Attitudes towards migrants and preferences for asylum and refugee policies before and during russian invasion of ukraine: The case of slovakia

Abstract

Extant literature shows that well-being is one of the key drivers of attitudes towards migrants as well as preferences for asylum and refugee policies. Less in is known, however, about the relationship between well-being and attitudes towards migrants during sudden micro-level events that may elicit the sense of existential threat. To investigate the underpinnings of these relationships, two studies on samples of 600 Slovaks each were conducted before the Russian invasion of Ukraine and during its initial phase. The results show that well-being had a stable positive relationship with attitudes towards migrants across the studies, albeit not with preferences for asylum and refugee policies. During the Russian invasion of Ukraine, the negative feelings elicited by the war predicted preferences for asylum and refugee policies beyond well-being. The results indicate that incorporating psychological factors, such as emotional responses to the looming threat of war, may considerably inform the debate surrounding the support for inclusive asylum and refugee policies.

Introduction

Although the number of migrants as the share of the global population remains relatively stable over time (de Haas, 2014), migration patterns are considerably affected by social changes and a complex network of macro-structural processes such as urbanisation, industrialisation and globalisation (de Haas, 2021). Additionally, persecution, conflicts and violations of human rights increase numbers of forcibly displaced people who flee their homes to seek refuge in wealthier, better protected and safer regions (UNHCR, 2024). Migration, therefore, became one of the key social issues in the modern world and urges researchers to delve deeper into drivers shaping people’s attitudes towards migrants¹ (ATM) and their preferences for asylum and refugee policies. Literature provides evidence that one of the most important factors in this context is subjective well-being (e.g., Welsch et al., 2021). However, several studies show that ATM and preferences for asylum and refugee policies may not be robust in relation to momentous events such as large-scale economic downturns, military conflicts or pandemics (Adam-Troian & Bagci, 2021; Andrighetto et al., 2016; Esses & Hamilton, 2021; Muis & Reeskens, 2022). Abdelaaty and Steele (2022) pointed out that rather than with micro-level concerns (e.g., well-being), preferences for asylum and refugee policies could be associated with macro-level factors (e.g., terrorist attacks or macro-economic situation including GDP and unemployment). Such macro-level situational factors may be perceived as existential threats that elicit strong emotional responses with effects on policy preferences going beyond well-being. Inspired by those findings, the primary aim of the present study was to investigate the role of negative emotions associated with the Russian invasion of Ukraine (a macro-level situational factor) in shaping ATM and preferences for asylum and refugee policies.

We chose to situate our study in Slovakia—one of the European Union (EU) states that neighbours Ukraine but also historically scores high in terms of negative ATM and shows persistent preferences for restrictive asylum and refugee policies (Bozogáňová & Piterová, 2020). Despite remaining relatively unaffected by previous migration crises as a country that rarely is the final destination for migrants, Slovakia has one of the most restrictive migration laws in the EU (Stojarová, 2019) and together with other countries in the region continuously opposes mandatory relocation schemes (quotas) (Csanyi & Kucharčík, 2023). All these make Slovakia an interesting country to investigate potential shifts in ATM and policy preferences in the aftermath of the outbreak of violent conflict. The Russian invasion of Ukraine caused a major migration crisis in Europe. The Office of the United Nations High Commissioner for Refugees (UNHCR, 2022) estimated that by the end of November 2022, more than 7.8 million people had fled Ukraine, over half of whom had found refuge in the EU. From the very beginning, Slovakia, as a neighbouring country, was among the EU states most actively involved in managing the crisis. Despite its image as a conservative country with negative attitudes towards migrants (Grežo et al., 2024), both the Slovak government and the Slovak people provided vast humanitarian aid and support to the refugees. Common discourse and media were filled with compassion, and communications were much more positive than in the case of migrants and refugees originating from the Global South (Dingott Alkopher, 2018; Schmidt, 2021). Moreover, little voices opposed to accepting the refugees were heard during the initial phase of the invasion. Instead, there was an exceptional wave of spontaneous help offered by ordinary people.

This anecdotal evidence seems to indicate that people might have changed their ATM as well as preferences for asylum and refugee policies in the aftermath of the Russian invasion. Therefore, to delve deeper into the underpinnings of this possible attitudinal shift, the present paper reports and compares the results of two studies on the relationships between well-being, ATM, and preferences for asylum and refugee policies that were conducted on Slovak samples before the Russian invasion of Ukraine (Autumn 2021) and during its initial phase (March 2022). In Study 1, we aimed to examine the general levels of and the relationships between well-being, ATM, and preferences for asylum and refugee policies. In Study 2, we hypothesised that the effect of well-being upon ATM as well as preferences for asylum and refugee policies would dampen, while the negative feelings of worry, anxiety and helplessness elicited by the ongoing war in Ukraine would come to the foreground. Unlike most previous studies, we focused on negative feelings elicited not by the migrants or refugees themselves, but by the war which could directly threaten national safety. In line with the common ingroup identity model (Gaertner & Dovidio, 2000), the close war might evoke a sense of shared fate and build up the understanding of the refugees’ experiences. Therefore, we expected the negative feelings of worry, anxiety and helplessness to incline people towards showing more consideration for migrants and refugees, as well as more support for inclusive asylum and refugee policies.

