Religiosity, trust, and the acceptance of the SARS-CoV-2 pandemic restrictions: the case of Poland

ABSTRACT

The aim of the current study was to investigate the relationships between personal religiosity, trust, and the acceptance of restrictions which could be imposed on individuals during the COVID-19 pandemic in order to overcome the crisis. The study was carried out in Poland, a country with one of the highest declared levels of religiosity in Europe. Interpersonal and institutional trust were measured. The acceptance of the pandemic restrictions was positively related to personal religiosity and institutional trust (trust in the Church, trust in the Government, and trust in the health authorities). However, there was no association between the acceptance of the restrictions and interpersonal trust. Trust in the Church turned out to mediate the relationship between religiosity and the acceptance of most restrictions associated with the pandemic. The results of the study are discussed in the context of other studies on the relationship between religiosity and health behavior.

KEYWORDS:

• Trust
• religiosity
• Centrality of Religiosity Scale
• SARS-CoV-2-related restrictions
• coronavirus pandemic
• health behavior

Introduction

In December 2019, SARS-CoV-2, a new type of coronavirus, was diagnosed for the first time in Wuhan, in Hubei Province in China. The virus started to spread rapidly, reaching almost every country in the world in the first quarter of 2020. Initially, the only fact the experts knew was that the spread of the virus occurred through direct contact with an infected person. Therefore, the main method of protecting the general public was an order for social distancing. Soon afterwards, governments introduced further restrictions resulting from this order—borders were closed, workplaces switched to remote work, and many limitations were introduced on the use of essential public services. The measures could be effective, provided that the residents accepted the restrictions and were willing to comply fully with them. The process of adaptation to rapid changes was not equally smooth in every country and the problems some encountered were usually attributed to cultural factors.

The current study fits in the line of research investigating the factors that may affect attitudes toward the restrictions imposed, aimed at limiting the spread of the SARS-CoV-2 virus. Combating it required consistent behavior on the part of citizens and respecting the measures that were implemented by state authorities to support the fight against the pandemic. From the point of view of the research findings showing the protective role of religiousness, it would be interesting to see whether religiousness supports attitudes that accept the restrictions. The present study was conducted in Poland, where a fairly uniform Catholic identity prevails. Its specificity can be observed in the relations between the Church and the central Government, which additionally provides potential support for processes implemented to counteract the pandemic from Catholic hierarchs as figures of authority. Consequently, trust in the Catholic Church was included in the study as a variable mediating the relationships explored. The aim of the present study was, therefore, to seek links between the degree of acceptance of various types of restrictions, the level of religiousness of the general public, and the strength of religious figures of authority in the country, measured by the level of trust.

Religiosity in Poland

Religiosity in Poland has a particular nature related to the history of Church–State ties spanning over 1,000 years, but also to contemporary events (Turska-Kawa and Wojtasik 2017, 46–49), such as the almost 50-year period of Communism (with the USSR imposing a totalitarian political system and eliminating religion from the public space) or the transformation of the political system. The Catholic Church played a special role, actively mediating in the development of democratization. Historical events shaped a particular relationship between Church and State in Poland and, consequently, between national and religious identity (Pilch and Turska 2015, 108). Religion acted many times as a defender of the nation in difficult moments. In the period of Communism, the Catholic Church, led by Cardinal Stefan Wyszyński, managed to convince the nation to take its side. Church buildings provided physical space and freedom to act as a democratic opposition. Consequently, national identity was preserved and renewed in the church environment (Modrzejewski and Potulski 2022, 1–2).

The role of the Catholic Church in Poland is also visible in the levels of social trust. For many years, in Polish society, general distrust of other people has been much more common than attitudes based on openness and trust (CBOS 2020). The majority of Poles (79% in 2002, 76% in 2020) invariably believe that their neighbors should not be trusted and that one should be very cautious in relations to others. The level of trust toward others is significantly higher among those who often participate in religious practices. The levels of trust in important institutions, such as political parties, the Government, the Lower and Upper Chamber of the Polish Parliament, the courts, and the media, were 24%, 46%, 33%, 42%, and 32%, respectively. Compared to this, trust in the Catholic Church is at a high level (64%), in spite of the decrease recorded in recent years. (CBOS 2020)

It should be emphasized that, in recent years, cracks in the uniform Catholic structure have been observed in Poland. According to the Pew Research Center (PRC), which examined religious practices in 34 European countries, Poland is ranked eighth. However, it is emphasized that it is the country where the number of Sunday service participants drops the fastest. (PRC 2018) The latest report of the Catholic Information Agency (Bukłaha 2021) shows that, although as many as 91.9% of Poles declare that they belong to the Catholic Church, there is a sharp decline in religiosity among young people. It is difficult to assess to what extent this is a temporary or permanent process. The end of the second decade of the twenty-first century has been a period of major crisis in the Catholic Church in Poland. The main reasons for this include the successive scandals of child abuse by priests revealed in the media, the decline of the image/authority of Cardinal Stefan Dziwisz (the long-standing private secretary of John Paul II, who was accused of covering up cases of pedophilia in the Church), and, finally, the increasing interference of the Catholic Church in politics (Wojtasik 2021, 87).

