Spanish LGBTQ+ Youth and the Role of Online Networks During the First Wave of Covid-19

During the lockdownmeasures put in place at the time of the first wave of the Covid‐19 pandemic in Spain (March through June 2020), LGBTQ+ youth lived through a particularly stressful situation that has so far received little attention. Confined in homes that are often hostile to their sexuality, struggling with the transition to online classes, they reached out to Internet social networks to obtain the support most of them lack in person. This article explores the role of technology for LGBTQ+ youth during a period when the educational environment was not supportive of students’ sexuality and iden‐ tity needs. The research assesses correlations between the use of online social networks and the perceptions of support received from others (using the concepts of social support, thwarted belongingness and burdensomeness, and cohabita‐ tion in their homes). The study involves a sample of 445 Spanish participants aged 13 to 21. A descriptive multivariate analysis of variance and bivariate correlations was performed. We found that social networks were very important for LGBTQ+ youth during the pandemic, helping them to explore their identities, but could also be a source of violence. In this regard, while trans and nonbinary youth’s use of social networks to contact acquaintances show important differences when compared to that of gays, lesbians, and bisexuals, the former group also experiences more violence coming from these networks, finds less social support through them, and feels a stronger sense of burdensomeness in relation to them. Additionally, they were often living with people other than family members during the lockdown. This data suggests the need to offer specific support and online services for LGBTQ+ youth, particularly for trans and nonbinary youth.


Introduction
In Spain, unlike other European countries, childhood and youth were strongly stigmatized during the first wave of Covid-19 (Chmielewska, 2020), which required harsh confinement and social distancing measures between 15 March and 21 June 2020. High schools and universities moved their classes online until the end of the semester. Faculty felt overwhelmed and unsupported in this technological transition, not knowing whether their students had the means to keep up with their classes or under what conditions (Ozamiz-Etxebarria et al., 2021). The requirements of young people were often overshadowed by the urgent need to address the pandemic crisis, and by the common view that sexuality and bullying are not only minor issues but also politically controversial. Additionally, many families faced economic uncertainty and job losses, with unemployment rising to 16% (INE, 2021).
Having access to a device (computer, phone, tablet, etc.) and the Internet was very important for these young people to attend online classes and maintain ties with their peers. Probably due to the widespread use of cell phones (Pérez Díaz et al., 2021;Powell et al., 2010;Qustodio, 2019), young people spent a significant amount of time online during the pandemic and, depending on their use of these resources, were able to access academic resources and maintain communication with one another. However, at the same time, they exposed themselves to potential mental health risks (Hamilton et al., 2020).
Before Covid-19, the literature indicated specific Internet usage by LGBTQ+ youth particularly related to their need to explore their sexuality (González-Ortega et al., 2015). Most notably, LGBTQ+ youth can make sense of their identities using this form of communication (Austin et al., 2020). On social networks, LGBTQ+ youth explore their desires and make friends, practice their social skills, and seek resources to cope with a world that tells them they are too young to know about sexuality (Tortajada et al., 2020;Zheng et al., 2017). They find a "public intimacy" on these networks, having intense online experiences with their devices to which their families and other people around them are oblivious (Jenzen, 2017). LGBTQ+ youth can see how their identity is received online and "come out" to a small group before talking to their families. In particular, during the first wave of the Covid-19 pandemic in Spain, access to these social networks was vital for LGBTQ+ youth to make sense of who they were: to stay in touch with other people, being able to express an identity that they themselves have chosen and which is not always known to all around them (Fish et al., 2020). For those who lack family support, these online connections can alleviate stressful situations linked to their non-normative sexuality and gender identity (Green et al., 2020). However, online social networks are also a space where many LGBTQ+ youth are subjected to harassment (Tortajada et al., 2020).
Overall, it is important to consider how having the social support of their families, schools, friends, and neighbors is essential for LGBTQ+ youth to cope with the stigma of being outside cisgender and heterocentric norms (Frost et al., 2016;Moody & Grant Smith, 2013;Platero, 2014;Warner, 2002), avoiding what is known as "minority stress" (Meyer, 2003). Having this support helps them avoid feelings of loneliness and isolation, but also the sensation of burdensomeness (Green et al., 2020). This is even more true in a situation of a pandemic. Alarming data exist on the lack of support for LGBTQ+ youth and, in particular, trans or nonbinary youth (Buspavanich et al., 2021;Jones et al., 2021;Mirabella et al., 2021).
During the pandemic, support networks helped these young people cope with the associated challenges and imposed restrictions (Mirabella et al., 2021). This occurred in a context in which not only were their social relationships reduced, but they faced potentially hostile situations around their sexuality and greater discomfort linked to their uncertainty regarding the immediate future. They also encountered a standstill in the public administration that affected individuals who wanted to change their names or receive hormonal or retroviral treatments, among others (Green et al., 2020;Platero & López-Sáez, 2020).
Acknowledging the fact that LGBTQ+ youth in Spain often sought support online during the pandemic (Platero & López-Sáez, 2020), we want to understand to what effect this technology was used at a time when the institutional educational environment was not able to provide a comprehensive response for youth development. With an awareness of the contradictory idiosyncrasies of the Internet, which both provides LGBTQ+ youth with an opportunity to explore their sexuality but also exposes them to significant risks, we explore the role of online networks for the vulnerable group of LGBTQ+ youth during the first wave of the pandemic. In particular, we pose questions related to the interrelationship between using online networks, social support, LGBTQ+ youth housemates, and the self-perception of burdensomeness. Finally, we analyze their experiences from the viewpoint of their age, sexual identity, and identification as cis or trans to offer information about a social group that is insufficiently studied in Spain (CIMOP, 2010;Coll-Planas et al., 2009).

