A Longitudinal Assessment of the Impact of Democratic Outcomes on Individuals

A. Intro The present research has three specific goals. It aims at investigating the impact of outcomes of democratic processes on i) perceptions of others’ opinions in the society (i.e., societal norm perceptions; Cialdini et al., 1991) and ii) the psychological well-being of sexual minorities. Finally, it aims at iii) exploring how these changes in perceptions of others’ opinions explain potential changes in psychological well-being. To address these goals, we rely on a longitudinal study in Switzerland, which consists of data that were collected before and after the national voting on extending protection against discrimination to sexual minorities. B. Hypotheses This research aims at providing a better understanding of the impact of outcomes of democratic processes on individuals’ lives. First, because democratic processes are a direct indicator of public support for different issues, this research seeks to investigate how outcomes of democratic processes about sexual minorities impact perceptions of others’ opinions (i.e., perceived societal norms) Research Question 1: How do outcomes of democratic processes about sexual minorities impact perceptions of societal norms toward different sexual minorities’ issues? Initial evidence in the literature suggests that new rights influence perceptions of societal norms (e.g., Eisner et al., 2020; Tankard & Paluck, 2017). Yet, this research also shows that the shift in perceived norms does not necessarily spread to other issues (e.g., a new law on step-child adoption in Switzerland impacted perceived norms toward same-sex parenting but not toward same-sex marriage; Eisner et al., 2020). Based on this evidence, we expect that: The outcome of democratic processes will influence societal norm perceptions. More specifically, we expect that outcomes of democratic processes in favor of sexual minority rights will positively (i.e., consistent with the direction of change) influence norm perceptions by sexual minorities (H1.1). This might, in particular, be the case for norms related to the object of the referendum. Second, this research aims to investigate whether outcomes of democratic processes in favor of sexual minorities impact individuals’ psychological well-being: Research Question 2: How do outcomes of democratic processes about sexual minorities impact the psychological well-being of sexual minorities? Sexual minorities will report experiencing a higher level of psychological well-being following outcomes of democratic processes (H2.1). This effect should be stronger for sexual minorities who highly identify with people who support LGBTIQ+ rights (H2.2) Finally, bringing these goals together, this research aims at understanding the mechanisms behind changes in people’s psychological well-being following outcomes of democratic processes: Research Question 3: To what extent do changes in perceived societal norms in the population explain changes in the psychological well-being of sexual minorities following democratic processes? According to the minority stress framework (Meyer, 2003; see also Hatzenbuehler, 2009), stigmatized minorities’ exposure to unique stressors such as discrimination and rejection lead to detrimental health outcomes and lower psychological well-being. The minority stress model suggests that the effect of distal stressors (e.g., laws or political campaigns) on mental health outcomes is mediated by proximal stressors such as perceived hostile social norms (Hatzenbueler, 2009). Based on the minority stress model, one would expect that sexual minorities experience less detrimental health outcomes following a supportive public ruling, as it should decrease their expectation of rejection by others in the society (e.g., perception of a less intolerant societal norm). Changes in psychological well-being will be partially explained by the shift in norm perceptions (perception of the status quo in the society) for sexual minorities (H3.1), especially those who highly identify with supporters of sexual minorities' rights (H3.2). C. Swiss Study To investigate the impact of outcomes of democratic processes on perceptions of societal norms among sexual minorities, their psychological well-being, and the mechanisms behind these effects, we will rely on Swiss data that were collected among sexual minority members around a public referendum on discrimination. On the 9th of February 2020, 63.1% of Swiss citizens voted in favor of ‘adding protection against discrimination on the basis of sexual orientation in the Swiss law’. We will use the results of this public referendum as a natural experiment – relying on data collected before (December 2019 to early February 2020) and one month after the public referendum (data collection started on the 9th of March 2020). It may be noted that this second data collection coincides with the outbreak of the Covid-19 pandemic in Switzerland – the first semi-lockdown was implemented on the 16th of March 2020, one week after the start of the data collection. For this reason, it is possible that the outbreak of the Covid-19 pandemic impacted participants’ psychological well-being as well (particularly if they participated after