Gender Expression as a Basis for Employment Discrimination in Gay, Lesbian, and Bisexual Populations

Note: This is a preprint version of a paper submitted to the International Journal of Transgenderism on August 7, 1999. Until it is accepted for publication, it is copyrighted © 1999 by Sarah Fox, Ph.D., all rights reserved. The contents of this page may not be reproduced or redistributed in any form and are available here for informational purposes only.

by Sarah D. Fox, Ph.D.
Chrysalis Continuing Education Services
and It's Time, Ohio!

Correspondence and requests for materials should be sent to:
Sarah Fox
% It's Time, Ohio!
P.O. Box 21310
Columbus, Ohio 43221, USA
(614) 470-4245
email: sarah@gender.org

• Abstract
• Introduction
• Methods
• Results and Discussion
• References

ABSTRACT
In the US, national gay, lesbian, bisexual, and transgender (GLBT) political organizations are currently urging passage of the Federal Employment Non-Discrimination Act (ENDA), which would prohibit employment discriminations on the basis of sexual orientation but not gender expression. A survey by the NGLTF, the NCLR, and GenderPAC indicated that this gap in coverage would leave 37% of the GLB (and not transgender) employment discrimination victims vulnerable to continued discrimination. In the current study, a similar survey was conducted of participants at the 1999 Pride March in Columbus, Ohio. Of 80 respondants, 61% had experienced employment discrimination, with 76% of that group citing gender expression as a basis. This figure is markedly higher than that found by NGLTF/NCLR/GenderPAC and clearly indicates that ENDA's language should be expanded to include protections based on gender expression.

INTRODUCTION
Gays, lesbians, bisexuals, the transgendered, the intersexed, and other gender-variant people have long faced employment discrimination in the United States and elsewhere based on their sexual minority status. In the 106th US Congress, the Federal Employment Non-Discrimination Act (ENDA) has been re-introduced that if enacted would "prohibit employment discrimination on the basis of sexual orientation." (H.R. 2355; see http://thomas.loc.gov/cgi- bin/query/z?c106:H.R.2355.IH:). While ENDA promises to offer employment protection to gays, lesbians, and bisexuals (GLB people), those protections may be incomplete. Many gay men have effeminate mannerisms or styles of dress, and many lesbian women have unusually masculine mannerisms, hair styles, and manners of dress. Thus, they are guilty of breaking societal "rules" of gender expression in the same manner as transgendered people. Their violation of gender expression norms is itself objectionable to many "straight" people (people of sexual majority) and therefore would be a likely basis for employment discrimination.
Discrimination on the basis of gender expression (irrespective of sexual orientation) would not be covered under the language of ENDA, leaving many GLB people subject to the same employment difficulties as the transgendered and intersexed. Leaders of the US transgender community have lobbied persistently for inclusion of language in ENDA that would extend protections on the basis of gender expression, thereby filling this gap in ENDA's coverage and providing protection for transgendered and intersexed people too. However, their efforts have been rebuffed by key lawmakers and by the Human Rights Campaign (HRC), which has been a major proponent of ENDA.
In a joint effort by the National Gay and Lesbian Task Force (NGLTF), the National Center for Lesbian Rights (NCLR), and GenderPAC, attendees of the 1999 NGLTF Creating Change Conference in Pittsburgh, PA were surveyed to determine the prevalence of sexual orientation based and gender expression based employment discrimination in their population (Wilchins, 1999a). Out of 120 respondants, 103 did not indicate having transgender issues. Because of the GLBT focus of the conference, the latter group likely had a high GLB representation. Of that 103, 58% had experienced job discrimination based on either sexual orientation (real or perceived) or gender expression. Sexual orientation had been the only basis for discrimination against 63% of that group, gender expression had been the only basis for 8%, and both sexual orientation and gender identity had been the bases for 29% . Thus, 37% of the group had experienced job discrimination based at least in part on gender expression; these people would not have been protected under the current language of ENDA.
The sample base of this survey was later expanded (Russell and Wilchins, in preparation; Wilchins, 1999b) with nearly identical results: Of over 275 total respondants (again, likely to include a high number of GLB people), 58% reported having suffered employment discrimination. Of the nontransgendered respondants having reported employment discrimination, gender expression was a basis of discrimination for 41% of the respondants.
Although these findings are compelling, the data had some difficult problems. First, it was not certain whether respondants not indicating transgender status were indeed GLB. Second, respondants were mostly civil rights activists and thus had been sampled from an artificially selected subpopulation. Third, only 120 of the 400 distributed surveys were returned, thus introducing a self-selection bias. In order to obtain data from a more representative sample of GLB people and to overcome some of these problems, a survey was conducted at the 1999 Pride celebration in Columbus, Ohio. Besides being useful for replication purposes, the resulting data were intended to provide Ohio-specific information for use in statewide lobbying efforts.

