Between the city and the country: Heterogeneous victimization experiences among LGBTI individuals

Abstract

The wide range of victimization experiences and life conditions of LGBTI individuals (lesbian, gay, bisexual, trans, intersex), far from being homogeneous, shows a significant geographical variation. Within countries and global regions, but even within provinces and smaller territories, unequal experiences have caught both academic and media attention, providing key concepts and insights. The disparities between European Union member states, for instance, have fuelled comparisons between Western and Eastern countries. Drawing from Spanish survey data on victimization experiences and life conditions of LGBTI individuals, this paper explores how said experiences relate to the respondents’ place of residence, within a rural–urban continuum. The chi-squared test of independence and the use of standardized residuals allows us to reject the apparent homogeneity of experiences among LGBTI individuals, as well as to distinguish between the specificities of each participant category along the spatial continuum. This paper argues for a nuanced perspective that prioritizes disparities and inequalities rather than commonalities, while also considering the shared system of oppression that informs said differences.

Keywords:

• LGBTI
• test of Independence
• victimization
• Spain
• inequality
• rurality

PUBLIC INTEREST STATEMENT

LGBTI (lesbian, gay, bisexual, trans, and intersex) individuals experience varying life conditions and victimization rates, even within a single country or territory. How their experiences and life conditions vary along factors such as culture, place of residence, class, and political situation allow us to compare and analyze different countries and places. This article uses Spanish data from a European Union-wide survey in order to explore the relevance of the place of residence, as a way to understand the role of geography and of the rural-urban continuum in the experiences and life conditions of LGBTI individuals. Beyond the Spanish case, this article proves the relevance of geographical and social factors, as well as the need to consider the heterogeneity within the LGBTI community.

1. Introduction

Violence and discrimination experienced by LGBTI individuals (lesbian, gay, bisexual, trans, and intersex), and particularly those motivated by their perceived or actual sexual orientation and gender identity and expression, frequently involve long-term negative effects on their mental and physical health (Walters, 2014). Victimization experiences may even aggravate existing health issues and inequalities (Mereish & Son Taylor, 2021; Rees et al., 2021). These forms of violence, either physical or symbolic, may be compared to those not motivated by prejudice, and anti-LGBTI violence must be understood from the perspective of both direct and indirect victims and the consequences they endure. Throughout this paper, we use the term “victim” following its widespread use in the materials and data analyzed. This usage does not limit or bar contemporary debates that question or further consider the “victim” category (Chakraborti, 2018; Rothe & Kauzlarich, 2018). Drawing from these perspectives, we use “direct victim” to consider individuals directly affected by a violent or discriminatory practice, and “indirect victim” to refer to those individuals who have suffered further consequences due to their adscription to the group targeted. The experiences of direct victims may include physical damage, whereas both them and indirect ones may face psychological effects such as feelings of unsafety, anxiety, isolation, fear, loneliness, and depression (Walters, 2014, 2019). These experiences allow us to further study and understand the social dimension of trauma, if we are to understand “systemic trauma” as that historically and continuously endured by a group, with particular consequences regarding their feelings of safety, belonging, and dignity (Haines, 2019). How prejudices and either rational or irrational motives are deployed and experienced by the offenders of these experiences also signals towards the social and societal understanding of certain groups or minorities as deserving protection, or lack thereof.

Due to its current track record of equal marriage and other LGBTI-specific laws in several countries, the European Union (EU) may be seen as a global reference in political and activist goals. However, according to the EU Fundamental Rights Agency’s (FRA) last survey on LGBTI individuals’ victimization and life conditions (FRA: European Union Agency for Fundamental Rights, 2020a), more than a fifth of the respondents had been discriminated at work because of their sexual orientation or gender identity, and more than a tenth of them had endured a prejudice-based physical or sexual attack in the five years prior to the survey. Not only does the survey data counter a potentially idyllic image of the EU, but it also allows us to question the homogeneity of experiences among LGBTI individuals: the average of physical and sexual attack victimization rate was 11% throughout the survey, whereas this rate was of 17% and 22% for trans and intersex respondents, respectively. Similarly, 17% of intersex respondents had reported their most frequent discrimination experience, compared to 9% of bisexual women and to an average of 11% throughout the survey.