The contribution of this study is twofold. Firstly, the study allows observing whether the relationship between subjective well-being and ATM and policy preferences is robust in relation to macro-level situational factors eliciting negative feelings of worry, anxiety and helplessness. Most importantly, however, our research extends the traditional focus by incorporating emotional responses to the threat that could be associated with attitudinal and behavioural aspects of the debates surrounding asylum and refugee policies.

Literature review and hypotheses

The relationship between well-being and ATM and policy preferences

The relationship between well-being and ATM is well established in literature (Nowakowski, 2021; Pinillos-Franco & Kawachi, 2022a, 2022b; Poutvaara & Steinhardt, 2018). Pursuing an explanation for the relationship, Welsch et al. (2021) suggested that those experiencing more well-being can afford being generous to others and may also draw additional utility from their more open, altruistic attitudes. Thus, people who are satisfied with their lives and feel happy may feel less threatened by the inflow of migrants and refugees and more supportive of inclusive asylum and refugee policies. On the contrary, those who experience less well-being may attempt to make up for it.

Indeed, European Social Survey (ESS) data supported the view that low subjective well-being is consistently associated with more support for populism and populist parties (Nowakowski, 2021; Pinillos-Franco & Kawachi, 2022b). In particular, throughout the EU, people who felt less happy and satisfied with their lives as well as those who considered their health to be poor were more likely to show more hostile ATM and support a populist party in the national elections. Populist parties—particularly right-wing parties—and their supporters tend to hold negative views on migrants and refugees (Chan, 2022; Glinitzer et al., 2021; Golec de Zavala et al., 2017, 2019; Hartman et al., 2021; Van Hauwaert & Van Kessel, 2018; Wirz et al., 2018; Xia, 2021). Consequently, low subjective well-being as a correlate of support for such parties could be further related to the support for those parties’ political programmes and their firm opposition to migration and open asylum and refugee policies.

According to realistic group conflict theory (Sherif, 1966), one of the reasons as to why well-being may be related to less favourable ATM and lower support for inclusive policies is the sense that migrants and refugees pose a threat to the economic status, to safety, or a symbolic threat to cultural and traditional values (Goodman & Schimmelfennig, 2020; Landmann et al., 2019). People who already have low subjective well-being, caused by either health-related and financial issues or generally low life satisfaction, may have limited resilience or ability to accommodate additional external threats. Common narratives showing migrants and refugees—particularly in new EU states—as threats to both individuals and countries’ economy, culture and traditions aggravated the fears. Indeed, literature shows that both symbolic and realistic threats are associated with negative attitudes towards migrants and refugees and with support for dismissive asylum and refugee policies (Landmann et al., 2019; von Hermanni & Neumann, 2019). Based on these findings, we hypothesise that there is a positive relationship between well-being and attitudes towards migrants (H1) as well as preferences for more inclusive asylum and refugee policies (H2).

Situational factors that could influence ATM and policy preferences

Negative shifts in ATM and policy preferences

Less is known about how macro-level situational factors could be associated with the relationship between well-being and ATM or preferences for asylum and refugee policies. Previous studies have shown that large-scale crises eliciting strong negative feelings, such as pandemics or economic cycles, may be associated with less favourable ATM and disturb the support for asylum and refugee policies (Adam-Troian & Bagci, 2021; Andrighetto et al., 2016; Esses & Hamilton, 2021; Muis & Reeskens, 2022). This is in line with realistic group conflict theory indicating that when resources—e.g. jobs, social transfers, or efficiency of the healthcare system—are (perceived to be) scarce, newcomers may be seen to be ripping natives off regarding what rightly belongs to them (Abeywickrama et al., 2018; Alarcon et al., 2016; Meuleman et al., 2020; Mols & Jetten, 2016; Sherif, 1966). Recent studies showed that during economic crises and the COVID-19 pandemic, common narratives pointed to migrants and refugees as additional risk factors increasing uncertainty among natives (Triandafyllidou, 2018). Migrants and refugees were then pointed towards as being responsible for spreading the disease and/or posing an economic threat by intercepting jobs and overburdening welfare and healthcare systems. Throughout the world, declaring war against the invisible enemy—the virus—prompted calls for an even firmer stance on migration as the threat was seen as coming from the outside, brought by migrant invaders (Giacomelli et al., 2020). The war-like narratives provided an excuse to legitimise extraordinary measures taken to stop the disease and secure the Fortress Europe borders by imposing further constraints on mobility that often violated basic human rights (Montagna, 2024). Despite the claims that the measures were driven exclusively by humanitarian and health-related concerns, the resulting climate of fear and distrust towards migrants led to discriminatory and xenophobic distinctions being made between people migrating from various regions (Pacciardi, 2023). Ultimately, politicisation and securitisation of the issue of migration grew stronger. For instance, Hartman et al. (2021) showed that while generally the relationship between right-wing authoritarianism and anti-immigration sentiment is weak, threats of COVID-19 strengthen it. In other words, among people who held more authoritarian views and reported greater COVID-19-related anxiety, anti-migrant attitudes were stronger than among those who did not see the pandemic as a threat.