The present study

On 13 March 2020, the Polish Government acknowledged the SARS-CoV-2 epidemic as a serious threat to the general public and declared a state of epidemic emergency. As the number of infections increased, further restrictions were introduced to combat the pandemic. The announcement of the state of the epidemic in Poland (20 March 2020) caused a great sense of uncertainty in many Poles. A clear threat to health and life appeared, heightened by the news coming from the rest of the world about new victims of the virus, and Polish people’s sense of economic security was shaken. However, during the first wave of the pandemic, Poles exhibited very different levels of behavioral compliance with the preventive measures against COVID-19. In a study of eight countries, Poland (and Russia) had the lowest rate of adherence to preventative measures regarding COVID-19 (Margraf, Brailovskaja, and Schneider 2020, 18).

The main aim of the present study was to investigate the relationship between religiousness and the degree of acceptance of the restrictions imposed on people by state authorities to stop the spread of the COVID-19 pandemic. The acceptance of these restrictions seems of crucial importance for compliance with the rules the state introduced. In many everyday situations, it is difficult or even impossible to control people’s behavior effectively or to force them to be obedient. Thus, convincing people that some restrictions are necessary to provide protection to vulnerable persons, prevent healthcare collapse or stop the pandemic may be the only way to prevent them from breaking the pandemic rules. As the pandemic outbreak exposed the powerlessness of international and national (health) authorities, one might think that other authorities are needed to support pandemic measures.

In the present study, religiousness was treated as a factor that may be associated with the acceptance of the pandemic restrictions. Additionally, we proposed a set of other individual features which could predict the acceptance of these restrictions, namely, interpersonal trust and trust in religious and state institutions. In this study, we attempted to answer the following questions: (1) what is the relationship between declared religiosity and the acceptance of the restrictions associated with the pandemic? (2) What is the relationship between levels of interpersonal trust and acceptance of the pandemic restrictions? (3) What are the relationships between trust in (a) religious, (b) state, and (c) health authorities and acceptance of the pandemic restrictions? (4) Will the association between religiosity and acceptance of the pandemic restrictions (if any) be mediated by trust in the (Catholic) Church?

The studies referred to above showed quite clearly the protective role of religiousness, diagnosed both in the actual health indicators (see e.g. Ironson, Stuetzle, and Fletcher 2006, S62; Kaplan et al. 2006) and in the intentional behaviors of individuals aimed at undertaking non-risky activities (see e.g. Kang and Romo 2011, 771–772; León and Pfeifer 2017, 99). Similar results were obtained in studies conducted in Poland (Domaradzki and Walkowiak 2021, 1514; Przepiórka and Sobol-Kwapińska 2018, 261). However, the results show the need to distinguish the dimensions of religiosity due to the possible different directions of influence (Szałachowski and Tuszyńska-Bogucka 2021).

This made us assume that religiousness may be a factor supporting the efforts of authorities and governments in the fight against the pandemic, because people with a higher level of religiousness may display more accepting attitudes toward the restrictions. We also argue that the pandemic restrictions could be justified by protective purposes (to protect oneself and others against the infection), while the fundamental biblical and theological principle of ‘loving one’s neighbor’ implies concern for the life of each person. Hence Hypothesis 1: Religiosity will be positively related to the acceptance of the pandemic restrictions.

Both interpersonal and institutional trust could be of crucial importance during the COVID-19 pandemic, which posed a serious threat to people’s lives and health as well as the global economy. Research results linking interpersonal trust to acceptance of restrictions are not consistent. For example, higher interpersonal trust was associated with a lower rate of reduction in mobility during the first wave of the coronavirus pandemic (Deopa and Forunato 2021), whereas contradictory results were obtained by Abel Brodeur, Idaliya Grigoryeva, and Lamis Kattan (2021). A person who trusts others believes that people’s future actions “will be beneficial, favorable or at least not detrimental” (Robinson 1996, 576). They think that, in general, people are honest and trustworthy and that they can take responsibility for their behavior and make ethically sound choices (Lewicki, Tomlinson, and Gillespie 2006, 996; Rotter 1980). Therefore, we expect that people with higher levels of interpersonal trust believe that forcing obedience during the pandemic is not needed. This leads to Hypothesis 2: Interpersonal trust will be negatively related to the acceptance of the pandemic restrictions.

Trust in the Church, which may reflect believers’ conviction that the (Catholic) Church is trustworthy and that the ideas it teaches are reliable, is proposed as a mediator of the relationship between religiosity and acceptance of restrictions related to the pandemic. In Poland, religious society has a fairly homogeneous structure. As mentioned, historical events have shaped the strong role of the Catholic Church as a social and political actor. This position has now many faces, but the Church still comments on public and political issues.