The survey data also signals towards the wide range of geographical disparities among the EU member states. For instance, when asked about their openness regarding their sexual orientation or gender identity, respondents who said they were “never open” reached 18% of those from the Netherlands, 23% of Spanish respondents, and 24% of those from Germany. On the other hand, the same response amounted 49% of Cypriot respondents, 53% of those from Romania, and 60% of those from Lithuania (FRA: European Union Agency for Fundamental Rights, 2020a). Similarly, when asked about discrimination experiences while searching for a job, victimization rates reached 17% in Bulgaria, 18% in Cyprus, and 19% in Greece, as opposed to Denmark’s and Sweden’s 5% and to Finland’s and the Netherlands’s 6% (ibid.). The fact that not all participant countries had nor have specific legislation on hate crimes and protection for specific minorities may play a significant role in the country-wise differences in victimization rates (FRA: European Union Agency for Fundamental Rights, 2021; Howard, 2019).

Spain’s data allows us to single out its recent history of legislation and relevance of its LGBTI activism. Despite critiques and instances for which the Spanish legislation could improve, such as those highlighted by the International Lesbian, Gay, Bisexual, Trans, and Intersex Association’s Europe barometer (ILGA Europe, 2022), the legal and political state of Spanish LGBTI individuals displays a positive or favourable evolution. The recent history of the local and nation-wide LGBTI social movements includes both cooperation and clashes with different political parties, but also a clear process in which the most relevant reform-oriented organizations have increasingly closer relations with public administrations (Jordi & Jubany, 2019; Martínez, 2017). Within this trajectory, however, most social movements singled out by historical research and accounts such as those by Ramón Martínez (2017) or Jordi Petit (2003) focus on large cities, such as Madrid and Barcelona. The centrality of large cities within LGBTI social movements is not limited to the Spanish case, as it has been seen as a consequence of “metronormativity”, or a series of discourses and expectations on how LGBTI individuals should live in large and dense urban areas (Stone, 2018).

Metronormativity, rather than as an abstract concept, should be seen as a relational dynamic according to which specific cities and urban areas, such as Madrid and Barcelona in Spain, attract individuals, investments, and cultural and media attention (Adiego & Antonio, 2020; Adiego et al., 2019; Langarita et al., 2021). This concentration should not be a surprise due to the historical relevance of Madrilenian and Barcelonan NGOs in these movements’ discourses and historical narratives (Martínez, 2017). This metronormative concentration, however, may also be seen in the academic concentration of spatial diversity in sociological and geographical studies of LGBTI experiences and realities (Dwyer et al., 2015; Forstie, 2020; Stone, 2018). Despite early academic research and publications focused on dense urban areas, and particularly on cities, such as London, New York City, Paris, or San Francisco, current research has explored LGBTI lives throughout a wide range of geographical settings. This research has focused on issues, such as rural trans masculinities (Abelson, 2016), the experiences of LGBTI youth (De Hulko & Hovanes, 2018; Pedro et al., 2018), migratory experiences, and motivations of LGBTI individuals (Thorsteinsson et al., 2022), and health disparities in rural areas or between rural and urban areas (Fisher et al., 2014; Glon et al., 2021; Horvath et al., 2014; Lyons et al., 2015). The geographical periphery of Western Europe (Butterfield, 2018; Kuhar & Švab, 2014), but also geographical heterogeneity within Catalonia in Spain (Adiego & Antonio, 2020; Adiego et al., 2019; Jubany et al., 2021; Langarita et al., 2021) have also attracted academic attention.