Similar patterns were observed in the context of economic crises. European Social Survey (wave 5) data show that across European countries, the 2007–2008 economic downturn—expressed as slower GDP growth—was consistently associated with more negative ATM (Billiet et al., 2014). Besides investigating macroeconomic factors, a study conducted by Billiet et al. (2014) showed that the perception of threats was stronger among unemployed and blue-collar workers who could be more easily replaced by fresh migrants than among more qualified employees. What is more, people whose income decreased due to or during the crisis perceived the threat posed by the migrants to be more profound. Meuleman et al. (2020) corroborated the view that the sense of an ingroup’s relative deprivation could be associated with more of an ethnic threat and mediate the relationship between objective economic indicators and an ethnic threat. Indeed, Andrighetto et al. (2016) observed that after the 2012 earthquakes, native Italians who were more affected by the disaster felt more threatened by migrants and, consequently, were less likely to report helping intentions towards outgroups. In other words, people experiencing some hardships or mishaps—be they objective or subjective—may feel that migrants and refugees add yet another threat to their already-precarious existence and may oppose migration more strongly than in times of prosperity and peace.

Positive shifts in ATM and policy preferences

In the present study, we reasoned that the threat and worry driven by the Russian invasion can hardly be attributed to Ukrainian people fleeing their war-torn country. In accordance with the common ingroup identity model (Gaertner & Dovidio, 2000), in such a case the threat posed by the war to both refugees and natives may build up a sense of shared fate (Vollhardt, 2009). This, in turn, may encourage people to show more understanding of the refugees’ situation and behave altruistically towards people previously viewed as being members of an outgroup (Briciu, 2020; Doidge & Sandri, 2019; Klimecki et al., 2020). Following an individual’s dire experiences or vivid perceptions of violence and suffering, psychological changes may transform the trauma into empathy. The sense of one’s own vulnerability may, thus, enhance feelings of compassion and intentions to help others, for instance, by awakening awareness that others also suffer, perceiving the similarity between oneself and those who suffer, and looking at their ordeal from a non-egocentric perspective, whereby establishing the sense of a bond with those more severely affected. Staub and Vollhardt (2008) refer to this type of helping behaviour arising from formative experiences as altruism born of suffering. Indeed, Andrighetto et al. (2016) showed that shared traumatic experiences of the Italian earthquakes fostered minority members’ notion of belonging to a common group with native Italians and inclined them towards showing more altruistic intentions towards the majority. Concurrently, Adam-Troian and Bagci (2021) showed that the perception of the COVID-19 threat was directly and negatively associated with anti-immigration attitudes, at the least, as long as the migrants were not perceived to be a threat themselves. Thus, through shared fears, those experiencing more war-related negative feelings could better understand and empathise with the refugees. Consequently, we expected negative feelings elicited by the war to explain ATM (H3) and preferences for more inclusive asylum and refugee policies (H4) above and beyond well-being.

Furthermore, in line with realistic group conflict theory (Sherif, 1966), migrants and refugees representing various ethnic groups may be perceived differently depending on the salience of threat that they pose. As van Hootegem et al. (2020) observed, some groups of refugees may be perceived to be more deserving of help than others. Meanwhile, others may be viewed with suspicion and there may be a tacit assumption that their reasons for migrating or seeking asylum may not be candid (Triandafyllidou, 2018). In Europe, this particularly applies to young males fleeing North Africa or the Middle East. In the case of the post-invasion migration wave, we speculated that there would be little reason with which to question the true motives of the asylum seekers. Again, in line with the common ingroup identity model, we expected that the level of negative attitudes towards migrants would decrease over time (H5), while the preference for inclusive asylum and refugee policies would increase (H6).

