Active pro-vaccine and anti-vaccine groups: Their group identities and attitudes toward science.

Vaccine rejection is a problem severely impacting the global society, especially considering the COVID-19 outbreak. The need to understand the psychological mechanisms underlying the active involvement of the pro-vaccine and anti-vaccine movements is therefore very important both from a theoretical and practical perspective. This paper investigates the group identities of people with positive and negative attitudes towards vaccination, and their attitudes toward general science. A targeted sample study of 192 pro-vaccine and 156 anti-vaccine group members showed that the group identity of pro-vaccine individuals is higher than of anti-vaccine individuals. and that both pro-vaccine and anti-vaccine individuals had a positive attitude toward science. Results are discussed in context of the heterogeneity of motivations causing vaccine rejection and the relation between active involvement in online discussion and group identity.

1. Introduction

1.1. Anti-science and vaccine skepticism

The achievements of modern society are built on scientific discoveries and their applications. However, in recent years, a concerning decrease in public confidence in science has been observed, and a strong anti-scientific counterculture has emerged [1]. Based on their own sources of information and selectively ignoring research that contradicts their beliefs, many laypeople believe that they have the background to challenge established scientific facts, and there is no indication that this trend will slow or reverse in the near future [2, 3]. There are several well-researched instances of anti-science. For example, although the theory of evolution is a fundamental idea of modern biology and one of the greatest achievements of Western thought, some people literally interpret the biblical Book of Genesis in asserting that God created the world and mankind within six days, and that it happened no more than 10,000 years ago [4, 5]. Moreover, despite substantial data and climatological analyses, some people deny climate change or reject the assumption that humans contribute to it [6, 7]. Similarly, although vaccination is widely considered one of the most important achievements of medicine and has saved millions of lives, vaccine skepticism is gaining increasing support. On the basis of theory of planned behavior [8-10] one can assume that vaccine skepticism strongly predicts its behavioral counterpart—vaccine rejection. People’s refusal to be vaccinated or have their children vaccinated poses increasing danger not only for them but also for the population’s resistance to infectious diseases. The problem of vaccine rejection is of particular importance in the current situation—the entire world is affected by the COVID-19 pandemic, and a large percentage of citizens is not willing to vaccinate against it (or hesitates to do so), putting the prospect of achieving herd immunity at risk. In Poland, where we conducted our study, only about 50% of the population was vaccinated for COVID19 as of 15% below the European average [11].

Understanding the predictors of both negative and positive attitudes towards vaccination is therefore extremely important both from a purely scientific and applied perspective. In studies conducted before the outbreak of the COVID-19 pandemic, particular attention was paid to the individual differences which might predict attitudes toward vaccines and vaccination. For example, anti-vaccine attitudes are associated with orthodox religiousness, moral purity concerns [12], conspiratorial thinking and individualistic/hierarchical worldviews [13]. We wanted to add to the existing research by taking into account the perspective of group processes. We are observing an important social phenomenon, which is of particular significance during COVID-19—the presence of highly organized anti-vaccination movements, which remain very active even during the pandemic. We assume that the functioning of vaccination opponents, as well as vaccination supporters can be viewed from the perspective of group processes. Contemporary social psychology recognizes that the key processes for group formation are identification with the group and commitment to group functioning. Group identity is seen from a cognitive-motivational perspective and the perspective of intergroup relations [14].

Based on this assumption, we decided to conduct a study of individuals who hold extremely positive or negative attitudes toward vaccination and are heavily involved in various activities (e.g., discussions, conference attendance) that align with their attitudes. The basis for referring to involvement in activities is the contemporary approach to group formation, in which involvement is treated as a fundamental process [15]. In addition to the question of defining oneself in terms of belonging to a group, contemporary analyses place a strongly emphasis on different forms of involvement in group functioning [16].

Our main goal was to investigate the group identity of active supporters and opponents of vaccination, comparing its level in both groups. We also intended to describe these oppositional groups in terms of four modes of group identity [17] (according to Roccas’ theory).

The aim of our study was also to test the attitudes toward science in pro and antivaccine group. Since anti-vaccine individuals often reject the scientific evidence supporting vaccine safety and efficacy and question the voice of medical authorities (e.g., [18-20], there is a fairly widespread perception that they are anti-scientific. However, results of existing research show that that rejecting vaccinations is not necessarily the same as rejecting science in general [18-20].

We wanted to test whether attitudes toward science are significantly different in vaccination supporters and opponents, and what the relationship is between the level of group identification and attitudes toward science in the two groups. Another important object of interest for researchers of group processes is the issue of perceptions of other social groups; in our research we were interested in how these two groups perceive each other’s knowledge about vaccines.