Churches in Poland, contrary to other public places, were not closed during the first wave of the pandemic. The number of participants in church services was only temporarily limited. Thus, the activity of priests was not focused on opposing government restrictions, as was the case in many sectors of the economy and social life, but directed to support the faithful. Consequently, trust in the Catholic Church was included as a potential mediator, as strong religious authorities seemed to be capable of influencing believers’ attitudes and behaviors in the pandemic situation. Therefore, Hypothesis 3 states that Trust in (a) religious, (b) state, and (c) health authorities will be positively related to the acceptance of the pandemic restrictions and Hypothesis 4 posits that Trust in the Church will mediate the relationship between religiosity and acceptance of the pandemic restrictions.

Age and sex were treated as control variables. The elderly (who are considered more vulnerable) would endorse the pandemic restrictions more, as these rules were introduced to protect vulnerable people from infection. Also women, compared to men, would endorse the pandemic restrictions more, because caring for others is related more strongly to the female social role.

Method

Participants and procedure

Three hundred and twenty-five adults from the general population of Poland participated in an Internet convenience survey. The participants were recruited via announcements on Internet forums and web sites. They included the research topic, conditions for participation in the research, and a link to the online questionnaire. In order to obtain the most diverse sample, the announcements were placed both in secular and in Catholic forums.

The recruitment period was 16–31 May 2020. Participation was voluntary and anonymous, without any compensation. Informed consent to participate was provided before the survey. The sample was diverse in terms of sex (127 men, 198 women), age (M = 35.4, range 18–71), education (secondary: 94, Bachelor’s degree: 47, Master’s degree: 184), marital status (single: 104, married: 137, unmarried/cohabitation: 58, other: 26), and employment (employed: 208, not employed: 117). During the survey, the participants were asked to answer several questions about their socio-demographic characteristics and to complete the questionnaire. Some additional measures (not related to this study) were included at the end of the survey. The study was approved by the Ethics Committee of the University of Silesia in Katowice, Poland. The data from the present study were uploaded on to the Open Science Framework (https://osf.io/5bukm/).

Measures

Restrictions associated with the pandemic

The questionnaire, designed to measure acceptance of the restrictions which were imposed or could be imposed during the COVID-19 pandemic, was developed as follows. Firstly, an initial pool of test items was generated by a group of academics (sociologists and psychologists), considering (a) the restrictions imposed by the Polish Government, (b) restrictions which were announced but ultimately not imposed, and (c) restrictions whose implementation was potentially possible, due to fears that the state of emergency might be introduced (as a consequence of the problems encountered by the authorities aiming to conduct presidential elections during the pandemic). Among the restrictions were those associated with social distancing and those limiting political, economic, and property rights. Secondly, two independent experts critically reviewed the test items and reduced their quantity by eliminating synonymous phrases or sentences and merging similar sentences. After discussion, consensus was achieved for all items. Thirdly, two other experts checked the restrictions for content independence by comparing the statements in pairs using a grid. The agreement of the judges was a criterion for including the statements in the final pool of 25 restrictions. The final version of the questionnaire consists of the main question (“To what extent do you accept each of the above restrictions which were imposed or could be imposed by authorities in Poland to stop the pandemic?”), the list of restrictions and a 101-point answer scale (0 “definitely don’t accept”, 100 “fully accept”). The list of restrictions used in this study is shown in the Appendix.

Religiosity

Personal religiosity was measured using the Polish version of the “Centrality of Religiosity Scale” (CRS-14) (Huber 2003; Zarzycka 2007). This version consists of 14 items, with the answers on a five-point scale (1 “never”, 5 “very often”), and assesses five core dimensions of religiosity: Intellect (knowledge of religion and thinking about religious issues: “How often do you think about religious issues?”), Ideology (belief in the existence of transcendent reality: “To what extent do you believe that God or something divine exists?”), Public Practice (taking part in religious services: “How important is it for you to be connected to a religious community?”), Private Practice (intensity of private religious practices: “How often do you pray?”), Experience (experiencing transcendence: “How often do you experience situations in which you have the feeling that God or something divine is present?”). The CRS is a widely used measure of religiosity. In the current study, the internal consistency of the scale was high (overall α=0.96, Intellect α=0.90, Ideology α=0.91, Private Practice α=0.96, Public Practice α=0.92, Experience α=0.95). Additionally, the participants were asked the question “Do you feel deeply religious?”, with answers on an 11-point scale (0 “definitely not”, 11 “definitely yes”). The score on this scale (named the DR scale) was the second indicator of religiosity.

Trust

Interpersonal trust was measured using three statements—“Most people are good and kind”, “Most people are honest”, “Most people are trustworthy”—with answers on a seven-point scale (1 “definitely not”, 7 “definitely yes”). These items are often used in the measurement of trust, defined as the expectation that people are generally good and trustworthy (Rotter 1980). The items were averaged (α=0.88). Trust in the Church, trust in the Government, and trust in health authorities were measured with three separate one-item scales: “Determine your level of trust in the Church/Government/health authorities by marking a point on the scale” (0 “lack of trust”, 100 “full trust”).