METHODS
Subjects: The 1999 Columbus Pride celebration was broadly attended, with an estimated attendance in excess of 10,000 (Scahill, 1999). Approximately 2/3 of the survey forms were distributed by the author to most of the GLB individuals in three separate areas of the gathering. The remainder of the survey results were obtained from individuals visiting the It's Time, Ohio! information booth at the end of the parade route.
Survey Forms: The half-page survey forms had operations definitions of both sexual orientation and gender expression, along with examples of discrimination scenarios on both bases. This information was placed conspicuously at the top of the survey in a box. Response areas were found immediately below the large box (see Figure 1).

Figure 1. The full text of the employment discrimination survey.

It's Time, Ohio! EMPLOYMENT DISCRIMINATION SURVEY PLEASE READ!! ...... GLBT people frequently experience employment discrimination, including unfair hiring, firing, promotion, demotion, pay, and job assignments. Employment discrimation against GLBT people may be based on two factors: sexual orientation -- the fact that someone is gay, lesbian, bisexual, or straight, irrespective of how he or she behaves, appears, or presents him or herself in the workplace. (e.g. "Bill, I don't like it that you have sex with other men. It's unnatural, and you're fired.") gender expression -- how a person behaves, appears, or presents him or herself with regard to societal expectations of gender. (e.g. "Bill, your limp wrists and 'gay' speech mannerisms bother me. You're a poor excuse for a man, and you're fired.")

Name:______________________________________ I am (check one): ___ GLB (gay, lesbian, or bisexual) ___ transgendered (crossdresser, transsexual, drag queen, etc.) ___ both transgendered and GLB I have experienced employment discrimination based on (check one): ___ both sexual orientation and gender expression ___ sexual orientation only ___ gender expression ___ I have not experienced discrimination on the basis of either sexual orientation or gender expression.

Verbal Instructions: All respondants were told of the importance of reading the definitions of sexual orientation and gender expression. They were observed as they took the survey, to ensure that they indeed read the definitions.
Statistics: Data were evaluated with simple chi-square tests and Z statistics on binomial probabilities. Only two-tailed probabilities were used.

RESULTS AND DISCUSSION
All respondants apparently complied with verbal directions, and no respondants expressed confusion over the definitions, except for one "straight" man who was excluded from the sample on the basis of his sexual majority status. Only three individuals declined to take the survey. Considering the size of the gathering and the selection methods used, the results are believed to be relatively unbiased and uncomplicated by extraneous variables.
A total of 89 forms were distributed and collected. Nine of those forms were excluded from the sample, either because responses were not given or because the respondants indicated they were transgendered. Data from the 80 valid responses, compared to the corresponding data collected at the Creating Change conference, are listed in Figure 2.

Figure 2. (a) Data collected at Pride, Columbus, Ohio. (b) Data collected at Creating Change. GE = Gender Expression. SO = Sexual Orientation. Answer of "yes" or "no" indicated by "+" and "-," respectively.

a.		SO -	SO +	Total		b.		SO -	SO +	Total
	GE -	21	12	43			GE -	43	38	81
	GE +	7	30	37			GE +	5	17	22
	Total	38	42	80			Total	48	55	103

The total frequency of employment discrimination was remarkably similar between the Columbus Pride and Creating Change populations (61 and 58%, respectively), as shown in Figure 3. This difference is nonsignificant (c2=0.168, 1 df), suggesting that both populations had been similarly marginalized in their employment history.