Method
This is an exploratory ex post facto prospective and cross-sectional study (Montero & León, 2002), with the independent study variables being age (adolescents and young adults) and gender identity (cis and trans/non-binary).

Participants
A total of 445 people aged 13-21 (M = 1.8, SD = 0.40) living in Spain completed the questionnaire, as part of a larger study that included the participation of 2,833 people of different ages.
With specific regard to Covid-19, 3.6% stated that they had had symptoms related to the illness. Concerning their place of residence during the lockdown, 26.5% were in large cities, 44% were in small cities, and 29.5% were in towns; 29.4% stated that they had to change their residence due to lockdown measures.

Procedure
In May 2020, a group of researchers in gender psychology from the Rey Juan Carlos University, the Autonomous University of Barcelona, and the University of Barcelona designed a study to assess the psychosocial impact of Covid-19 on the LGBTQ+ population. Participants were recruited through advertisements in different social networks and by reaching out to feminist and LGBTQ+ non-governmental organizations, between 4 April and 10 May 2020.
Different scales with the appropriate psychometric properties were used to design the instrument battery based on substantive relevance and consistency for our study. In addition, two experts in gender psychology reviewed the final battery to assess whether each item adequately represented the dimensions of interest. The items were also given to a pilot group consisting of two Black lesbians, two Caucasian gay men, two Caucasian trans persons, and one Caucasian intersex woman, who judged each item in terms of comprehensibility. Four people in this pilot group were under 22 years of age (respectively 15, 17, 19, and 21). Lastly, the items were reviewed by an expert in inclusive language and an expert in psychometric analysis. These revisions improved the clarity, simplicity, and comprehensibility of the questionnaire. Likewise, control items were incorporated to avoid acquiescence bias and loss of veracity, and the non-inclusion of intermediate response options was considered adequate to avoid central tendency bias and social desirability bias when responding to questions related to intimacy.
All the participants received the same instructions and were informed of the voluntary nature, confidentiality, and anonymity of their responses. Before participating, they had to read and accept an informed consent form.

Instruments
Except for the socio-demographic questionnaire, the scales used a response format from 1 (strongly disagree) to 6 (strongly agree). The different instruments used, along with their corresponding consistency indexes according to the authors of each scale, are discussed below.

Socio-Demographic Questionnaire
This questionnaire gathered information about gender identity, sexual orientation, age, education, Covid-19 status, access to treatment, and changes in place of residence. The change in residence variable asked whether a change in residence had occurred and about the participants' housemates before and after the lockdown measures were enacted.