Results

Preliminary analyses

An exploratory factor analysis (EFA) was performed to understand the factor structure of the initial list of 25 restrictions. The principal axis factoring method was applied (Fabrigar et al. 1999, 277). To ensure reliable results, we followed the recommendations of Dennis Child (2006) and Andy Field (2013, 692) regarding decisions made on the number and composition of the factors (see Appendix). The results of the EFA are shown in detail in the Appendix (Table A1). Finally, four factors, which explain 56% of the total variance, were extracted: (A) restrictions regarding social distancing and isolation (nine items, α=0.91, “Obligation to wear masks in public places”), (B) restrictions associated with limitations of labor rights (three items, α=0.72, “Work order for healthcare professionals to combat epidemics”), (C) restrictions limiting civil rights and increasing safety (seven items, α=0.83, “Controlling and tracking people in quarantine, e.g. via mobile applications”), (D) restrictions limiting civil rights without any direct impact on safety (three items, α=0.57, “Restriction of access to public information guaranteed by law”). The scores on each factor were computed by averaging their respective item scores. The factor scores were used in subsequent analyses.

Descriptive statistics are shown in Table A2 and Table A3 in the Appendix. Correlations between the four groups of pandemic restrictions and other study variables are displayed in Table 1. There were no associations between age, sex, and the acceptance rates of the four groups of restrictions. Thus these variables were ultimately not included in the analyses as control variables.

Table 1. Intercorrelations between four groups of pandemic restrictions and other variables (N = 325).

Study variables	Restrictions regarding social distancing and isolation (A)	Restrictions associated with limitations of labor rights (B)	Restrictions limiting civil rights and increasing safety (C)	Restrictions limiting civil rights without a direct impact on safety (D)
Religiosity (overall)	0.05	0.12*	0.15**	0.24***
Intellect	-0.01	0.20	0.02	0.14*
Ideology	0.05	0.10	0.13*	0.18**
Private practice	0.07	0.11*	0.18**	0.23***
Experience	0.06	0.10	0.16**	0.25***
Public practice	0.03	0.13*	0.15**	0.25***
Deeply religious	0.09	0.11*	0.19**	0.22***
Interpersonal trust	0.07	-0.078	0.00	0.01
Trust in Government	0.23***	0.34***	0.42***	0.36***
Trust in Church	0.07	0.20***	0.23***	0.25***
Trust in health service	0.25***	-0.00	0.28***	0.05
Age	0.02	-0.02	-0.07	0.10

***p < .001, **p < .01, *p < .05

Testing hypotheses

Religiosity (overall scores), two sub-scales of the CRS (Private Practice and Public Practice), and the DR scores were correlated with the acceptance of three groups of restrictions (B, C, D). The Ideology and Experience sub-scales of the CRS were associated with the acceptance of two groups of restrictions (C, D), but Intellect was connected with Group D only. These findings are partially in line with Hypothesis 1 because the association with religiosity was not observed for Group A restrictions. Generally, the relatively strongest correlations were those between religiosity (and their sub-scales) and acceptance of Group D restrictions.

Contrary to expectation, there were no relationships between interpersonal trust and acceptance of the restrictions. Thus, Hypothesis 2 did not receive support. The pattern of associations between trust in institutions and acceptance of the restrictions was more complex. As predicted, trust in the Government correlated moderately with acceptance of all groups of restrictions (Hypothesis 3b). Trust in the Church was associated with acceptance of all but one group of restrictions (B, C, D). In turn, trust in health authorities was connected to compliance with two groups of restrictions (A, C). This means that Hypothesis 2a and 2c were partially supported.

To test Hypothesis 4, the PROCESS macro (Hayes 2012), Model 4, was run using SPSS version 26. The macro calculates bootstrapped bias-corrected 95% percentile confidence intervals for indirect effects (10,000 samples). The intervals that do not contain zero show a significant indirect effect. Following Andrew Hayes and Li Cai’s (2007, 709) recommendation that heteroscedasticity-consistent standard error and covariance matrix estimators should be routinely used, we applied them in our analyses. The acceptance rates of three groups of restrictions (B, C, D), which correlated with both religiosity and trust in the Church, were the outcome variables, the overall religiosity was a predictor variable, and trust in the Church was a mediator. All relations between the predictors and outcome variables were approximately linear. One variable (restrictions limiting civil rights without any direct impact on safety) demonstrated substantial non-normality; thus the Box-Cox transformation (Osborne 2010, 7–9) was applied to normalize the distribution (see Appendix for a description).

The results of the mediation analyses are shown in Table 2. Consistent with the correlational analysis, religiosity was a significant predictor of trust in the Church (B = 20.66, t = 18.05, p < 0.001, CI[18.40, 22.91]). When religiosity alone predicted acceptance of restrictions, the models were significant (Group B, F(1,323) = 4.71, p = 0.03, R²=0.02; Group C, F(1,323) = 7.10, p = 0.008, R²=0.02; Group D, F(1,323) = 13.60, p < 0.001, R²=0.05). When trust in the Church was added to the equations, all models remained significant (Group B, F(2,322) = 5.82, p = 0.003, R²=0.04; Group C, F(2,322) = 10.07, p < 0.001, R²=0.05; Group D, F(2,323) = 7.66, p < 0.001, R²=0.06), but religiosity lost significance (see Table 2). Thus, trust in the Church turned out to be a full mediator of the relationships between religiosity and acceptance of three groups of restrictions. Religiosity influenced the participants’ acceptance of these restrictions indirectly through trust in the Church. However, the indirect effects reached significance for Group B (β_IND= 0.16) and Group C (β_IND= 0.17) restrictions, but for Group D restrictions (β_IND= 0.09), the effect was insignificant.