Figure 3. A comparison of employment discrimination rates between the two populations. ("Discr." = Discrimination; see Figure 2 for other abbreviations.)

	Discr -	Discr +	Total
Columbus	31 (39%)	49 (61%)	80
Pittsburgh	43 (42%)	60 (58%)	103
Total	74	109	183

Despite the similarity in rate of employment discrimination between the two samples, the pattern of discrimination was markedly different, as shown in Figure 4. With respect to the groups that had suffered discrimination, a much higher proportion of the Columbus Pride respondants (76%) than Creating Change respondants (37%) indicated that gender expression had been the basis for employment discrimination against them. This difference was highly significant (c2=40.74, 1 df, p<<.01)

Figure 4. Incidence of discrimination based on gender expression, compared between populations. (See previous figures for abbreviations.)

	GE -	GE +	Total
Columbus	12 (24%)	37 (76%)	49
Pittsburgh	38 (63%)	22 (37%)	60
Total	50	59	109

While the Creating Change population had a much higher frequency of discrimination based on sexual orientation than on gender expression (See Figure 5; Z=4.32, p<.0001), there was no significant difference between these figures for the Columbus Pride population (Z=.625, ns).

Figure 5. Comparison of rates of sexual-orientation-based discrimination and gender-expression-based discrimination between the two populations. (Note that the GE+ and SO+ categories are not mutually exclusive. See previous figures for abbreviations.)

	GE +	SO +	Total
Columbus	37 (76%)	42 (86%)	49
Pittsburgh	22 (37%)	55 (92%)	60

The reasons for the differences between these populations might be attributable to differences in surveying techniques. At the Creating Change Conference, respondants were not individually instructed to heed definitions relating to differences between gender expression and sexual orientation. Thus, many hurried respondants might have filled out answers without understanding the terminology. An understanding of these definitions would not affect one's response as to whether one had experienced employment discrimination based on sexual minority status. However, respondants unfamiliar with distinctions between gender expression and sexual orientation might have been more inclined to attribute discrimination to the familiar factor "sexual orientation" than to the relatively unfamiliar "gender expression." If this interpretation is correct, the Ohio data may more accurately reflect the extent of employment discrimination base on gender expression.
The implications of these data are that only 24% of Ohio's GLB employment discrimination victims would have been fully protected by legislation pertaining only to sexual orientation. Under the current language in the Federal ENDA bill, 76% (95% confidence limits from 54 to 97%) of discrimination victims would still be victimized.
This figure is very disturbing in light of the 61% total victimization rate (95% confidence limits from 48 to 75%). If these figures are accurate, the legislative policy of excluding gender expression language from the current ENDA bill will mean that almost half of the GLB population (46%; limits from 36 to 56%) will continue to be victimized by employment discrimination without any protections under the law. In light of these findings, it would prudent not only for the transgender community but even for the GLB community to reevaluate this exclusionary policy.

REFERENCES

King, Loren (1999) Task Force drops support for federal anti-bias bill. Bay Windows, June 10, 1999.

Scahill, Andy (1999) Pride Holiday 1999 Still a Success. Stonewall Journal (Columbus, OH) 10(6):4-6.

Russell, Stephen T., and Wilchins, Riki A. (in preparation) Gendered discrimination: Sexual orientation and gender as bases for employment discrimination in a national survey. Working paper, Department of Human and Community Development, University of California, Davis.

Wilchins, Riki A. (1999a) Unpublished data, NGLTF, NCLR and GenderPAC.

Wilchins, Riki A. (1999b) First National Gender and Employment Discrimination Survey: Major Results Report. GenderPAC: Washington, DC.

Back Next Webpage © 1999, Arctic Fox Designs