Drawing from current and recent research on the specificities and disparities of LGBTI realities and experiences throughout specific countries, this paper aims to study the spatial dimension of victimization experiences and life conditions faced by LGBTI individuals in Spain. The FRA survey’s data informs the quantitative analysis undertaken in this paper, which focuses on the intersection between the diverse range of daily experiences and victimization instances, on the one hand, and the different geographical places of residence within a rural–urban continuum. After this introduction, the paper explains the methodology employed: a statistical analysis of the dependence between the responses to a series of key questions and the identification of the respondents regarding their main category (based on sexual orientation or gender identity) and their place of residence. The null hypothesis of independence between categorical variables is tested on the secondary data produced by the EU Fundamental Rights Agency, and the chi-squared test results are further explored via the analysis of their standardized residuals. Finally, a series of conclusions contrast our results with recent research and signals towards potential future research, both qualitative and quantitative.

2. Methods and materials

This paper and its underlying analysis are informed by the EU Fundamental Rights Agency’s FRA LGBTI-II survey (FRA: European Union Agency for Fundamental Rights, 2020a). Conducted in 2019, this was the second edition of the FRA’s survey on victimization experiences and life conditions of LGBTI individuals. It obtained 139.799 valid participants, self-identified LGBTI individuals of at least 15 years old, living in the then-28 EU member states, plus Northern Macedonia and Serbia (ibid.). The survey was self-administered via an online platform, following a Europe-wide media campaign that also included the help of significant LGBTI NGOs. Despite the centrality of victimization experiences, it was a broader type of survey as it asked about several issues or domains: physical or sexual attacks, discrimination, harassment, physical and mental health assessment, visibility and social life, perception of social prejudices and tolerance, and demographic context questions.

This survey faced a series of methodological and conceptual limitations that may be linked to its target population, that of self-identified LGBTI individuals. The lack or difficulty of access to this target group, understood as a difficult or invisible population (Bell, 2017) makes random sampling difficult or an outright impossibility. The scarcity or legal impossibility of widespread surveys or census with questions on sexual orientation or gender identity (Guyan, 2022), on the other hand, make the quantification and definition of this target population a further difficulty. An additional conceptual limitation is the dependence on the self-identification of participants, as we may encounter a wide range of diverging or even opposing understandings and usages of the survey’s main categories (lesbian, gay, bisexual woman, bisexual man, trans, intersex), as highlighted by Doan (2019).

These limitations were considered and even countered by the survey design, as it followed, according to its technical report, a wide or flexible categorization scheme that aimed to facilitate the participants’ identification among the aforementioned categories (FRA: European Union Agency for Fundamental Rights, 2020a, p. 9). Regarding access, the FRA survey depended on a media campaign and on a frequent recommendation for surveys targeted at LGBTI individual, a combination of stratified and snowball samplings, as participants are asked to refer the survey to other potential respondents (Stuart et al., 2019). As for the population’s quantification, the FRA’s survey sample design entailed a comparative analysis of pre-existing surveys and research in order to approximate to the population self-identified as LGBTI (FRA: European Union Agency for Fundamental Rights, 2020b). Drawing from this analysis, they estimated a series of population percentages according to age groups, with a total of an average of 7,68% of the population to be expected to self-identify as LGBTI. Following this, the survey included the estimation of precise sample sizes per main category and age group for each one of the 30 participant countries, adjusted to facilitate minimally representative samples and to each country’s population (ibid.). Despite the impossibility of a random sampling due to the population’s characteristics, the survey’s technical report states that the comparison between the designed and actually obtained samples, as well as the theoretical and response saturation attained, allows us to accept the results as the most representative possible under the circumstances. This acceptance of the best result or the most representative one under the circumstances of scarcity of detailed data, furthermore, links this study to current debates on the possibilities of data on queer lives and on the “queer(y)ing” of social science research (Browne & Nash, 2016; Compton et al., 2018; Guyan, 2022).