Methods

Participants and procedures

Two separate studies were carried out to address our aims. The first data collection was collected in autumn 2021, approximately half a year before the start of the Russian invasion of Ukraine. The second data collection took place in March 2022, right after the start of the invasion. Each of the two studies was conducted on a gender-balanced sample of 600 Slovaks (300 women and 300 men) and used a self-report design. Table 1 shows sociodemographic characteristics of the two studies. The procedures were the same for both studies. Participation was anonymous and voluntary and the participants could withdraw from the survey at any time. The data were collected by a research agency via an online survey hosted on Qualtrics. After reading and signing an informed consent form, participants answered sociodemographic questions on age, sex and education. Thereafter, in both studies, they completed scales measuring their well-being, attitudes towards migrants, and preferences for asylum and refugee policies. Additionally, participants in the second study also completed scales measuring negative feelings elicited by the Russian invasion. The complete questionnaire in English along with the dataset are available at the Open Science Framework repository (https://osf.io/qsugz/). Both studies were pre-registered (https://osf.io/djr9w and https://osf.io/6hgd7) and approved by the research ethics committee of the Centre of Social and Psychological Sciences of the Slovak Academy of Sciences.

Table 1 Descriptive data for two studies

Measures

Independent variables

Well-being. Well-being was assessed using four distinct measures that were combined into one well-being latent factor score computed by means of a least squares regression method. Firstly, the Circles of life (Kjell et al., 2016) method was used to ask participants the extent to which they feel in harmony and interconnected with the world. Participants had to select one picture from 10 pairs of circles that differed in the degree to which they overlap.

Secondly, a four-item Subjective happiness scale (Lyubomirsky & Lepper, 1999) was used to assess global subjective happiness. The scale asks participants to describe themselves via both absolute ratings and ratings relative to their peers (e.g. “Some people are generally very happy. They enjoy life regardless of what is going on, getting the most out of everything. To what extent does this characterization describe you?”). Participants answered on a seven-point scale.

Thirdly, an eight-item Loneliness scale (Roberts et al., 1993) was used to measure feelings of loneliness. The scale asks participants about their feelings of isolation, being in tune with other people, or lacking companionship (e.g. “I am no longer close to anyone”). Participants answered on a five-point scale (1 = strongly disagree; 5 = strongly agree).

Fourthly, a five-item The satisfaction with life scale (Diener et al., 1985) was used to measure global life satisfaction. The scale narrowly asks about what participants think about their lives without tapping any related constructs such as positive affect or loneliness (e.g. “In most ways, my life is close to ideal”). Participants answered on a five-point scale (1 = strongly disagree; 5 = strongly agree).

Negative feelings elicited by the Russian invasion of Ukraine were assessed using three specific measures that were combined into one latent factor score by means of a least squares regression method. A 10-item Climate change worry scale (Stewart, 2021) was adapted/modified to ask about feelings of worry with regard to the Ukraine war (e.g. “Thoughts about Ukrainian war cause me to have worries about what the future may hold”). Participants answered on a five-point scale (1 = strongly disagree; 5 = strongly agree).

In addition, a short six-item Spielberger state–trait anxiety inventory (Marteau & Bekker, 1992) was adapted to ask about current feelings of anxiety with regard to the Ukraine war (e.g. “I feel tense”). Participants answered on a four-point scale (1 = not at all, 5 = very much).

Finally, a four-item Helplessness scale (Čavojová et al., 2022) was adopted to ask about feelings of helplessness regarding the Russian invasion of Ukraine (e.g. “I feel helpless when thinking about the Ukrainian war”). Participants answered on a five-point scale (1 = strongly disagree; 5 = strongly agree).

Dependent variables

Attitudes towards migrants. A latent factor of attitudes towards migrants, computed by means of a least squares regression method, was created by combining two scales. Firstly, a nine-item Fear-based xenophobia scale (van der Veer et al., 2013) was used to measure how people feel about immigration and migrants (e.g. “Interacting with migrants makes me uneasy”). Participants indicated their attitudes on a five-point Likert scale (1 = strongly disagree; 5 = strongly agree).

Secondly, three questions from the European Social Survey (ESS, 2018) were used to extend the assessment of people’s Immigration threats. The questions map what people think about migrants and immigration in terms of their effects on economics or culture (e.g. “Would you say it is generally bad or good for Slovakia’s economy that people come to live here from other countries?”). Participants answered on a seven-point scale.

Higher scores for attitudes towards migrants latent factor indicated more negative attitudes towards migrants.