the 9th of March 2020). Thus, a question about the perceived impact of Covid-19 is included as a control variable for well-being. Measures We assessed the following variables for the postulated model. We included additional variables for exploratory analyses and research unrelated to the current project. The complete questionnaire can be found online on OSF. Dependent variable Psychological Well-being. Psychological well-being was assessed using measures of positive and negative emotions. Six items adapted from Diener et al. (1985; see also Branscombe et al., 1999) were assessed on a 7-point-Likert scale (1 = very rarely, 7 = very frequently). Participants rated the extent to which they experienced different negative and positive emotions in the last month: (a) sadness, (b) helplessness, (c) shame, (d) enthusiasm, (e) happiness, and (f) pride. Independent Variables Perceived Societal Norm. The perceived societal norm was assessed using three sets of items. First, participants indicated their perception of how freely sexual minorities can disclose their sexual orientation in Switzerland, on a 7-point-Likert scale (1 = not at all, 7 = totally). Then, participants indicated their perception of the general climate toward sexual minorities in Switzerland (1 = very negative, 7 = very positive). Finally, we used two items adapted from the European Social Survey (2006). These items have been assessed on a 7-point-Likert scale (1 = would totally disapprove, 7 = would totally approve). Participants rated their perception of most Swiss people’s opinion toward (a) improving the rights of sexual minorities and (b) same-sex marriage (i.e., “If a same-sex couple wants to get married, most people in Switzerland would…”). Campaign. In the pre-referendum survey, we assessed campaign awareness: Participants indicated whether they had recently seen a political campaign related to the 9th of February voting on adding protection from discrimination on the basis of sexual orientation. The response categories were “1 = Yes, I have seen a campaign from the side which is against adding protection from discrimination on the basis of sexual orientation in the law”, “2 = Yes, I have seen a campaign from the side which is in favor of adding protection from discrimination on the basis of sexual orientation in the law”, “3 = Yes, I have seen campaigns from both sides”, “4 = No, I did not see any campaign”. In the analyses below these measures are coded as two dummy codes for sight of pro-protection campaign (0 = no, 1 = yes) and sight of anti-protection campaign (0 = no, 1 = yes). In the post-referendum survey only, we assessed campaign toxicity with the following item: Participants indicated the extent to which they think that the campaign “against extending the discrimination law” used tactics that could have negative consequences for LGBTIQ+ people and their families. This item has been assessed on a 7-point-Likert scale (1 = not at all, 7 = totally). Identification with Opinion-based Groups. Two items (one item, past-referendum) were assessed on a 7-point-Likert scale (1 = not at all, 7 = totally). In both surveys, participants rated the extent to which they: (a) identify with people who support the rights of sexual minorities. In the pre-referendum survey only, participants rated the extent to which they: (b) feel strong ties with people who support the rights of sexual minorities. Potential Mediators Minority Stress. We assessed several items commonly used in the minority stress literature. These items have been assessed on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). Participants indicated whether they: (a) worry a lot about what others think about them being a sexual minority member, (b) think very carefully before coming out to someone, (c) are very open about their sexual identity (this item will be reversed), (d) ever avoid expressing your sexual orientation for fear of a negative reaction from others (this item was only assessed in the pre-referendum survey). Feeling of Inclusion. Feeling of inclusion was assessed using a set of two items. These items have been assessed on a 7-point scale (1 = not at all, 7 = totally). Participants indicated the extent to which they (a) feel that they are included in the Swiss society, (b) feel that they are part of the Swiss society (this item was only assessed in the pre-referendum survey). Covid-19 Control Variable Impact of Covid-19. In the post-referendum survey only, participants indicated the extent to which their psychological well-being has been affected in the past month by different factors (e.g., political change, climate change) and the outbreak of viruses (i.e., covid-19). The Covid-19 item is included as a control variable, assessed on a 7-point scale (1 = very negatively affected, 4 = not affected, 7 = very positively affected). Sample and Exclusion Criteria We recruited members of sexual (e.g., lesbian, gay, and bisexual individuals) minorities living in Switzerland through online platforms (e.g., social networking sites, snowball sampling, and contacting relevant organizations in Switzerland) and on university campuses. Approximately 400 sexual minorities participated in the pre and in the post referendum survey. We will include participants