From within this survey’s results and its open data, we have selected a series of key questions regarding issues such as everyday life and openness, physical and sexual attacks, discrimination and harassment, physical and mental health self-evaluation, and perception of the social and political context. The response data from Spanish participants was filtered according to their self-identification as one of the six main categories, but also as one of five possible places of residence: city, suburbs or outskirts of a city, town, country village, or the countryside. Most questions produced sufficient results from Spanish participants, regarding sample size, with exceptions such as intersex respondents (see, Table 1). A significant case was that of questions understood to be secondary, insofar as they depended on a specific response to a previous one, regarding experiences of victimization or lack thereof. These questions frequently produced insufficient filtered responses for most or all categories and places of residence, and as such they were not explored from the detailed, intersection-based perspective, but only for the broad six main categories and the broad five places of residence.

Table 1. Estimated and actually obtained sample sizes for Spain. Source: adapted from FRA 2020b

	Total	L	G	BW	BM	T	I
Estimated optimal sample	8.814	1.616	2.925	1.635	1.130	1.284	224
Actually obtained sample	20.180	2.458	7.339	6.406	1.796	2.067	114

The filtered response data was thus classified according to the intersection of six main categories and five places of residence, and as such the statistical analysis has focused on the independence of categorical variables, based on the chi-squared test. This statistical tests aims to contrast the frequencies of two categorical variables according to a specific criterion or distribution (Agresti, 2012; Geher & Hall, 2014); Hanneman et al., 2012; Wagner & Joseph Gillespie, 2019). A frequent, and in this case useful, application of this test is the contrast of an independence null hypothesis, according to which the two categorical variables should behave independently. Should the null hypothesis be rejected, a dependence relation could be argued for. Accordingly, said null hypothesis has been tested for each survey question according to the main category, to the place of residence, and to the intersection of both. The first analysis aims to test the heterogeneity or homogeneity of experiences among or between LGBTI individuals. The rejection of null hypotheses in this case would signal towards significant differences between the L, G, B, T, and I categories. The second analysis acts as a continuation, seeking to test whether the experiences and life conditions of LGBTI individuals as a group depend rather on their place of residence. Finally, the third analysis targets specific differences within each main category, as in this case the rejection of null hypotheses would allow us to posit the existence of significant differences within a main self-identification category.

The rejection of independence null hypothesis, however, is a limited form of analysis, as this test does not signal the relation’s direction or strength (Geher & Hall, 2014); Kraemer & Blasey, 2016). As such, a significant p-value may allow us to affirm the existence of a relation between two categorical values, without contributing to the study of how said relation is. To do, so we have added an additional layer of analysis with the use of standardized Pearson residuals, as a further tool that does allow us to signal towards a relation’s direction. Therefore, we have estimated the standardized residuals for those cases in which the independence null hypothesis may be rejected, in order to signal towards the specific direction of the dependence relation. This additional level of analysis, however, is still part of a correlation analysis, with no clear causal relation. Besides this methodological limitation, we must consider how particularly in the case of intersex respondents the small sample size may limit the validity of the analysis, whereas the rest of the response data allows us to affirm their sufficient representativeness.

3. Results

The use of the independence chi-squared test allows us to estimate significance p-values for each intersection of categorical variables, including the responses, the main categories, and the places of residence (see, Table 2). The independence null hypothesis has been tested with confidence levels of 90%, 95%, and 99%, and standardized residuals have been estimated only for those cases in which the null hypothesis has been rejected. Regarding said residuals, they have been deemed significant beyond an absolute value threshold of 3.

Table 2. P-values for the Independence chi-squared test. Source: author’s estimation drawing from the EU-LGBTI II survey data for Spain