Preferences for asylum and refugee policies. The original six-item Preferences for asylum and refugee policies scale was created to assess people’s preferences for what specific asylum and refugee policies a state should follow (e.g. “Each EU country should make its own decisions on asylum applications within its territory”). We were inspired by the work of Jeannet et al. (2021), who identified six core dimensions that conceptualise the asylum and refugee policies of high-income countries. These dimensions relate to: (1) the right of refugees to apply for asylum, (2) the resettlement of already-recognised refugees, (3) the return of asylum seekers whose applications for protection have been unsuccessful, (4) family reunification for recognised refugees, (5) the state’s independence concerning their migration policies, and (5) financial solidarity with countries that host refugees. Participants responded on a five-point Likert scale (1 = strongly disagree; 5 = strongly agree). Higher scores indicated preferences for more restricted asylum and refugee policies.

Control variables

To control for the effects of sociodemographic characteristics, participants were asked questions on their gender, age, education and religiosity. Religiosity was measured using a single question of “How important is religion and believing in God for you personally?” answered on a seven-point scale (1 = not at all; 7 = very important).

Results

Table 1 reports means and standard deviations for sociodemographic variables as well as for all variables that represented the latent factors of well-being, attitudes towards migrants and negative emotions elicited by the Russian invasion. The main focus of our study was the investigation of how well-being relates with attitudes towards migrants and preferences for migration policies. Since we examined these relationships in two independent waves, our data were characterised by a nested structure. To deal with the nested structure of the data, we aimed to apply a multilevel linear regression analysis. The examination of the unconditional (null) model, however, showed that the variation between two studies on mean attitudes towards migrants (Estimate = 0.01; SE = 0.02; Wald Z = 0.89; p = 0.37; 95% CI [0.00; 0.12]) as well as preferences for migration policies (Estimate = 0.07; SE = 0.07; Wald Z = 0.99; p = 0.32; 95% CI [0.01; 0.47]) was not significant. Therefore, there was no evidence of clustering and thus not a need to model the data using multilevel approach. Therefore, we have decided to use multiple linear regression analysis for each study separately.

Study 1—before the russian invasion of ukraine

We performed two separate multiple linear regressions to examine how well-being relates to attitudes towards migrants as well as preferences for asylum and refugee policies before the Russian invasion of Ukraine. In both analyses, biological sex, age, education, and religiosity were entered as covariates to control for their effects.

As shown in Table 2, the model with well-being and covariates as predictors and ATM as the outcome was significant, explaining 6% of the variance of the attitudes. People with higher levels of well-being and education showed more positive attitudes towards migrants. In contrast, age and religiosity was associated with more negative ATM. The relationship between sex and ATM was not significant.

Table 2  Total and direct effects in the multiple linear regression model with attitudes towards migrants as an outcome variable—Study 1

Considerably different results were shown for the model including preferences for asylum and refugee policies as an outcome variable (Table 3). Although the predictors accounted for significant variance in the preferences, and the explained variance (5%) was very similar to the model with ATM, the direct relationships of specific predictors were different. In particular, only age showed a significant relationship with preferences for asylum and refugee policies, with older people having more restrictive preferences. Well-being, sex, education, and religiosity showed no significant relationships.

Table 3  Total and direct effects in the multiple linear regression model with a preference for asylum and refugee policies as an outcome variable—Study 1

Study 2—during the initial phase of the russian invasion of ukraine

We performed two separate hierarchical linear regressions to examine how well-being and negative feelings predict ATM as well as preferences for asylum and refugee policies during the Russian invasion of Ukraine. As in Study 1, biological sex, age, education, and religiosity were entered as covariates to control for their effects.

Table 4 reports the models with ATM as an outcome. The first model (including well-being and covariates) accounted for significant variance in ATM. Like before the invasion, people with higher levels of well-being and education reported more positive ATM. Sex, age and religiosity showed no significant relationships. In the second step, negative feelings elicited by the Russian invasion of Ukraine were entered into the model, along with well-being and covariates. The model accounted for significant variance in ATM, and the explained variance increased significantly (ΔR² = 0.02, ΔF(1, 593) = 12.91, p < 0.001). This model showed that people with higher levels of well-being, education, and negative feelings elicited by the invasion reported more positive ATM. Religiosity showed a non-significant relationship with ATM. Lastly, including negative feelings in the second model caused sex and age to show a significant positive relationship with ATM.

Table 4  Total and direct effects in the multiple linear regression model with attitudes towards migrants as an outcome variable—Study 2

Table 5 reports the results of hierarchical linear regression with preferences for asylum and refugee policies as an outcome. In the first step, well-being was entered along with biological sex, age, education, and religiosity. The model accounted for significant variance in the preferences. People with higher levels of well-being and education reported more inclusive preferences for asylum and refugee policies, while sex, age and religiosity showed non-significant relationships. In the second step, negative feelings were added into the model. The model accounted for significant variance in the preferences, and the explained variance increased significantly (ΔR² = 0.05, ΔF(1, 593) = 34.42, p < 0.001). In the second model, both well-being and negative feelings related with more inclusive preferences, but the negative feelings showed a considerably stronger relationship. As in the first model, education continued to have a significant negative relationship, while sex, age and religiosity showed a non-significant relationship with the preferences.