Items on Social Network Usage
Five specific items were included in the questionnaire on the use of social networks, with one question focused on the perception of anti-LGBTQ+ aggression on virtual social networks ("I have received/observed more LGTBQphobic aggression on virtual social networks") and four on how they used networks ("I use online media like social networks/calls/video calls for sexual practices") and whether they made voice or video calls for different purposes (flirting, sexual interactions, talking to friends, talking to family members). These items referred back to two moments: currently (the period of the state of alarm in Spain, between 15 March and 21 June) and before the Covid-19 pandemic. These items were selected because of the usual importance of social networks for young people and adolescents, especially for those with nonnormative sexualities and gender identities (Craig et al., 2015), at a particular time that usually required coping with living in very close quarters with family members and being isolated from peers and other people who help them have a sense of self that is more in line with their self-perceived identities.

Social Support Frequency and Satisfaction Questionnaire
This survey comprises 12 items that measure perceived social support on an emotional, informational, and instrumental plane. The tool has a factorial structure composed of four dimensions: (a) social support received from a partner (Social Support Frequency and Satisfaction Questionnaire [SFSQ]-P), (b) social support received from the family (SSFSQ-F); (c) social support received from friends (SSFSQ-FR), and (d) social support received from the community (SSFSQ-C). Higher scores reflect a greater perception of social support. García-Martín et al. (2016) indicated a high reliability with internally-consistent alpha coefficients of .95, .91, .92, and .92, respectively.

Interpersonal Needs Questionnaire
This questionnaire is composed of nine items, six related to the dimension of self-perception as a burden to others, i.e., burdensomeness (Interpersonal Needs Questionnaire [INQ]-PB) and three related to the sensation of loneliness and a lack of reciprocal support, i.e., thwarted belongingness (INQ-TB). Feelings of burdensomeness and thwarted belongingness are two risk factors strongly linked to suicidal ideation (Van Orden et al., 2010). Higher scores reflect a greater self-perception of burdensomeness. Silva et al. (2018) reported a good overall reliability with an omega coefficient that ranged between .85 and .95.

Analysis
Descriptive statistics were obtained for each item and instrument, along with visual histograms and normality tests. The scores were calculated for each dimension by averaging the items.
Differences in age (adolescents aged 13-17 or young adults aged 18-21) and gender identity (woman, man, or gender non-binary) were analyzed using a multivariate analysis of variance (MANOVA). The different kinds of cohabitation, changes in residence, scales, and the items related to social network usage were considered dependent variables, while age group and gender identity were independent variables. Lastly, correlations between the different variables were estimated using Pearson's coefficient.

Descriptive Statistics
Regardless of whether there was a change of residence, most participants chose to live with family members: 39.3% lived with family members before 15 March, a figure that increased to 89.2% after that date. Adolescents and young people living with friends decreased from 26.1% to 4.3%, those living with their partner(s) increased from 1.3% to 3.1%, and those living alone decreased from 4.7% to 3.1%. During confinement, 28.1% felt little or no support from their partner(s), 27.2% from their family, and 18.2% from friends. During the same period, 31.6% had a feeling of burdensomeness or thwarted belongingness either moderately, frequently, or very frequently. Social networks before the lockdown were used to talk to friends (94.6%), talk to family (79.6%), flirt (51.9%), or engage in cybersex (29.7%). The percentages during confinement were as follows: (a) talking to friends, 98.7%, (b) talking to family, 84.7%, (c) flirting, 40.4%, and (d) cybersex, 30.3%. Some 80% perceived anti-LGBTQ+ aggression before the lockdown, while 71.5% perceived it during confinement. Table 1 shows means and standard deviations, divided by age group (adolescents 13-17 years old × young adults 18-21 years old) and gender identity (cis × trans × nonbinary gender).