Table 2. The mediating effect of trust in Church in relationships between religiosity and acceptance of the pandemic restrictions.

	B	t	LLCI	ULCI	β	Boot CI
Total effect of religiosity on Group B restrictions	2.61	2.17*	0.25	4.97	0.12
Direct effect of religiosity on Group B restrictions	-0.75	-0.49	-3.71	2.22	-0.03
Indirect effect of religiosity on Group B restrictions through trust in Church	3.35	-	1.01	5.69	0.16CS	0.05, 0.26
Total effect of religiosity on Group C restrictions	3.48	2.66**	0.91	6.05	0.15
Direct effect of religiosity on Group C restrictions	-0.60	-0.33	-4.14	2.94	-0.03
Indirect effect of religiosity Group C restrictions through trust in Church	4.08	-	1.72	6.50	0.17CS	0.07, 0.28
Total effect of religiosity on Group D restrictions	0.41	3.69***	0.19	0.63	0.22
Direct effect of religiosity on Group D restrictions	0.24	1.62	-0.05	0.53	0.13
Indirect effect of religiosity on Group D restrictions through trust in Church	0.17	-	-0.05	0.39	0.09CS	-0.02, 0.20

Note: A heteroscedasticity-consistent standard error and covariance matrix estimator were used. β—standardized coefficient. A subscript _CS means completely standardized indirect effect. Boot CI—percentile bootstrap confidence intervals (10,000 samples) for completely standardized indirect effect.

***p < .001, **p < .01, *p < .05

After performing mediation analyses, the assumptions regarding residuals were checked. There was neither substantial violation of the assumption of homoscedasticity nor normality of residuals in the analyses for Group B and Group C restrictions. However, only partial normality of residuals was observed for Group D restrictions.

Discussion

In this research, our aim was to investigate the relationships between religiosity and acceptance of restrictions which can be imposed on individuals during the pandemic in order to overcome the crisis. Our study was carried out in Poland which, as a post-Communist country with a difficult history, has a strong tradition of societal resistance and disobedience to the rules introduced by the State. At the same time, Poland is a country with one of the highest declared levels of religiosity in Europe and where the status of the Catholic Church is still high. For these reasons, investigating the role of religiosity in shaping the attitudes and behaviors of Poles in the health crisis seemed especially appropriate.

A wide range of possible restrictions associated with the COVID-19 pandemic was taken into account. The majority of them were imposed in Poland during the first wave of the pandemic and had a great impact on people’s everyday lives. Some of the restrictions examined in our study had not been imposed in Poland yet (e.g. controlling the contents of parcels and letters or monitoring conversations and emails). However, they could have been imposed as the state of an epidemic was introduced in Poland in March 2020 and the introduction of the state of emergency was discussed later. The factor analysis reveals 4 factors, including 22 of the 25 restrictions. The final set of factors contains restrictions regarding social distancing and isolation (A), restrictions associated with limitations of labor rights (B), restrictions limiting civil rights and increasing safety (C), and restrictions limiting civil rights without any direct impact on safety (D). For the sake of simplicity, only relationships at the factor level will be presented.

Our first research question concerns the association between religiosity and acceptance of the pandemic restrictions. It was predicted that the relationships between religiosity and acceptance of the restrictions would be positive (H1). Contrary to expectation, the acceptance rate for the first group of restrictions (A) was not related to religiosity. This may be the case because the restrictions belonging to this group involve social distancing and isolation and were generally endorsed by most participants. However, the acceptance rate of three groups of restrictions (B, C, D) showed significant associations with religiosity: higher declared religiosity was accompanied by higher agreement with the restrictions in these groups. As most of these restrictions are difficult to enforce, as they were set to regulate a wide range of behaviors, an individual act of acceptance (which is commonly related to the belief that these rules and limitations are useful and beneficial for the individual and/or society at large) seems to be the best way to encourage this kind of behavior and make people obedient. Our findings suggest that religiosity can facilitate this process.

There are different reasons why religiosity may be associated with higher levels of agreement with the pandemic restrictions. Some of them, such as the theological principle of loving one’s neighbor or other important elements of the Catholic social teachings, have already been mentioned. Many sociological studies also show the connection between religious axiology and social values (Leege 1993, 10–12). Religion stimulates prosocial activity, determining the aspects of engagement and how it manifests itself (Aghazadeh and Mahmoudoghli 2017, 6–7). Some researchers prove the link between religiousness and the need to help others (Becker and Dhingra 2001). Compliance with the restrictions could be an expression of protecting others in this context.

