Discrimination against LGBTI People in the Workplace

Work environment has the most precise legislation in the European Union regarding discrimination against LGBTI people. Yet, a European-wide survey conducted by the European Agency for Fundamental Rights (^{FRA, 2012}) revealed that 16% of LGBTI persons surveyed in Spain reported having felt discriminated against at work due to their sexual orientation or gender identity.

In the above-mentioned study carried out in 2013 by the Spanish Federation of Lesbian, Gay, Transgender, and Bisexual People (^{FELGTB-COGAM, 2013}), this percentage increased to 31.2%. This may be due to the fact that participants in this study were mainly members of LGBTI organizations, who are theoretically more sensitive to discrimination. In the United States, a review of fifty studies examining job discrimination against LGBT individuals found that 16% to 68% of LGBT people reported discrimination at work (^{Badgett et al., 2007}). Data from a national probability survey representative of the U.S. population in 2008 also showed that 38% of LGB employees reported being harassed at work, and 27% experienced employment discrimination based on their sexual orientation (^{Sears & Mallory, 2011}). Discrimination against LGBTI workers is reflected in many indices, such as those related to hiring decisions or salaries. Accordingly, ^{Flage (2019)} found, in a meta-analysis about discrimination against gays and lesbians in hiring decisions in OECD countries, that openly homosexual applicants faced similar discrimination as ethnic minority applicants. Discrimination in the selection process for LGBTI individuals was also significantly greater for low-skilled than for high-skilled job candidates, and significantly higher in European countries than in North America. Another meta-analysis of studies published between 1995 and 2012 about the effects of sexual orientation on earnings (^{Klawitter, 2015}) found that gay men earned less and lesbians earned more than their heterosexual counterparts. Sexual orientation discrimination was an important explanation for the results of gay men. However, explanations were less conclusive regarding the results of lesbian women (e.g., lesbian women tend to have higher educational level, heterosexual women tend to be more subjected to traditional gender roles and may therefore give greater priority to their family than their career).

Specifically, studies reveal that LGBTI individuals report discrimination at work, which is associated with lower physical and emotional well-being, and negative work results (^{DeSouza et al., 2017}). In Spain, a study conducted by ^{Di Marco et al. (2018)} using 39 in-depth semi-structured interviews with LGBTI employees found that they experienced incivility at work in the form of jokes, use of language, stereotypes, and nosy behaviors. Although subtle forms of prejudice appear to predominate over hostile and overt ones in the workplace, overt forms of prejudice still occur with a high frequency. Specifically, in the aforementioned study by ^{FELGTB-COGAM (2013)}, the types of discrimination experienced by those who reported feeling discriminated against at work were (in decreasing order): jokes (72.73%), discriminatory treatment by colleagues (47.52%), harassment (23.97%), obstacles to career advancement (20.25%), and obstacles in their access to work (19.83%). Similarly, ^{Molero et al. (2017)} conducted a study with a Spanish sample of lesbian women and gay men. Participants reported experiencing subtle discrimination (an example of item measuring this dimension is “Even when people seem to accept LG persons, I think that, deep down, they have some misgivings”) to a greater extent than overt discrimination (an example of item capturing this dimension is “In Spanish society LG people are visibly rejected”). However, differences were not considerable (mean scores were 3.34–subtle–vs. 3.09–overt–in the entire sample). In another study, ^{Molero et al. (2013)} had previously found in various stigmatized groups in Spain (e.g., Latin American immigrants, Romanian immigrants, people with HIV, gays and lesbians) that subtle forms of discrimination were more related to two measures of psychological well-being–affect balance and self-acceptance–than blatant forms of discrimination. In any case, the negative consequences of both subtle and overt forms of prejudice seem to be rather similar (^{Jones et al., 2016}).

Therefore, it seems important to gain further insight on the job discrimination experienced by LGBTI employees. It is key to determine whether these experiences actually take place and, if so, to assess their level of impact on work stress and psychological well-being of LGBTI individuals, testing the main contribution of the minority stress model. One of the first questions we intended to address was whether LGBTI employees perceive discrimination due to their sexual orientation to a greater or lower extent than heterosexual employees. Our first hypothesis (H1) was that LGBTI employees, compared to heterosexual ones, would report experiencing higher sexual orientation job discrimination. Next, we analyzed work stress–a possible consequence of job discrimination against LGBTI employees–and its potential influence on psychological well-being focusing on two of its indices: common mental disorders and depression.

Work Stress among LGBTI Employees

Discrimination of LGBTI workers due to their sexual orientation is likely to be reflected in higher levels of work stress. ^{Lazarus and Folkman (1984)} defined stress as “a particular relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being” (p. 19). It is common to differentiate between individual and social stressors. The former are related to the events that can occur to a particular person at an individual level, such as loss of a job or a loved one. By contrast, social stressors refer to conditions in the social environment, such as the belongingness to stigmatized social categories (e.g., based on socioeconomic status, race/ethnicity, gender, or sexuality).