for our analyses who participated in both waves of data collection and with less than 20% missing data on relevant items. We will further examine the patterns of missingness. Main Analyses Data analysis will follow the standard procedure to estimate moderation and mediation using within-participants designs (Judd et al., 2001) and recent recommendations to test for indirect effects (see Montoya & Hayes, 2017; Yzerbyt et al., 2018). First, we will examine the impact of the public voting on people’s perceptions of the norms (Research Question 1). To do so, we will match participants’ answers to the pre-voting and the post-voting study. Because it is possible that only norms closer to the topic of the referendum (i.e., extending the anti-discrimination law) will be affected by it (e.g., general climate), we will begin by comparing the impact of the voting on perceptions of norms toward different issues (e.g., same-sex marriage, rights in general, general climate) using one sample t-tests in an exploratory manner (we will correct for multiple-testing). If we find similar patterns and high correlation among the items, we will combine the measures of perceived norms. We will run a confirmatory factor analysis to test whether our decision fits the data well. Next, we will investigate the impact of the public voting on people’s psychological well-being (Research Question 2). Using a similar methodological approach, we will test for the difference in reported psychological well-being in the past month before and after the public voting. We will first run a confirmatory factor analysis to test whether negative and positive emotions load, as expected, on two factors as in previous studies. If not, we will construct one factor. Because research mostly either focuses on negative or positive emotions, it is not clear if the effects will be the same or different for positive and negative emotions. Depending on this decision, we will conduct one or two one sample t-test(s). We will further look at whether participants’ level of identification at Time 1 (i.e., pre-referendum survey) moderates the relationship between the public voting and psychological well-being. Finally, we will test for the impact of changes in norm perceptions on changes in psychological well-being (Research Question 3). To do so, we will create difference scores between perceived norms at Time 1 (T1) and Time 2 (T2) and well-being at T1 and T2. We will then regress changes in norms on changes in well-being using these difference scores (see Judd et al., 2001; Yzerbit et al., 2018). In line with past research, we will also include the sum of perceived norms (T1 and T2) in the model to account for the fact that the effect of perceived norms on well-being can differ between T1 and T2 (see Judd et al., 2001; Montoya & Hayes, 2017). We will further add participants’ level of identification at Time 1 in the model to test for the moderation effect of identification. This last set of analyses (i.e., addressing Research Question 3) is contingent on the results of the first two research questions – that is whether there are differences in both norm perceptions and psychological well-being between Time 1 and Time 2. First, we will create scores for the different variables on the basis of factorial analyses. The analyses would examine whether norm changes affect all psychological well-being indicators equally, or whether some are more closely affected, and whether the spillover of the norm change to broader perceptions of a more favorable climate also broadens the impact on psychological well-being. Additional Exploratory Analyses As additional exploratory analyses, we will also test the possible roles of exposure to the pro- and anti-campaigns as a source of perceived norm changes or well-being and as moderators of the associations between changes in norms and psychological well-being. Because data for the post-voting survey were collected right at the beginning of the rise of Covid-19 in Switzerland, participants’ psychological well-being might have been altered. In particular, the first semi-lockdown in Switzerland was implemented one week after the start of the data collection of the post-referendum survey. While many participants participated before the implementation of the semi-lockdown, it is unclear whether their psychological well-being had already been impacted by Covid-19. We assessed the impact of Covid-19 on participants’ psychological well-being using a direct question. For this reason, we will run additional exploratory analyses (e.g., by including this additional construct in the model). Besides investigating the impact of the campaign and Covid-19 on changes in psychological well-being, we will also explore whether characteristics among the participants (i.e., age, gender, sexual orientation) and/or contextual variables (i.e., language-speaking region) affect the findings. Further, to address our last Research Question, we will run an additional exploratory model including items of minority stressors as additional mediators. In addition, if our findings support our hypothesis H3.1, we will additionally explore whether “feeling of inclusion” mediates the impact of perceived norms on psychological well-being (see Badgett, 2011).