How to explain why the relatively strongest association in our study was the relationship between religiosity and agreement with restrictions of unclear/doubtful usefulness in the pandemic situation? In Poland, during the lockdown, religious authorities explicitly supported the decisions of the Polish Government associated with the lockdown measures. Therefore, it may be possible to understand this result better by referring to the authority of the Catholic Church as an institution. When rational explanations of the limitation are not available, the role of authority might be even more important.

Our second research question concerns the association between levels of trust and acceptance of the COVID-19 restrictions. We proposed trust in people (H2) and trust in institutions (H3) as factors associated with attitudes toward the pandemic restrictions. The expectation was that the relationship between interpersonal trust and acceptance of the pandemic restrictions would be negative. However, trust in people was not related to the endorsement of the restrictions in our sample. This could be the case because higher interpersonal trust might be connected with attitudes toward the pandemic restrictions in different ways. Participants with higher interpersonal trust might come to the conclusion that restrictions are unnecessary, as most people can be trusted. At the same time, other trustful individuals might endorse the restrictions because they believe that they are beneficial and that most people will obey the rules for the good of others. Thus, contradictory inferences that can be drawn from the belief about people’s trustworthiness may contribute to this result; it is impossible to make inferences about the attitude toward the pandemic restrictions from the level of interpersonal trust. The justification of this result may also be sought in the general social distrust among Poles (Czapinski and Panek 2011). The pandemic situation required quick and clear suggestions of what to do to maintain or even regain the sense of security that was shaken virtually overnight. The stability of attitudes of distrust supports the result obtained, showing that citizens looked for support not in other individuals but in institutions. Generally, the lack of relationships between interpersonal trust and acceptance of pandemic restrictions is congruent with the findings of Frederik Jørgensen, Alexander Bor, and Michael Petersen (2021, 679).

It was also expected that trust in institutions, i.e. trust in the Church (H3a), trust in the Government (H3b), and trust in health authorities (H3c), would be positively related to acceptance of the restrictions. These expectations were confirmed by the data. Trust in the Government correlated moderately with the acceptance of all groups of restrictions. The present study was conducted during the outbreak of the pandemic; most of the restrictions which were included in our survey were indeed imposed on society by the Government at that time. Therefore, it seems justified that, when people are asked to judge whether the decisions of governments are right or wrong, trust in the government (as a decision-maker) may be of primary importance. Positive relationships between trust in government and health behaviors during the pandemic were also obtained in other studies (see e.g. Clark et al. 2020, 79).

Trust in the Church was associated with acceptance of all but one group of restrictions (B, C, D). Only Group A, which concerns mainly social distancing and has the highest acceptance rates, was not associated with trust in the Church. This result mirrors the relationship between religiosity and the acceptance of three groups of restrictions described above. Trust in health authorities was connected to compliance with two groups of restrictions: Group A (limitations associated with social distancing and isolation, which were indeed highly recommended by health practitioners and institutions) and Group C (restrictions limiting civil rights and increasing safety). The results obtained turned out to be consistent with those obtained by researchers in analyses concerning the H1N1 pandemic (Prati, Pietrantoni, and Zani 2011; Quinn et al. 2009) and the COVID-19 pandemic (Sailer et al. 2021). Trust in institutions significantly supports the process of compliance with restrictions in difficult and urgent situations. The results clearly suggest that it is important to build institutional trust, which can prove crucial for rapid social mobilization in an emergency situation.

Our third research question concerns trust in the Church as a mediator of the relationship between religiosity and acceptance of the pandemic restrictions (H4). For all the models examined, the effect of religiosity on acceptance of the restrictions was fully mediated by trust in the Church. No direct effect of religiosity on acceptance of the restrictions occurred when trust in the Church was taken into account. Therefore, believers were more willing to accept the pandemic restrictions than non-believers. It was not so much because of their personal beliefs, but rather because of their trust in the Church and the interpretation of its teachings (or the statements of its representatives) as fostering obedience to the pandemic rules. Thus, trust in the Church seems to have a strategic role in influencing religious persons to accept the pandemic measures. However, it should be noted that the effect sizes for all models tested in this study were significant but low (R² range 0.02–0.05). Also, the standardized indirect effect for the Group D restrictions (0.09) did not reach significance as it was clearly smaller than the coefficients for Group B (0.16) and Group C (0.17).

Limitations, conclusions, and future directions

This study is not without limitations. It was conducted with a convenience sample of participants and had a cross-sectional design. Thus, true causal inference based only on regression analyses was not possible. Recruitment of participants was based on self-selection, which may be a source of bias (Bethlehem 2010). In terms of the socio-demographic features of our sample, the proportion of male and female participants as well as the structure of age and education were different from the whole population, which also limits the generalizability of the results. However, there is a strong theoretical rationale concerning the unquestionable role of authorities (including religious figures of authority) on people’s beliefs, attitudes, and behavior, which can support our conclusions (see e.g. Milgram 1974). Only self-report measures were used in this study; thus, common method bias could occur. To minimize this problem, some techniques recommended by Philip Podskakoff, Scott MacKenzie, and Nathan Podsakoff (2012) were used: each questionnaire was presented separately and had separate instructions. Well-established and valid tools were used in order to reduce the ambiguity of the statements. Personal data were not collected to preserve anonymity. Nevertheless, in future studies, other sources of information such as behavioral observation or experiments are needed. The results of mediation analysis for the Group D restrictions should be interpreted with caution because some of the statistical assumptions for mediation analysis were not fully satisfied. Finally, the data were collected during a specific period (the first wave of the COVID-19 pandemic), which had an impact on participants’ beliefs and attitudes.