According to the minority stress model (^{Meyer, 1995}, ²⁰¹³), the stressors that LGBTI people can experience vary in a continuum, from distal to proximal. Distal minority stressors can be defined as objective stressors that do not depend on an individual's perceptions or appraisals—although certainly their report depends on perception and attribution. Stressors related to expectations of rejection and discrimination are located in the middle of the continuum, including the surveillance that these expectations require and, sometimes, the concealment of the minority identity. Finally, proximal stress processes are more subjective and therefore related to self-identity as a lesbian, gay, or bisexual person (e.g., internalized homophobia). According to this model, circumstances in the environment and/or having a certain minority status lead to exposure to stressors —some general and others specific— in minority group members. These stressors can positively or negatively affect mental health outcomes, depending not only on their intensity but also on individuals' coping strategies and social support available.

Some of the main job stressors are related to the job itself, such as job contentment (e.g., fragmented or meaningless work), workload and work pace, work schedule, roles in organization (e.g., role ambiguity or conflict), and low participation in decision-making. Other job stressors are more related to the organizational culture (e.g., poor communication, low levels of support for problem-solving and personal development), interpersonal relationships at work (e.g., social or physical isolation, interpersonal conflict, lack of social support, bullying, harassment), and career development (e.g., career stagnation and uncertainty, under- or overpromotion, job insecurity, low social value of work) (^{Leka & Jain, 2010}).

Stress is one of the factors with the highest influence on well-being at work (^{Bliese et al., 2017}). Specifically, work stress has been associated with mental health concerns (^{Bonde, 2008}; ^{Stansfeld & Candy, 2006}) and poor physical health (^{Kivimäki et al., 2002}). With this in mind, ^{Nixon et al. (2011)} conducted a meta-analysis of 79 studies analyzing the relationships between job stressors (e.g., organizational constraints, interpersonal conflict, role conflict) and physical symptoms (e.g., backache, headache, eye strain, disturbed sleep, dizziness, fatigue, loss of appetite, gastrointestinal problems). They found a significant relationship between job stressors and physical symptoms in cross-sectional and longitudinal analyses. Occupational stress has also been associated with diminished performance (^{Sullivan & Bhagat, 1992}; ^{Virtanen et al., 2009}) and poorer safety results in the workplace (^{Spurgeon et al., 1997}).

^{Meyer (2013)}, after reviewing research evidence on the prevalence of mental disorders in lesbians, gay men, and bisexuals (LGBs) using meta-analyses, concluded that LGB individuals are exposed to excess stress due to their minority position and that this stress leads to mental disorders. Thus, taking the aforementioned references into account showing the persistence of prejudice towards LGBTI individuals and their experience of minority stress, our second hypothesis (H2) was that LGBTI employees would report higher levels of work stress than heterosexual ones.

Psychological Well-being among LGBTI Employees

Common mental disorders, most often involving mood, anxiety, and drug use disorders (^{Steel et al., 2014}), are very widespread and represent a considerable proportion of the global burden of disease (^{World Health Organization - WHO, 2008}). Data from the World Mental Health Surveys reveal that 12% to 47% of a country's population will suffer one or more mental disorders in their lifetime (^{Kessler et al., 2007}). These disorders include insomnia, anxiety, fatigue, irritability, depressive mood, difficulty concentrating, and somatic complaints. Mental disorders are a serious public health concern (^{World Health Organization - WHO, 2003}) since they are well-known causes of major functional disability and distress. They also have a major socioeconomic impact due to work absenteeism and the increased demand on health services.

A particularly relevant mental problem is depression, which is “a mental disorder that presents with depressed mood, loss of interest or pleasure, decreased energy, feelings of guilt or low self-worth, disturbed sleep or appetite, and poor concentration” (^{Marcus et al., 2012}, p. 6). Major depression is one of the most arduous health conditions, both at individual and population levels (^{Vilagut et al., 2016}). It is also the most common mood disorder, with a lifetime prevalence reported to range between 7% and 21% (^{Bromet et al., 2011}). Major depression is also related to a considerable functional disability, reduced quality of life, an increased burden —both for patients and caregivers—, and a greater risk of death (^{Eaton et al., 2008}). According to the World Health Organization (^{WHO, 2017}), the proportion of the worldwide population affected by depression in 2015 was estimated at 4.4%, with a higher prevalence among women (5.1%) than men (3.6%). Its prevalence also varies depending on geographical location and age: it is higher in older people, but depression is also present in adolescents.

As indicated previously, many studies have shown a higher prevalence of depression and anxiety disorders among lesbian, gay, and bisexual groups when compared to heterosexuals (e.g., ^{Bostwick et al., 2014}; ^{Meyer, 2013}). Thus, our third hypothesis (H3) was that LGBTI individuals —versus heterosexual individuals— would report higher levels of mental disorders and depression.