Domain	Question	Categories	Place	L	G	BW	BM	T	I
Openness	Avoid holding hands	0,0134	0,8165	0,6114	0,3745	<,0,0001	0,1789	0,2539	0,2734
	Avoid certain places	0,0971	0,3995	0,8830	0,0254	0,0082	<,0,0001	0,0084	<,0,0001
	Places where openness is avoided	0,1920	0,9877	0,9968	0,9999	0,0002	0,8652	0,4227	0,5768
	Satisfaction with life	0,3038	0,9999	0,9703	0,7504	0,0958	<,0,0001	<,0,0001	<,0,0001
	Perceived health status	0,0021	0,3950	0,7537	0,2138	0,3569	<,0,0001	<,0,0001	<,0,0001
	Feeling downhearted/depressed in the past 2 weeks	0,0009	0,7591	0,9825	0,9392	0,0381	<,0,0001	0,0002	<,0,0001
	Openness about being LGBTI	<,0,0001	0,1388	0,9947	0,0117	0,0100	0,0571	<,0,0001	<,0,0001
	Involvement in one or more organisations for LGBTI people	0,4957	0,9997	0,9642	0,9999	0,8888	0,2429	0,0963	0,0004
Discrimination	Discrimination in the past 12 months in 8 areas of life	<,0,0001	0,2283	0,9705	0,6128	<,0,0001	0,2968	0,0011	0,9876
	Reporting the last incident of discrimination	0,7498	0,5229	0,3801	0,7573	0,2276	0,9176	0,0908	—
	Reasons for not reporting the last discrimination incident	0,5891	0,2051	—	—	—	—	—	—
Physical and sexual attacks	Experiences of physical or sexual attacks due to being LGBTI in past 5 years	0,0032	0,8597	0,7157	0,3154	0,7752	<,0,0001	0,0004	0,0604
	Experiences of physical or sexual attacks due to being LGBTI in the past 12 months	0,0007	0,7791	0,6595	0,5456	0,9723	0,6158	0,0266	0,3516
	Reporting the last incident of hate-motivated physical or sexual attack to any organisation	0,0147	0,2732	0,0035	<,0,0001	0,2744	0,0278	0,0478	—
	To who did you report the last incident of hate-motivated physical or sexual attack	0,0004	0,2877	—	—	—	—	—	—
	Reasons for not reporting the last incident of hate-motivated physical or sexual attack to the police	0,0001	0,1544	—	—	—	—	—	—
	Effect of last hate-motivated violent incident on health and well-being	<,0,0001	0,8203	—	—	—	—	—	—
Harassment	Experiences of harassment due to being LGBTI in the past 12 months	0,0854	0,6814	0,2781	<,0,0001	0,0003	0,0003	0,0009	<,0,0001
	Reporting the last incident of hate-motivated harassment to any organisation or institution	0,0466	0,1505	—	—	—	—	—	—
	To who did you report the last incident of hate-motivated harassment	0,3640	0,8100	—	—	—	—	—	—
	Reasons for not reporting the last incident of harassment incident to the police	0,0134	0,6821	—	—	—	—	—	—
Social attitudes and governmental response	Perception of increase or decrease of prejudice and intolerance against LGBTI people in the past 5 years	0,1738	0,9290	0,9870	0,5128	0,0104	0,1846	<,0,0001	<,0,0001
	Perception of increase or decrease of violence against LGBTI people in the past 5 years	0,2429	0,6799	0,9859	0,0995	0,0085	<,0,0001	<,0,0001	0,0001
	Main reasons for increase in prejudice, intolerance and/or violence?	0,9442	0,9975	0,5639	0,2872	<,0,0001	0,6322	0,0349	0,5426
	Main reasons for the decrease in prejudice, intolerance and/or violence?	0,8619	0,8388	0,1046	0,9931	<,0,0001	0,3717	0,2104	—
	Effectiveness of government to combat prejudice and intolerance against LGBTI people	0,0022	0,7424	0,7998	0,9190	0,2274	<,0,0001	<,0,0001	<,0,0001
	Government response to safety needs of LGBTI people	0,0052	0,9896	0,8337	0,9888	0,3054	<,0,0001	0,0006	<,0,0001

3.1. Main categories and places of residence

A first level of analysis allow us to test the independence null hypothesis based on the broad set of main categories. Such initial analysis was thus based on the contrast between the response data for each one of the 27 questions and the main categories. The null hypothesis rejection can thus be related to a dependence relation according to which the self-identification within one of the main categories significantly affects the responses. Stated otherwise, this level of analysis allows us to contrast, using the Spanish responses, the comparative homogeneity of victimization experiences and life conditions of LGBTI individuals. As summarised in Table 2, we have been able to reject the null hypothesis for 17 out of 27 questions, and with a 99% confidence for 11 of them: those regarding health self-evaluation, feelings of depression or sadness, visibility as LGBTI, discrimination experiences, physical or sexual attacks in the last year and in the last five years, who where they reported to, why were they not reported, last attack’s effects, and the perception of the government’s efforts regarding the fight against prejudice or violence and for securing LGBTI individuals’ safety needs.