Multivariate Analysis of Variance
The results of the MANOVA indicate significance in the interaction (age group × gender identity), F(14,427) = 1.78, p = .05, 2 p = .05. This occurred with the items concerning SSFSQ-FR, F(1,440) = 5.70, p < .05, 2 p = .01. This kind of interaction requires an analysis of the simple effects in order to be interpreted without error (see León & Montero, 2015). The analyses of the simple effects of the age groups showed that there were significant differences between cis-and trans/nonbinary adolescents (F(1,87) = 8.53, p = .005, 2 p = .09), but not among young adults. Thus, cis adolescents scored higher in perceived support from friends than trans and nonbinary adolescents. Analyses of the simple effects of gender identity showed that there were significant differences between trans/nonbinary adolescents and young adults (F(1,138) = 9.50, p < .005, 2 p = .06), with young adults scoring higher on perceived support from friends. No such differences appeared between cis young adults and adolescents.
Differences also appeared in the perception of aggression on social networks, F(1,440) = 7.01, p < .05, 2 p = .01. The analyses of the simple effects of the age groups showed that there were significant differences between cis-and trans/nonbinary young adults (F(1,354) = 7.23, p = .05, 2 p = .02), but not between adolescents. That is, trans/nonbinary young adults scored higher on perceived aggression than cis young adults. Analyses of the simple effects of gender identity showed that there were significant differences between adolescents and trans/nonbinary young adults (F(1,138) = 6.07, p < .05, 2 p = .04), where young adults scored higher. No such differences appeared between cis young adults and adolescents.
Lastly, regarding the use of networks for flirting, the analyses found (F(1,440) = 5.62, p < .05, 2 p = .01). The analyses of the simple effects of the age groups showed that there were significant differences between cis-and trans/nonbinary adolescents (F(1,87) = 4.74, p < .05, 2 p = .05), but not between young adults. Thus, cis adolescents scored higher on the use of social networks and dating apps than trans/nonbinary adolescents. Analyses of the simple effects of gender identity showed that there were significant differences between adolescents and trans/nonbinary young adults (F(1,138) = 6.58, p < .05, 2 p = .04), and trans/nonbinary young adults, with young adults scoring higher. No such group differences appeared between cis young adults and adolescents.
With cohabitation with family members, the analyses of the main effects of age showed the existence of significant differences between adolescents and young adults (F(1,440) = 6.41, p < .05, 2 p = .01), with adolescents of all groups scoring higher in family cohabitation. Similarly, the analyses of the main effects of living with friends showed significant differences between adolescents and young adults (F(1,440) = 5.37, p < .05, 2 p = .01), with young adults of all groups scoring higher on living with friends.
Regarding the use of social networks and applications for cybersex, the analyses of the main effects found differences close to significance between adolescents and young adults (F(1,440) = 2.92, p = .08, 2 p = .01), indicating that young adults in all groups scored higher in the use of networks for cybersex.
In the social support received from family, SSFSQ-F, the analyses of gender identity F(1, 440) = 13.22, p < .001, 2 p = .03) showed the existence of significant differences between cis and trans/nonbinary people, with cis adolescents and young adults scoring higher. These data reveal that cis people perceive that they have more support from their families than trans and nonbinary people.
On the negative self-perception of burdensomeness to others and thwarted belongingness, INQ-PB (F(1,440) = 14.90, p < .001, 2 p = .03) and INQ-TB (F(1,440) = 11.23, p < .05, 2 p = .02), the main effects analyses of gender identity showed the existence of significant differences between cis-and trans/nonbinary people, with adolescents and trans/nonbinary young adults scoring higher. That is, both trans and nonbinary adolescents and young adults have more self-perceptions of burdensomeness and more thwarted belongingness.
Finally, for the use of social networks and other apps to talk to family, the analyses of the main effects of gender identity (F(1,440) = 7.91, p < .05, 2 p = .02) showed the existence of significant differences between cis and trans/nonbinary people, with cis adolescents and young adults scoring higher. Homologously, although bordering on significance (F(1,440) = 2.92, p = .08, 2 p = .01), this relationship was also found in the use of social networks and other applications to talk to friends. Thus, cis people of all ages use networks more to talk to family and friends. Table 2 shows the correlations for the whole sample according to the following variables: gender identity, age, SSFSQ, INQ, housemates, change of residence, use of networks, and perception of aggression. The correlations were calculated using Spearman's coefficient due to the breakdown of the assumptions of continuity or normality in all the pairs of variables.