The Present Research

According to ^{Meyer (2013)}, evidence of the stress suffered by LGBTI people is based on two methodological approaches: studies examining within-group processes and their impact on mental health and studies comparing the prevalence of mental disorders between minority and non-minority groups. Our research falls within the second approach. However, in studies on between-group differences, only exposure —minority status–and outcomes—prevalence of disorder–are usually assessed. Unfortunately, minority stress processes that may lead to a higher prevalence of disorders are inferred but not analyzed (^{Meyer, 2013}). One of the contributions of the present research is the fact that participants are explicitly asked not only about their minority status, but also about their experience of discrimination at work and their stress levels, providing a direct test of the minority stress model.

The literature review included in the previous sections showed the existence of the following relationships: a) between being an LGBTI person and experiencing psychological and well-being problems (e.g., ^{Bostwick et al., 2014}); b) between experiencing discrimination at work for being LGBTI and physical and psychological well-being (e.g., ^{DeSouza et al., 2017}); c) between discrimination and stress in general (e.g., ^{Bruce et al., 2015}); and d) between stress and psychological well-being (e.g., ^{Nixon et al., 2011}). However, no studies have addressed all these constructs in relation to each other in LGBTI people focusing on work environment. Moreover, studies conducted in Spain have found that there is perceived (manifest or subtle) discrimination of LGBTI persons in general (^{Domínguez-Fuentes et al., 2012}; ^{Molero et al., 2017}) and specifically in the workplace (^{Di Marco et al., 2018}; ^{FELGTB-COGAM, 2013}). However, the relationship between perceived discrimination, stress, and well-being in LGBTI workers has not been analyzed. In this research we tried to include all the aforementioned relationships. Thus, we expected to find (Hypothesis 4) that these higher levels of mental disorders and depression among LGBTI people would be caused by work stress resulting from the discrimination experienced for being LGBTI. Specifically, we proposed the following mediation model: being LGBTI will lead to greater sexual orientation discrimination, resulting in higher stress levels and consequently in an increase of common mental disorders and depression. The main contribution of our research, in addition to showing that LGBTI people in Spain still perceive discrimination at work, is to highlight the mediating role of job discrimination and the stress experienced at work for being LGBTI. That is, being a LGBTI worker does not necessarily lead to psychological and well-being problems in itself; it does so through being discriminated at work, which increases work stress.

Participants

The study was conducted with 377 participants, although 11 did not answer the question about sexual orientation and were removed from the study. Of those remaining, 137 participants defined themselves as heterosexual, 134 as gay, 61 as lesbian, and 34 as bisexual. Individuals self-defined as gay, lesbian, or bisexual were grouped into the same category, with a total of 229 LGBTI individuals in the scale answered by the largest number of participants. Regarding level of education, 62.1% had a university degree, 30.9% had completed secondary studies, and 7.1% had completed primary school. In terms of marital status, 46% were single, 28.4% married, and 6.8% separated or divorced. In the sample, 95.7% of participants were working at the time (70% full-time) and 4.3% were not employed. Regarding gender identity, 50.5% defined themselves as cis men, 37.2% as cis women, 1.7% as trans men, 2.9% as trans women, and 7.7% as non-binary.

We analyzed patterns of missing data for each scale. In total, 2.7% of participants did not answer the measure of work stress, 4.1% did not complete the CESD-7, and 6.3% did not complete the SRQ-20. Analyses were conducted with a sample of 356 participants for the measure of work stress, 351 for analyses concerning the SRQ-20 and 343 for the CESD-7. Supplemental analyses where missing data were replaced using multiple imputation produced no substantive differences from the results presented below.

Procedure

Participants were contacted through a mainstream Spanish trade union organization. Specifically, the LGBTI section of that union broadcast a message among its affiliates that included a link to access the survey. The message explained that the research, conducted by the Department of Social Psychology of a university in southern Spain and the trade union, was intended to assess the working conditions of LGBTI people and determine the psychosocial risks to which they are exposed. The anonymous nature of responses and confidentiality was clearly indicated, as well as the possibility of leaving the survey when participants deemed appropriate. It was clarified that when asked about work-related issues, participants should respond having in mind the job they currently had or, if they did not have a job, thinking about the last one they had had. Finally, participants were thanked for their participation.

Participants completed a questionnaire that included various measures and scales via the Qualtrics platform. For the present research we shall analyze the measures included below.