Table 5  Total and direct effects in the multiple linear regression model with a preference for asylum and refugee policies as an outcome variable—Study 2

Comparing the levels of well-being, attitudes towards migrants, and preferences for asylum and refugee policies in two studies

To examine whether there was any shift in people’s ATM and their preferences for asylum and refugee policies over time, we performed independent t-tests that compared the data on levels of attitudes towards migrants and preferences for asylum and refugee policies before and during the Russian invasion of Ukraine. Since factor scores are centralised and, thus, not comparable, we were not able to use factor scores of ATM for these analyses. Therefore, we used observed variables that were representing these factor scores.

Table 6 shows that people reported significantly lower fear-based xenophobia and immigration threats during the Russian invasion of Ukraine than before the invasion. Moreover, they also expressed significantly stronger preferences for inclusive asylum and refugee policies.

Table 6 The comparisons of the levels of attitudes toward migrants and preferences for asylum and refugee policies in two studies

Discussion

The study investigated the role of negative emotions in shaping ATM and preferences for asylum and refugee policies in the aftermath of the Russian invasion of Ukraine in Slovakia, one of the countries directly neighbouring Ukraine. The research comprised two studies performed before the invasion and soon after it had started. Firstly, the studies explored whether well-being has stable associations with ATM despite a sudden geopolitical crisis. Secondly, the studies examined whether negative emotions elicited by the Russian invasion of Ukraine may be associated both with more positive ATM and with enhanced support for inclusive asylum and refugee policies. The results indicate that in the aftermath of the Russian invasion of Ukraine, Slovaks’ attitudes towards migrants have become more favourable and they were more open to the arrival of refugees. Specifically, people reported lower levels of fear-based xenophobia and sense of threat posed by immigration, and also expressed stronger preferences for inclusive asylum and refugee policies. Although the results may not easily translate into attitudes towards other groups of migrants, they suggest there is a reason to incorporate macro-level situational factors that elicit strong emotional responses in the study of attitudinal and behavioural aspects of debates surrounding asylum and refugee policies. Finally, our results shed light on antecedents of the wave of altruism towards Ukrainian refugees during the first weeks of the Russian invasion of Ukraine.

The roles of well-being and emotions in shaping ATM and policy preferences

In line with previous literature (Nowakowski, 2021; Pinillos-Franco & Kawachi, 2022a, 2022b; Poutvaara & Steinhardt, 2018), the two studies show that well-being has a stable relationship with attitudes towards migrants in general. Both before the invasion and soon after it started, well-being was consistently associated with more positive ATM, thus supporting Hypothesis 1. These results corroborate the view that subjective well-being is robustly associated with ATM despite the macro-level situational factors eliciting negative feelings of worry, anxiety and helplessness. However, our study provides less convincing evidence on the relationship between well-being and preferences for asylum and refugee policies. Inconsistent with Hypothesis 2, well-being was positively associated with preferences for more inclusive asylum and refugee policies in Study 2, but the two were not related in Study 1.

A potential explanation for the findings may be associated with the geographical proximity of the conflict and the fact that it could be perceived as a threatening macro-level factor (Abdelaaty & Steele, 2022). In line with the common ingroup identity model (Gaertner & Dovidio, 2000), the close war might elicit more negative emotions than those elicited by conflicts in distant parts of the world. This, in turn, could build up the understanding of the refugees’ experiences prompting people to support more inclusive asylum and refugee policies. The literature also offers a complementary explanation of the emergence of common identity and the increased acceptance of refugees in the aftermath of conflict by pointing to shared animosities and rivalries that allow to see adversaries' victims as symbolic allies (Jackson & Atkinson, 2019). The authors argue that ideological disputes provide a considerable incentive to accept refugees despite the admission costs. Alas, this explanation does not seem plausible in the Slovak context where strong pro-Russian attitudes prevail and the sense of symbolic or ideological threat posed by the West is greater than that of Russia (Šrol & Čavojová, 2024). This recent research provides persuasive evidence pertaining to strong pro-Russian and anti-West sentiments showing that many Slovaks hold pro-Russian stances, see Russia as the main geopolitical partner and blame NATO, Ukraine and the U.S.—rather than Russia—for provoking the war. Consequently, the sense of ideological disagreements with Russia and associated strategic considerations—the cornerstone of the Jackson and Atkinson (2019) argumentation—are likely to be absent in a large share of the Slovak population. This not only makes Slovakia an even more interesting place to investigate attitudes towards Ukrainians after the Russian invasion but also lends additional credence to our results about the role of well-being and affective cognition in accepting refugees fleeing the close war. Finally, the geographical proximity may have yet another impact. Before the Russian invasion of Ukraine, Slovakia was among the countries with the lowest numbers of asylum applications submitted and even fewer granted (Bozogáňová & Pethö, 2022). Migration crises were mostly observed from a distance and via media coverage. The lack of experiences gained through regular and meaningful contact with migrants may make people more distrustful towards members of outgroups (Bilgic et al., 2019; Briciu, 2020; Green, 2009; Knappert et al., 2021). In the aftermath of the invasion, not only was the media coverage much more sympathetic, people were also more likely to gain first-hand experiences with the refugees (Troszyński & El-Ghamari, 2022).