The analysis of the standardized residuals for those questions for which we have been able to reject the null hypothesis signals towards those main categories and questions that deviate more significantly from the independence hypothesis:

• For lesbian women and gay men, we identified significantly more open or out respondents.

• For bisexual women, we identified significantly fewer respondents reporting the last physical or sexual attack.

• For bisexual men, we identified significantly more respondents avoiding holding hands with a same-sex partner and fewer respondents being open or out. On the other hand, significantly more respondents reported the last physical or sexual attack to the media, rather than to police forces or public institutions.

• For trans respondents, we identified significantly fewer respondents with a positive health self-evaluation and more respondents with depression or sadness experiences. We also identified significantly more discrimination experiences, and the last physical or sexual attack was significantly reported to generic or mainstream victim rights NGOs. Regarding the last attack experienced, trans respondents did not report it mostly due to three reasons: fear of a transphobic police reaction, because of them being doing sex work when experiencing the attack, and because of their migratory status. The last attack’s effects were also significantly severe for trans respondents.

• For intersex respondents, we identified significantly more respondents with depression and sadness feelings, discrimination experiences, and physical and sexual attacks in the last year and in the last five years. The last harassment attack was not reported mostly due to fear of reprisals from the same offenders.

The comparison between the main categories signals towards an overwhelmingly better situation for respondents self-identified as lesbian women and gay men, whereas the other main categories have significantly worse experiences and life conditions regarding attacks, discrimination, and health issues. The life conditions and victimization experiences of trans and intersex individuals signal towards significant differences with the other categories, as well as towards key reasons why some of these experiences were not reported by their respondents. We can thus reject the independence null hypothesis between most questions and the main categories, as well as the notion that LGBTI individuals as a whole have homogeneous or even similar experiences and life conditions. However, we cannot reject the null hypotheses for the responses according to the place of residence for LGBTI respondents as a whole. We can thus reject the homogeneity of LGBTI experiences and life conditions even when we consider those LGBTI individuals as a whole that live in a specific category of place of residence.

3.2. Geographical disparity within categories

3.2.1. Lesbian women

Out of 20 questions analysed for the intersection of main category and place of residence, we can only reject the independence null hypothesis for a single question for lesbian respondents’ data. We can only affirm a dependence relation with the place of residence of the question about whether the last physical or sexual attack was reported. We can identify thanks to the standardized residuals how significantly more respondents reported said attacks from the middle of the rural–urban continuum, as opposed to cities and the countryside. However, this analysis is surely limited by the small sample of respondents self-identified as lesbian women who responded from the countryside and country villages. This merits further research and the consideration of the limited scope of the results.

3.2.2. Gay men

As for gay men respondents, we may reject the independence null hypothesis for 5 out of 20 questions: those of the perception of variations in violence against LGBTI people as a whole, avoiding specific places out of fear, openness as a gay man, reporting or not the last physical or sexual attack to any organization, and harassment experiences in the last year prior to the survey. The standardized residuals highlight a significantly better experience for gay men in the rural side of the continuum, regarding visibility, not avoiding specific places, or having fewer or no harassment experiences. Residuals also signalled toward significantly higher reporting rates in suburbia and outskirts of cities.

3.2.3. Bisexual women and men

Whereas we could only reject one and five independence null hypothesis for the lesbian women and gay men categories, respectively, we could reject 12 out of 20 in the case of bisexual women and men. Half of these rejected hypotheses coincide, albeit with differing levels of confidence: avoiding specific places out of fear, general satisfaction with one’s life, sadness or depression feelings, openness or outness, harassment experiences in the last year, and perception of changes in violence against LGBTI individuals. The standardized residuals of most of these questions signal towards a significatively better situation for bisexual women and a significatively worse one for bisexual men in the rural extreme of the continuum. The question about the perception about the change in violence against LGBTI people is the only dissonant one, as rural-based responses from bisexual women were significatively negative, as opposed to those from bisexual men. It is also significant the fact that this increased perception does not match the violence experiences by bisexual women or most categories in the rural extreme. An additional outlier is the relatively better situation for bisexual men in suburbia, regarding harassment experiences.