Perceived job discrimination on the basis of sexual orientation. We assessed this construct with a purpose-built measure that evaluated the extent to which participants felt that their sexual orientation may have caused a problem or difficulty in several job-related aspects (e.g., promotion, relationships). Two items were selected from a survey by the Center for Sociological Research [Centro de Investigaciones Sociológicas] and another item was selected from a survey by the Spanish Federation of LGBT people and the Madrid LGBT Collective (^{FELGTB-COGAM, 2013}, p. 17) about discrimination due to sexual orientation and/or gender identity in Spain. This measure comprised 7 items that presented potential situations of job discrimination due to sexual orientation, with which participants had to show their agreement or disagreement on a scale ranging from 1 (totally disagree) to 5 (totally agree). The items of this scale were the following: [my sexual orientation] “has hindered me when applying for a job”, “has caused me problems in my relationship with my colleagues”, “has caused problems in the recognition of my work”, “has harmed me in accessing to positions of responsibility in my job”, “has led me to be the target of jokes at work”, “has caused me workplace harassment”, and “has made me be treated discriminatorily by colleagues”.

Higher scores in this measure indicate greater perceived discrimination. A total mean score on the scale was calculated, showing high internal consistency (α = .94) (α = .94 for the heterosexual subgroup, and α = .94 for the LGBTI subgroup).

Work stress. We used a Spanish adaptation (obtained through a back-translation procedure involving professional translators) of the Stress in General (SIG) scale by ^{Stanton et al. (2001)} to measure work stress using two dimensions: seven items to assess pressure (SIG-I) and eight items to assess threat (SIG-2). This measure has shown good convergent and discriminant validity. The measure is applied by presenting participants with a list of adjectives (e.g., “agitated”, “tense”) and asking them to consider whether or not this characteristic describes their workplace. To reply, participants must use a No (scored 0) or Yes (scored 3) response format. Higher scores indicate higher levels of work stress. In this study, the full scale showed high internal consistency (α = .87) (α = .86 in the heterosexual subgroup and α = .88 in the LGBTI subgroup). The internal consistency of the pressure subscale was α = .82 (α = .82 in the heterosexual subgroup and α = .83 in the LGBTI subgroup); the internal consistency of the threat subscale was α = .77 (α = .75 in the heterosexual subgroup and α = .77 in the LGBTI subgroup).

Common mental disorders. We used a screening tool to assess nonspecific psychological distress. Overall mental health was assessed with a Spanish translation of the Self-Reporting Questionnaire-20 (SRQ-20; ^{Harding et al., 1980}), which consists of 20 items referring to somatic, anxious, and depressive symptoms. This tool was developed by the World Health Organization to detect non-psychotic disorders and has since been widely used in many countries. It explores the various forms of emotional distress experienced by people in the last four weeks (e.g., symptoms of insomnia, fatigue, irritability, memory and concentration problems, as well as other somatic indicators such as headaches, tremors, or indigestion). The short format and dichotomous answers used (yes/no) make it a promising tool for the busy environment of primary care. Every Yes answer by a participant is scored with 1 point, resulting in a symptom score ranging from 0 to 20; higher values indicate lower mental health. Importantly, the cut-off point for considering the existence of a common mental disorder is 8 (^{Ludemir et al., 2008}). Numerous studies have shown the good psychometric properties of the SRQ-20 (^{Ludemir et al., 2008}). In this study we used the Spanish version by ^{Navarro-Mantas et al. (2018)}, provided by the World Health Organization. The scale had high internal consistency (α = .91) (α = .89 for the heterosexual subgroup and α = .91 for the LGBTI subgroup).

Depression. Given its relevance, we considered it essential to measure this specific form of mental disorder. To this end, we used the Center for Epidemiologic Studies Depression Scale (CESD-7), the short form of ^{Radloff's (1977)} CESD-20, as an instrument to measure depressive symptoms. This version was developed by ^{Herrero and Gracia (2007)} and validated in a Spanish general population sample. The scale was developed as a quick and reliable way to diagnose depression in the Spanish-speaking world and has shown adequate psychometric properties according to its authors. It consists of seven items that participants must answer about how they felt in the last week (e.g., “I felt as if I could not get rid of the sadness” or “I had trouble concentrating on what I was doing”). The scale has a response format ranging from 1 (rarely or never) to 4 (all or most of the time), with the intermediate options 2 (rarely or few times – 1/2 days) and 3 (a considerable number of times – 3/4 days). After inverting item 6, which is the only one drafted positively (“I enjoy life”), higher scores indicate a higher depressive mood. In this study the scale showed high internal consistency (α = .9) (α = .88 for the heterosexual subgroup and α = .96 for the LGBTI subgroup). Scores on this measure have been associated with sex (i.e., women tend to show higher scores than men), physical and mental health, and social integration (^{Herrero & Gracia, 2007}). Although this instrument is useful for measuring the risk of depression, it should not be considered as a measure of depression itself. It would be more appropriate to consider it as a measure of general distress and as a tool that facilitates the detection of depressive symptoms in clinically undiagnosed individuals who may be at risk of developing depression.