To delve deeper into the drivers of the shift in ATM and policy preferences, Study 2 also included a measure of negative emotions elicited by the Russian invasion of Ukraine. Our results indicate that negative feelings of fear, anxiety and helplessness are associated both with more positive ATM and with preferences for more inclusive policies. However, the results do not corroborate Hypothesis 3. Although adding negative feelings slightly increased the variance explained by the model, their presence also increased the role of well-being. Thus, although negative emotions were positively related to preferences for more inclusive policies—whereby corroborating Hypothesis 4—they were shown to be closely intertwined with well-being. It seems that in the aftermath of the invasion, immediate emotional responses to such an acute stressor may have close associations with subjective well-being. In general, acute stressors require time to accommodate and, thus, may more strongly resonate in subjective assessment of well-being through emotions than may any chronic factors. Similar mechanisms were observed in Slovakia at the beginning of the COVID-19 pandemic when sudden and acute economic stressors—compared to long-term economic hardship—had an enhanced impact on individuals’ sense of financial well-being (Adamus & Grežo, 2021).

ATM and policy preferences born of suffering

Our study shows that support for more inclusive asylum and refugee policies may be rooted not only in positive and meaningful contact with refugees or migrants (Knappert et al., 2021), but also in shared traumatic experiences. The data indicate that immediate negative emotions elicited by the invasion may have the potential to curb xenophobia and the sense of threat posed by the migrants. In other words, the newly emerged sense of shared threats may also include other ethnically or culturally more diverse groups of people in need (Muis & Reeskens, 2022). The fact that natives may perceive themselves to be equally threatened could, thus, reduce negative outgroup affect (Adam-Troian & Bagci, 2021; Andrighetto et al., 2016). Ultimately, those dire experiences may encourage more helping intentions and altruism. The results, thus, corroborate the hypothesis that formative experiences such as the invasion of a neighbouring country may foster altruism towards members of outgroups (Staub & Vollhardt, 2008; Vollhardt, 2009).

Nevertheless, we need to remember that dual standards of compassion are often applied to different groups of migrants (Pacciardi, 2023). Our results might thus be driven by the methods that we used in both studies. We asked about general ATM without specifying the target group of migrants or refugees. Although general ATM improved over time—whereby providing arguments in favour of Hypothesis 5—this attitudinal shift could be associated with the fact that participants thought about different groups of refugees when answering those questions on general ATM. For instance, extant literature indicates that Middle Eastern refugees—often young, working-age males—are seen to be seeking opportunities to improve their economic circumstances (Verkuyten, 2004; Verkuyten et al., 2018). As such, they could seize less complex jobs and burden the welfare system. Therefore, in line with social identity theory, people with a lower socioeconomic status could see refugees and migrants as being a group competing for the same pool of scarce resources (Billiet et al., 2014; Green, 2009; Meuleman et al., 2020; Van Hootegem et al., 2020). For them, migration may be associated with less job security, as their jobs may be more easily taken by newcomers who generally lack the (recognised) required education or professional experience. The sense of competition could enhance hostility. Non-negligible are also perceived symbolic threats allegedly posed by predominantly Muslim refugees from the Middle East. When we collected data for Study 1, the migration crisis was presented in the media in the context of Middle Eastern or African migrants and refugees. It is likely, thus, that Study 1 partly captured sentiments associated with this ethnic group.

