 (0.0) | 4 (2.7) | 6 (4.1) | 1 (0.7) | 7 (4.8) | 0 (0.0) | 35 (24.0) | 146 |
|
Other, n (%) | 0 (0.0) | 3 (1.8) | 5 (0.9) | 11 (2.0) | 5 (0.9) | 0 (0.0) | 39 (7.2) | 166 |
|
Not mentioned, n (%) | 2 (0.3) | 23 (1.8) | 30 (2.2) | 61 (4.5) | 23 (1.7) | 1 (0.1) | 275 (25.1) | 1347 |
|
.35 | .006 | .85 | .09 | .10 | .049 | 0.88 |
|
|
|
18-25, n (%) | 0 (0.0) | 11 (2.9) | 10 (2.6) | 20 (5.3) | 12 (3.2) | 0 (0.0) | 96 (25.2) | 381 |
|
26-35, n (%) | 0 (0.0) | 17 (2.5) | 17 (2.5) | 39 (5.7) | 17 (2.5) | 0 (0.0) | 177 (25.7) | 689 |
|
36-45, n (%) | 1 (0.2) | 1 (0.2) | 9 (1.7) | 21 (3.9) | 6 (1.1) | 0 (0.0) | 131 (24.4) | 537 |
|
46+, n (%) | 3 (0.8) | 5 (1.3) | 10 (2.6) | 9 (2.3) | 5 (1.3) | 0 (0.0) | 87 (22.7) | 384 |
|
Not mentioned, |
2 (1.0) | 0 (0.0) | 5 (2.4) | 7 (3.4) | 2 (1.0) | 1 (0.5) | 49 (23.5) | 209 |
|
.10 | .46 | .30 | .93 | <.001 | .99 | .56 |
|
|
|
Homosexual, |
0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (14.3) | 7 |
|
Straight, |
0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (3.1) | 2 (6.3) | 0 (0.0) | 11 (34.4) | 32 |
|
Bisexual, |
0 (0.0) | 5 (3.7) | 6 (4.4) | 6 (4.4) | 9 (6.7) | 0 (0.0) | 36 (26.7) | 135 |
|
Not mentioned, |
7 (0.4) | 29 (1.4) | 45 (2.2) | 89 (4.4) | 31 (1.5) | 1 (0.1) | 492 (24.3) | 2026 |
|
.26 | .96 | .77 | .47 | .89 | .93 | <.001 |
|
|
|
Negative, n (%) | 2 (0.7) | 5 (1.7) | 8 (2.8) | 16 (5.6) | 6 (2.1) | 0 (0.0) | 96 (33.3) | 228 |
|
Positive, n (%) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 2 (22.2) | 9 |
|
Not mentioned, |
5 (0.3) | 29 (1.5) | 44 (2.3) | 80 (4.2) | 36 (1.9) | 1 (0.1) | 532 (28.0) | 1903 |
|
.13 | .04 | .04 | .27 | .37 | .93 | <.001 |
|
|
|
DDF, n (%) | 3 (0.4) | 13 (1.9) | 22 (3.2) | 5.6 (39) | 15 (2.1) | 0 (0.0) | 275 (39.4) | 698 |
|
Not mentioned, |
4 (0.3) | 21 (1.4) | 29 (1.9) | 3.8 (57) | 20 (1.8) | 1 (0.1) | 265 (17.7) | 1502 |
|
.67 | .10 | .58 | <.001 | .21 | .65 | .14 |
|
|
|
Good looking, |
1 (0.3) | 10 (2.8) | 10 (2.7) | 44 (12.0) | 10 (2.7) | 0 (0.0) | 91 (24.7) | 368 |
|
Not mentioned, |
3 (0.2) | (24 (1.4) | 41 (2.2) | 52 (2.8) | 32 (1.8) | 1 (0.1) | 449 (24.5) | 1832 |
Total |
|
7 (0.3) | 34 (1.6) | 51 (2.3) | 96 (4.4) | 42 (1.9) | 1 (0.1) | 540 (24.6) | 2200 |
These findings provide insight into the representation of biases and intragroup stigma among MSM using Craigslist to seek sex with other men. Overall, very few biases were reported. This could be indicative of the unique formatting of Craigslist ads. Since there is no prescribed form for authors to complete, we found great variation in how ads were presented. In most cases, ads included very little information, resulting in limited reports of both author characteristics and biases. However, biases that were present still showed patterns and variation.
DDF bias was the most pervasive. The saliency of using the term “DDF” is consistent with previous research examining Craigslist Men Seeking Men sex ads [
Another important finding from these data is that when men reported sexual orientation, most men identified as straight or bisexual; bias against feminine gender expression was only present in ads by these authors. This bias was only present in 1.9% of ads overall, but this finding provides insight into who is perpetuating bias against feminine gender expression. Previous research has identified a subgroup of non-gay-identified men on Craigslist who seek sex from other non-gay-identified men (who they may believe will present as more stereotypically masculine) [
The presence of stigma in online sex ads may contribute to poor mental health and increased sexual risk for those who are seeking sex online. Men who have characteristics that are described in ads as undesirable may experience a fear of rejection, loneliness, and reduced self-esteem. These men may also perceive themselves as having less bargaining power when negotiating sex, possibly increasing sexual risk; previous research has examined how homophobic discrimination may influence behaviors associated with higher risk for HIV, including nondisclosure of HIV status [
The ad authors represented in this study are limited to men who are actively seeking sex partners online. Men who seek sex partners on Craigslist differ in characteristics from men seeking partners offline [
Despite an overall low reporting of biases in ads, these findings provide insight into patterns of stigma represented on Craigslist Men Seeking Men sex ads. These findings suggest that there is a need to address intragroup stigma within MSM communities; it is important to focus on HIV-related stigma among MSM, but it is also useful to understand other forms of intragroup stigma and how they may influence mental health outcomes and sexual risk-taking behaviors, especially for MSM who are seeking sex online. Understanding patterns in the perpetuation of intragroup stigma can help to better target messages aimed at making cultural and behavioral shifts in the perpetration of intragroup stigma within MSM communities.
Descriptive statistics of sample characteristics.
disease and drug free
Enhanced Comprehensive HIV Prevention Planning
metropolitan statistical area
men who have sex with men
sexually transmitted infection
None declared.