Considering those questions whose independence null hypotheses were rejected only for bisexual women, we must list the following ones: avoiding holding hands with a same-sex partner, the specific list of places avoided out of fear, discrimination experiences, the perception of variation in prejudice against LGBTI people, and the questions about the reasons for the changes in prejudice and violence. The related residuals follow the previous trend, as the rural side of the spatial continuum is significantly better for bisexual women regarding discrimination and perception about prejudice, as well as the effect of political and institutional support, despite a significant lack of support from community leaders and celebrities. As for which places are avoided out of fear, the rural-based respondents signal markedly towards visibility at work, public transportation, and public spaces.

As for the null hypotheses rejected only for bisexual men, we may list the perception of overall health, physical, or sexual attack experiences in the last five years, reporting or not said experience, and questions about institutional responses to LGBTI individual’s safety needs and to the fight against prejudice. Part of these rejected hypotheses signal towards the aforementioned tendency, as they highlight a significantly negative experience and set of life conditions for bisexual men in rural environments. The questions about the perception of governmental efforts, on the other hand, provide a polarized insight: on the one hand the rural side respondents tilt towards a positive perception of governmental efforts against prejudice and intolerance, but on the other they have a markedly negative view on said efforts to respond to LGBTI individuals’ safety needs.

3.2.4. Trans respondents

We have been able to reject the independence null hypotheses for 17 out of 20 questions for trans respondents: all questions related to physical and mental health, avoiding openness as a trans person in specific places out of fear, openness, participation in LGBTI NGOs, all questions about victimization experiences in different time periods and about reporting them or not, perceptions about changes in violence against LGBTI people and their motives, and the efficacy of governmental actions. The analysis of the standardized residuals allows us to highlight an overall better situation for trans respondents in the rural end of the continuum, regarding openness, avoiding specific places, feelings of depression or sadness, physical or sexual attacks in the last year, evaluation of governmental efforts, and perception about changes in prejudices or violence. Violence was perceived as being significantly rising in cities, on the other hand, and both overall health assessment and harassment experiences were markedly higher in rural settings. We may thus summarize the role of the place of residence for trans participants’ responses as a significantly better experience and set of life conditions in the rural end of the continuum despite the significance of more harassment experiences. The data, however, signals towards a significantly worse situation, violence-wise, in villages as opposed to the countryside and in suburbs as opposed to cities. Consequently, the intermediate places of residence appear to concentrate more cases of physical and/or sexual violence against trans respondents.

3.2.5. Intersex respondents

Last, we may reject the independence null hypotheses for 12 out of 17 questions for intersex respondents. Similar to the analysis for trans respondents’ data, we may reject the independence for all questions about health, avoiding openness out of fear, openness, participation in LGBTI NGOs, attacks experienced in the last five years, harassment experiences, perceptions about the change in prejudice and violence against LGBTI individuals, and assessment of governmental efforts. For questions related to avoiding openness, overall satisfaction with life, health self-assessment, sadness or depression feelings, discrimination, harassment, and the adequacy of governmental efforts, we find ambivalent results with responses significatively negative in suburbia or cities’ outskirts. Both the urban and rural ends of the continuum show relatively better results. However, we must highlight the limitations of the analysis used in this sample for this main category.