The results of the study clearly prove the strong position of the Catholic Church in Poland, understood institutionally rather than measured by level of faith. At a time when authorities are in decline and become incidental in the public space, this can be seen as positive. However, at the same time, our findings show a less positive and potentially harmful aspect of this situation. Polish believers declaring trust in the Church tended to accept more restrictions which were generally less related to the pandemic. This can be used not only in threatening situations but also in moderating expected social behaviors. It is alarming as the faith and trust of citizens could be used instrumentally to achieve extra-religious advantages. It is a situation already encountered repeatedly in Poland in the period of political elections, with Church hierarchs urging people to vote for a specific candidate, perceived as the one supporting the Church’s interests.

The results obtained in the study also raise the question about the specific nature of Polish people’s faith. Prosocial behaviors (understood as compliance with the restrictions) have turned out to be mediated to a greater extent by trust in the institution of the Catholic Church rather than by internal convictions stemming from the Catholic faith or Bible teachings. This points to the cultural embeddedness of religion in Poland and seems to confirm its ludic nature, connected, among other things, with the fusion of the sacred and profane spheres in everyday life. From the point of view of research, it seems extremely interesting to diagnose the motivational potential of both spheres and the relations between them.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Irena Pilch

Irena Pilch is Associate Professor in Psychology at the University of Silesia in Katowice, Poland. Her research interests focus on personality psychology and its applications in the field of clinical, political, economic, social, work, and organizational psychology. She is the author and co-author of numerous scientific articles published in Polish and international journals.

Agnieszka Turska-Kawa

Agnieszka Turska-Kawa is the Director of the Institute of Political Science and Associate Professor in Political Science at the University of Silesia in Katowice, Poland. Her scientific work focuses on electoral behaviors, political leadership, and civic and political participation. She is the coordinator of a number of scientific grants (e.g. Horizon Europe, COST CA15207, National Science Centre Poland) and author or co-author of numerous scientific articles published in Polish and international journals.

Natalia Galica

Natalia Galica is a research and teaching assistant and a PhD candidate in the Institute of Political Science of the University of Silesia in Katowice, Poland. She held a triple scholarship grant of the Ministry of Science and Education for exceptional scientific achievements. Her research interests focus on open science, international migration, political participation, social cohesion, and transnationalism. She has conducted research projects in Poland, Ireland, the UK, and Germany.

Appendix

Factor analysis

An exploratory factor analysis (EFA) was performed to understand the factor structure of the initial list of 25 restrictions. The Kaiser-Meyer-Olkin measure of sampling adequacy was high (KMO = 0.93) and the Bartlett’s test of sphericity yielded a significant result (chi-square = 3.526, df = 300, p < 0.001) which supports the use of factor analysis with our data. The principal axis factoring method (PAF) was applied using Oblimin rotation with Kaiser normalization. PAF is recommended if the assumption of multivariate normality is violated (Fabrigar et al. 1999, 277).

The Kaiser’s eigenvalue-greater-than-one rule and the Scree test were used to determine the number of factors. One item with communalities lower than 0.2 was removed from the analysis (“Salary reduction in some plants that suspended operation”, h²=0.14; see Child 2006). Based on Field’s proposition (2013, 692), factor loadings lower than 0.3 were suppressed during the analysis, which resulted in removing two items (which had all scores suppressed: “Mandatory isolation for people infected by coronavirus”, “Prohibition on raising prices of goods and services”). It is worth noting that 74% of the participants fully agreed with the first of these items (M = 92.4, SD = 18.7). Four factors, which explain 56% of the total variance, were extracted. The scores on each factor were computed by averaging their respective item scores. The four groups of restrictions distinguished were named as follows: (A) restrictions regarding social distancing and isolation (9 items, α=0.91), (B) restrictions associated with labor rights limitations (3 items, α=0.72), (C) restrictions limiting civil rights and increasing safety (7 items, α=0.83), (D) restrictions limiting civil rights without direct impact on safety (3 items, α=0.57).

Table A1. Pattern matrix with suppressing factor loadings less than 0.3 (recommendation by Field 2013, 692).