On the contrary, the Ukrainian refugees were mostly women of various ages, often accompanied by children (UNHCR, 2022). As Hudson et al. (2019) noticed, women tend to be seen in a more idealistic way and less as a competitive threat and, thus, may elicit more empathy and helping behaviour. The fact that Ukrainians may be more easily seen to be members of the European Christian community may additionally mitigate the sense of symbolic threat posed by members of outgroups. Ultimately, the ethnic and religious affinity may warm up responses to Ukrainian refugees and elicit more openness and helping behaviour. Therefore, the results of Study 2 could be influenced by the composition of the group of refugees. In conclusion, although we observed a positive shift in ATM, it could be associated with Ukrainians serving as an anchor in answering the questions on general ATM in Study 2. Consequently, it is possible that the effects we observed in our studies may not easily extend to migrants who come from more distant and less known environments, and whose ordeal is less palpable. Building on our results, however, it seems plausible that the experience of close war has a potential to create a shared identity and reduce moral disengagement (Bandura, 1999). This notion of reducing psychological distance and building common identity could be further exploited by policymakers and proponents of a more humanitarian and unified approach to migration. Even though it may be challenging in case of other migration crises, fostering the sense of shared experiences may become an effective strategy for increasing support for more inclusive migration and asylum policies beyond the context of the war in Ukraine. Our findings indicate that this approach could be additionally augmented by policies aimed at improving quality of life, providing more stability and enhancing people’s wellbeing. Ultimately, situational cues and related emotional responses could also manifest through enhanced preferences for more inclusive refugee and asylum policies, corroborating Hypothesis 6. Therefore, although we cannot draw far-reaching and definite conclusions, the findings provide a preliminary empirical argument in favour of the altruism born of suffering hypothesis (Staub & Vollhardt, 2008; Vollhardt, 2009).

Limitations and future recommendations

Despite our best efforts, the study is not free from limitations and thus the results need to be interpreted with some caution. Firstly, our results are based on a self-report design, which may be subject to several biases inherent to this method, including socially desirable responding, acquiescence, and extreme responding (Paulhus & Vazire, 2007). Extreme responding, in particular, might be more common when individuals are asked about issues of significant personal importance (Baumgartner & Steenkamp, 2001). This could indeed be the case for Slovaks, given their general levels of negative attitudes towards migrants and their ongoing rejection of inclusive policies (Csanyi & Kucharčík, 2023).

Secondly, our data were collected only a couple of weeks after the invasion had started; thus, we cannot draw a conclusion with regard to the robustness of the changes in ATM as well as preferences for asylum and refugee policies. As Adam-Troian and Bagci (2021) observed, the necessary condition for enhanced altruism is to perceive the refugees to be non-threatening. The cumulative effects of the war and the looming economic and energy crises may distort the perception of refugees and asylum seekers. If the economic threats become more pronounced, the sense of competition for scarce resources could swiftly set the attitudes and preferences back, whereby thwarting any small gains that we observed in this study. Van Hootegem et al. (2020) pointed out that those who are generally struggling with their economic circumstances are more opposed to migrants and refugees. A lower socioeconomic status and the subjective appraisal of economic threats seem to be consequential for individuals’ ATM and policy preferences. In the long term, economic threats allegedly posed by the refugees and asylum seekers may get to the fore. Consequently, altruistic impulses may soon be replaced by fears that refugees will deprive natives of resources and disturb the process of redistribution. In particular, when the efficiency of institutions responsible for redistribution and social assistance is viewed as being inefficient or deficient, the natives may withdraw their support and the anti-migrant attitudes may be revived (Andrighetto et al., 2016). The longitudinal design could be employed in future studies to track changes in ATM and policy preferences at multiple time intervals reflecting the current geopolitical and economic situation.

Lastly, well-being and negative emotions explained only a relatively small fraction of variations in ATM and preferences. The results suggest that there may be other factors that are more strongly associated with these outcomes. We are aware that particular variables such as political orientation, preferences for an intergroup hierarchy, or institutional trust may have an even-stronger impact on ATM and policy preferences (Hudson et al., 2019; Jennings et al., 2021). Future research could delve deeper into how political or value orientations are intertwined with emotional responses to acute stressors and migration crises. Perhaps conservative political orientation or values may prompt people to overcome immediate emotional responses, thus diluting any effects that emotions could have on ATM as well as support for more inclusive asylum and refugee policies (Grežo et al., 2024).

Conclusion

Our results indicate that the sense of shared fate and understanding have the potential to alleviate hostility towards refugees and migrants. The empathy-driven experience with Ukrainian refugees could encourage more favourable perceptions of migrants and refugees in general. In the current geopolitical and economic situation, this attitudinal shift—albeit frail—is precious and worth fostering. Enhanced sensitivity towards people suffering from war or climate change is a foundation of winning widespread support for policies aimed at bringing relief to those most affected. With the increasing threats—be they real or imagined—to the quality of life and safety, egoistic motivations may take precedence. Citizens of host countries may start to view refugees as competing for scarce economic resources and welfare benefits. Concurrently, those contributing to solidarity or relief funds may come to view the contributions to be disproportionate and unnecessary, thus nullifying potential successes in building the sense that we share our fate with people who differ from us or are far away. Our results indicate the promising path that expanding the sense of community through more empathy towards experiences of other people may become a step in uniting responses to global challenges.