4. Discussion

Just as Forstie (2020), we cannot affirm the existence of a unified or homogeneous experiences among LGBTI individuals, even when considering a single place of residence. The analysis of the independence hypotheses between categorical variables, drawing from the Spanish data of the FRA survey, signals towards significantly worse experiences in terms of health, victimization, and openness for trans and intersex respondents, particularly when compared to gay men and lesbian women respondents. The data analysed also allows us to confirm the significantly distinct experiences and life conditions endured by bisexual men and women. Finally, the data analysed also allows us to counter the expectations about big cities as the best places of residence for LGBTI individuals. The findings of the rural edge of rural–urban areas are much more favourable for LGBTI individuals, to the surprise of bisexual men, in that finding the side of protection from violence and distinct advantages for LGBTI individuals or victims.

Our findings support recent critical views on the rural–urban expectations, such as critiques that question “the common assumption that the big city is the inherent space of sexual liberation, as stated in the 1980ʹs and 90ʹs” (Jubany et al., 2021). This critical view on the matter is even more relevant when considering sexile, a migratory narrative or expectation of LGBTI individuals who flee from rural areas to large cities, looking for dimensions historically expected from them: anonymity, job market, and the possibility to be free from kinship and small-town relations (Adiego & Antonio, 2020; Jubany et al., 2021). These expectations have been interrogated by research, such as that by Adiego and Antonio (2020, p. 1355) in Girona, as this Catalonian medium city “exports sexiles, but it is also welcoming of LGBT people from other territories”. This critical perspective on the seemingly univocal or unidirectional migratory narrative from countryside to cities is further put into question by the difficulty to establish a clear-cut division between rural and urban settings (Butterfield, 2018; Jubany et al., 2021; Kuhar & Švab, 2014), as well as by the analytical difficulty to prove this narrative. This dichotomy’s fuzzy boundaries may be explicated were we to understand rurality not as a simple demographic or quantitative issue, but from the perspective of expectations, discourses, and images associated to the rural end of the spatial continuum (Langarita et al., 2021). This line of enquiry follows geographical and sociological points of view that have favoured the relational nature of geographical scales, as political, economic, and sociocultural realities rather than immanent, physical ones (Capdepón & Félix, 2008; Carr & Lempert, 2016; González, 2005).

Drawing from these geographical perspectives, this paper may thus be understood within the current research on the geographical or spatial dimension of LGBTI individuals’ experiences and life conditions. For instance, whereas Glon et al.’s (2021) study on gay and bisexual men’s health in Oklahoma signalled towards a lack of relevant relation between their health and rurality, we may affirm that Spanish data does show a significant and negative relation for bisexual men. Our data do confirm Fisher et al.’s (2014) findings about bisexual men and women in the Nebraska Midlands, whereas we may also confirm the direction signalled by health conditions of gay men and lesbian men in Australia, researched by Lyons et al. (2015). We may thus conclude with the local or particular nature of territories, such as Spain, and this fact makes us argue for the need for both quantitative and qualitative research to explore the specificities of victimization experiences. The need for qualitative, in-depth studies beyond dense and well-known areas, such as Madrid and Barcelona in Spain may be explained by the relevant role of mediating factors, such as a local community’s size and density, but also its mores and traditions, local history, and the distance to metro areas (Forstie, 2020; Kuhar & Švab, 2014). The need for in-depth perspectives also extends to quantitative and mixed-methods research, as our data shows how homogeneous or similar victimization experiences and life conditions among LGBTI individuals, as a whole, cannot be taken for granted. It may thus be more productive to understand victimization experiences and life conditions as epiphenomena that draw from the same system of oppression, but as distinct experiences informed by sociocultural, geographical, and economic inequalities.

Acknowledgements

I must thank Héctor Sánchez Herrero, Diego Parejo Pérez, and Olga Jubany Baucells for their support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author received no direct funding for this research.

Notes on contributors

Ignacio Elpidio Domínguez Ruiz

Ignacio Elpidio Domínguez Ruiz holds a PhD in Human Sciences (Social Anthropology) from the Universidad Autónoma de Madrid, and works as a researcher at the European Social Research Unit at the Universitat de Barcelona’s Social Anthropology Department. He also lectures at both the Universitat de Barcelona and the Universitat Oberta de Catalunya. His research focuses on LGBTI and queer studies, justice, and tourism.