	Factor
	A	B	D	C
Suspension of classes in kindergartens, schools, and colleges	.860
Closing of shopping malls and furniture stores	.795
Suspension of sports competitions and cultural events	.749
Limiting the number of participants in Mass/service in churches	.706
Closing of services such as hairdressers and hotels	.702
Limitations on the number of customers in stores, pharmacies, post offices	.685
Prohibition of movement (except for professional duties and basic needs)	.553
Obligation to wear masks in public places	.483
Prohibition of visiting patients in hospitals, people in nursing homes, etc.	.403
Mandatory isolation for people infected with coronavirus
High financial penalties for healthcare professionals for failing to perform forced labor		.754
Work duty to combat epidemics for healthcare professionals		.741
Forced work in some workplaces (no employer consent for sick leave, days off, etc.)		.501
Control of content of parcels and letters and content of conversations and emails			.470
Restrictions of access to public information guaranteed by law			.457
Seizing cars or flats in connection with the need to fight a pandemic			.341
Prohibition of organizing assemblies, protests, and employee strikes				-.607
Controlling and tracking people in quarantine, e.g. via mobile applications				-.564
Suspension of associations, political parties, and trade unions				-.535
High financial penalties for non-compliance with bans and orders				-.517
Mandatory quarantine/prohibition to leave apartment for people coming from abroad or after contact with an infected person				-.513
Closure of state borders				-.484
Providing personal data of people in mandatory quarantine, e.g. to police, social security institution, post office				-.453
Prohibition on raising prices of goods and services

Extraction Method: Principal Axis Factoring

Rotation Method: Oblimin with Kaiser Normalization

Table A2. Descriptive statistics for the acceptance of the pandemic restrictions.

	M	SD
Control of content of parcels and letters and content of conversations and emails	6.12	16.56
Restrictions of access to public information guaranteed by law	9.45	20.78
Seizing cars or flats in connection with the need to fight a pandemic	11.48	23.27
High financial penalties for healthcare professionals for failing to perform forced labor	18.97	27.78
Forced work in some workplaces (no employer consent for sick leave, days off)	22.57	27.99
Salary reduction in part of suspended plants	35.22	30.56
Suspension of associations, political parties, and trade unions	35.60	37.72
Work duty to combat epidemics for healthcare professionals	35.66	32.47
Providing personal data of people in mandatory quarantine, e.g. to police, social security institution, post office	39.95	39.51
Prohibition of movement (except for professional duties and basic needs)	49.69	37.38
Controlling and tracking people in quarantine, e.g. via mobile applications	51.26	39.87
Prohibition of organizing assemblies, protests, and employee strikes	52.65	37.19
Closing of services such as hairdressers and hotels	54.95	35.44
High financial penalties for non-compliance with bans and orders	57.75	35.30
Obligation to wear masks in public places	58.89	37.68
Closure of state borders	64.57	35.73
Prohibition on raising prices of goods and services	64.60	37.34
Closing of shopping malls and furniture stores	67.09	33.18
Suspension of classes in kindergartens, schools, and colleges	71.82	31.63
Suspension of sports competitions and cultural events	73.09	32.48
Prohibition of visiting patients in hospitals, people in nursing homes, etc.	73.11	31.40
Limitations on the number of customers in stores, pharmacies, post offices	75.80	29.57
Mandatory quarantine/prohibition to leave the apartment for people coming from abroad or after contact with an infected person	78.08	30.11
Limiting the number of participants in Mass/service in churches	81.42	29.43
Mandatory isolation for people infected with coronavirus	92.45	18.76

Table A3. Descriptive statistics for study variables.

	M	SD	skewness	kurtosis
Religiosity	2.51	1.11	0.39	-0.98
Intellect	2.43	1.11	0.59	-0.51
Ideology	3.17	1.35	-0.17	-1.20
Private practice	2.49	1.38	0.41	-1.24
Experience	2.36	1.25	0.54	-0.82
Public practice	2.13	1.21	0.77	-0.67
Trust in the Church	26.49	32.53	0.99	-0.35
Trust in the Government	20.11	26.86	1.53	1.56
Trust in the health service	63.95	25.74	-0.64	-0.00
Interpersonal trust	3.97	1.48	-0.05	-0.51
A restrictions regarding social distancing and isolation	67.32	25.34	-0.77	-0.13
B restrictions associated with labor rights limitations	25.73	23.66	0.90	0.29
C restrictions limiting civil rights and increasing safety	54.27	25.84	-0.23	-0.75
D restrictions limiting civil rights without direct impact on safety	9.01	14.94	2.14	4.57
Dn restrictions limiting civil rights without direct impact on safety (after normalizing data)	1.72	2.07	0.77	-0.78

Box-Cox transformation

Before mediation analyses were performed, we checked whether the normality assumption was met. Because one of the variables (restrictions limiting civil rights without direct impact on safety) demonstrated substantial non-normality (skewness 2.14, kurtosis 4.57), the Box-Cox transformation (Osborne 2010, 7–9) was applied. This procedure is a family of power transformations that helps to find the normalizing transformation which is optimal for one’s data. First, a constant (1.00) was added to the scores on the variable, so that the smallest score was equal to 1.00. After that, a set of Box-Cox transformation coefficients (lambdas range: from -2 to 1) was calculated using an SPSS syntax (Osborne 2010, 8–9). The transformation which offered the best improvement of normality in this variable (i.e. skewness .77, kurtosis -.78) was used in the subsequent analyses.

The Open Science Framework

The research data are available within the Open Science Framework (see https://osf.io/5bukm/).