Discussion and Conclusions
The intersection of age (adolescence and young adults) and gender identity (cis and trans/nonbinary) seems to influence the perception of support received from friends (especially for cis adolescents), as well as the use of networks and applications (with communicating with family and friends and flirting being more frequent among cis people). Age and being trans or nonbinary are key when it comes to perceiving more aggression on social networks during confinement. This perception is linked to the fact that they are, indeed, subjected to greater violence than their cis peers and that such violence is increasingly frequent on social networks, although it is not always reported (FELGTB, 2020a(FELGTB, , 2020b. Likewise, age itself seems to influence the choice to live with some people or others during the lockdown. As might be expected, the lower the age, the greater the likelihood of cohabitation with family members and the less likely cohabitation with friends or partners. This intensive cohabitation with relatives at a time of crisis, like the Covid-19 pandemic, forces adolescents to assess whether to reveal their identity in homes where they do not always receive support and are vulnerable to violence from which they cannot escape (Borraz, 2020;Gorman-Murray et al., 2018;Hawke et al., 2021;Momoitio, 2020). As some preliminary studies among young college students have shown, the people with whom one lives can influence anxiety levels (Iñiguez-Berrozpe et al., 2020). Age also influences the use of social networks. In general, during adolescence, there is a need to explore sexuality, and for those who have non-normative identities this often means facing potential rejection (Mustanski, Newcomb, & Clerkin, 2011). This encourages these teens to turn to social networks in search of support, places where they may find peers and potential partners that they do not have in their "offline life" (DeHaan et al., 2013). Our study found that adolescents are more interested in using networks to flirt, while youth use social networks more frequently for cybersex. These data must be compared with what has been found in other studies amongst the Spanish adolescent population, which is starting to use the Internet for sexual purposes (flirting, searching for information, watching pornography, cybersex, etc.) at an increasingly young age (Ballester-Arnal et al., 2016). Additionally, we also found a high prevalence of online "sexual activities" among Spanish young adults (Gutiérrez-Puertas et al., 2020), as well as the use of social networks for flirting, searching for sexual information, purchasing sexual materials, etc., which is also significant and coincides with the international literature (Shaughnessy et al., 2013;Zheng & Zheng, 2014).
The interactions between online and offline behavior shape the emerging identities, romantic relationships, sexual behaviors, and health of young people (DeHaan et al., 2013). For that reason, more studies are needed on the use of networks in this age group in the intersection with LGBTQ+ identities during times of crisis, such as the Covid-19 pandemic. In this regard, our data suggest the need to create spaces and resources for youth that are informative, accessible, educational, and involve their peers, both online and offline (Fish et al., 2020). This work can be done by the public authorities who work in youth intervention programs. Moreover, these adolescents and young people are already content producers and can thus be an active part of these institutional proposals (Jenzen & Karl, 2014), challenging the adult-centric view of intervention with young people.
On the other hand, gender identity determines the perception of support from friends, since the cis people in the sample perceived that they have more support than trans and nonbinary people. There is a greater self-perception of burdensomeness, having feelings of frustration and thwarted belongingness, which is more common among trans and nonbinary people as the results of other studies have also shown (Pullen Sansfaçon et al., 2019;Reisner et al., 2015). Finally, gender identity determines the type of use of social networks and apps to chat with family members, where cis people in the sample used them more frequently, perhaps because they receive more support from and have more communication with their families than trans and nonbinary people.
The correlations in the sample as a whole affirm some findings in the earlier literature, as well as information appearing in the press (Borraz, 2020;Momoitio, 2020). Gender identity correlated negatively and significantly with using networks to talk to friends. In other words, trans/nonbinary individuals make less use of networks to communicate with friends and-although not significantly but negatively-to flirt and talk with family members. This data raises two questions: What freedom did trans and nonbinary individuals have to communicate and talk during confinement about their identities while under intense family monitoring? Are trans and nonbinary youth finding friendships and bonds with peers that they may not find in their usual places of socialization (school, neighborhood, leisure spaces, etc.) elsewhere? (DeHaan et al., 2013). Contacts made through social networks can compensate for the absence of support "in real life," allow these youths to understand themselves and their processes, find peers with whom to share important experiences in the development of their identity, and potentially forge offline friendships (DeHaan et al., 2013;FELGTB, 2020b;Jenzen, 2017;Mustanski, Newcomb, & Garofalo, 2011;Subrahmanyam & Greenfield, 2008).
These data, in particular on the perception of support received, suggest that there is a specific need for support (both online and offline), not only for adolescents and young adults with non-normative sexualities but especially for those who are trans and nonbinary, who often do not find answers in the existing resources for young people, especially during times of crisis.
Living with a partner correlated positively and significantly with perceiving the existence of aggression on social networks, and negatively with using the networks to talk to family. This suggests that those who live with a partner may not need as much family support. Furthermore, interacting and communicating with a partner could contribute to making this violence on social networks more visible.
Feeling that one has family support positively correlated with using social networks to talk to both family members and friends. This data is consistent with the literature that has observed that, for adolescents, Internet use is a way to stay in touch with the world and explore opportunities (Ofcom, 2014;Procentese et al., 2019;Subrahmanyam & Greenfield, 2008). For youth, the Internet extends how they connect and communicate with other important people in their lives, such as family members (Neustaedter et al., 2013).
It is significant that perceived support from friends correlated positively with all uses of social networks and with perceived aggression. For these adolescents and young adults, social networks may expand social circles, and separate circles may mix, allowing users to explore different uses because of this support (Jenzen & Karl, 2014). The same happens with perceiving partner support, which correlates with all network uses (except flirting) and perceived aggression. If flirting and looking for a partner are two frequent activities on social networks for this age group (Pascoe, 2011), and their online and offline life is interconnected, it is not surprising that partners in monogamous couples are not encouraged to use social networks in this way. In addition, the visibility of a partner or one's very identity as an LGBTQ+ person with friends can be linked to greater exposure to online violence and, consequently, a greater perception of violence.
Self-perceived burdensomeness correlates positively with perceived aggression, but negatively with any use involving contact with others, except cybersex. This could indicate that cybersex is a poor protective factor, unlike other social network uses. Although cybersex can help young people to explore their sexual preferences (Shaughnessy et al., 2013), it also carries some potential risks (Ballester-Arnal et al., 2016), such as exposure to misinformation, reinforcing sexual stereotypes (Longo et al., 2002), and receiving unwanted sexual content (Castro et al., 2015). Furthermore, if it becomes an addiction, it can interfere with daily life (Döring, 2009).
Lastly, the perception of thwarted belongingness correlates negatively with all uses of social networks that involve contact with other people, which is consistent with the feelings of thwarted belongingness and lack of reciprocal care that characterize them. This data is consistent with the existing literature, which indicates that a perceived lack of belonging is related to the perception of loneliness and isolation, which together with a feeling of burdensomeness are risk factors for an active desire to commit suicide (Joiner et al., 2012;Silva et al., 2015;Van Orden et al., 2010).
These data need to be contrasted with more specific studies (that is, based on more representative samples) and comparative studies between countries. However, our data show the need to recognize a population with intersectional characteristics who experiences a particular type of violence and often lacks the necessary support from their environment and the institutions that serve young people.
As some studies have indicated (Espinosa, 2020), access to health protection related to Covid-19 must be better articulated as part of the basic human rights of adolescents and young adults. In Spain, this age group has been discriminated against because of their alleged "potential to spread the coronavirus" while, at the same time, they have not been sufficiently protected and their needs have been ignored. In particular, the lack of protection for LGBTQ+ young adults and adolescents during the pandemic has entailed significant health risks for a population that already has notable health disadvantages, intensifying the gap with their peers.
One lesson learned from the effects of the pandemic is that education and youth-related policies must address existing social inequalities, including sexual and gender diversity. Specifically, policies and youth programs should pay more attention to the use of social networks and apps by LGBTQ+ adolescents and young adults, offering more support services, both inside and outside these networks, particularly considering that young adults and adolescents are already content producers of online materials, in addition to being consumers (Jenzen & Karl, 2014). LGBTQ+ inclusive programs and policies could be extraordinarily helpful in providing much-needed support during these young people identity processes, especially for vulnerable adolescents who are trans and non-binary.
We would also like to acknowledge the support provided by the AFIN research group, the Flax Foundation, and the Rey Juan Carlos University.