1.2. Group identity

The strength, certainty and clarity of attitudes is strongly influenced by perceived social support for one’s attitude [21, 22]. The group to which people belong is one of the most important sources of such support. A strong sense of identity with the group is related to the intensity of contacts with its members and the exchange of information about important issues. Thus, contacts with members of one’s own group become both a source of different beliefs and a factor increasing the strength and subjective rightness of one’s attitudes [23, 24]. That is why we believe that the level of group identity may be related to attitudes towards vaccination in both vaccine-supporters and vaccine-rejectors. For instance, it has been demonstrated in research conducted on nurses, that their group identification (as a nurse) constitutes an independent predictor of the intention to vaccinate against influenza [25].

Roccas and her colleagues [17] emphasized that group identity is considered within different perspectives: interculturalism (individualism—collectivism), intergroup comparison, identifying with a nation (e.g. patriotism) and within professional organizations. Integrating these perspectives, they suggest that identification with groups involves four correlated but distinguishable modes: 1. Importance—concerns how much individuals see the group as part of who they themselves are. It is a cognitive type of identification; defining oneself by belonging to a group ("we" in opposition to "they"). 2. Commitment—the motivation to act for the good of the group even at the expense of one’s own individual interests. Identification in this sense means a strong positive feeling for one’s own group and a tendency to defend it at all costs. Commitment understood in this way is a typical component of patriotism as well as collectivism. 3. Deference—group organization based on hierarchical relations and subordination of individuals to group norms, symbols and leaders (this is characteristic of totalitarian regimes). 4. Superiority—own group is better than others (typical of nationalism and collective narcissism). Roccas et al. assumed that these four modes of identification are meaningful for large social categories, provided that they are sufficiently entitative for their members to think of them as groups entitativity is understood as “the extent to which a group is perceived as being a coherent unit in which the members of the group are bonded together in some fashion” [26 p. 131].

Roccas et al. created a questionnaire to measure these modes of identification and in their research showed that it retains a four-factor structure for both national and organizational identity. They created a questionnaire to measure these modes of identification and in their research showed that it retains a four-factor structure for both national and organizational identity.

In our project, we incorporated this model of group identity. We were interested in whether the four modes of identification are relevant to pro-vaccine and anti-vaccine groups-which are without a formal hierarchy or clear norms. Whether the identification of these groups maintains a four-factor structure and the differed modes of identification with these groups correlate positively with each other?

Group identity consists of what makes a person subjectively different as a member of his or her own group from the members of other groups (to which he/she does not belong). In principle, any domain (value, symbol, attitude) can become the basis for group identity, provided that it is accepted and considered valid by the members of the group. In accordance with Tajfel & Turner’s classic view [27] we assume that group identity develops not only during face-to-face interactions but also in indirect communication (including electronic), without the necessity for direct contact with other group members in the real world. In this case, “identification is largely symbolic rather than based on interpersonal relationships” [17 p. 281].

Opinion-based groups often form around controversial issues, while oppositely defined opinion groups tend to advocate contrary viewpoints on these issues. The social identity of both opposite groups is then defined by the shared in-group beliefs [28]. From the social-psychological point of view, vaccine rejecters and vaccine supporters can be described as social groups which center around a shared opinion, rather than more tangible characteristics such as gender, nationality or affiliations [29]. We assume that vaccine supporters base their identity on majoritarian common sense, concern for the common good (i.e. health) and respect for scientific authorities. It is difficult to assume a single leading factor in the group identity of vaccine opponents. A variety of factors may influence their involvement: business and political goals, individuals’ characteristics and dispositions (e.g., heightened conspiratorial thinking), parents’ fears for their children, etc. Additionally, anti-vaccinationists are probably aware of their group’s heterogeneity in terms of beliefs; for they differ in their attitudes toward science, their assessment of the existence of coronavirus and the real threat of a pandemic, the content and intensity of conspiracy thinking These factors might lead to a higher level of social identity in the pro-vaccine group than in the anti-vaccine group, even though the minority usually has a stronger group identity than the majority.

1.3. Are vaccine rejecters generally anti-scientific?

One may assume that one of the factors that differentiate vaccine supporters and vaccine opponents is the attitude towards science and scientific knowledge. Scientific knowledge is based on causal relationships between variables and on estimation of the probabilities of the occurrence of particular events. The belief in the effectiveness of vaccines is evidently based on such evidence-based data. Therefore particular, rare cases of post-vaccination negative reactions (like fever or chills) or correlational (not causal!) relationships between vaccination and autism are not enough for those who believe in science to consider vaccination harmful or not necessary. In contrast, common-sense knowledge is based on the observation of single events and on assigning them greater importance than evidence-based data. Therefore, people who do not believe in science may attach greater importance to single cases, correlation relationships, as well as conspiracy theories. Our aim was to examine the role of attitudes toward science in the construction of group identity of vaccination supporters and opponents.

It is often mentioned in the literature that anti-vaccine individuals tend to reject scientific evidence supporting the safety and efficacy of vaccines, and question the legitimacy of relevant research, expertise and medical authority [18-20]. One may think, therefore, that vaccine rejection is strongly associated with general anti-scientific attitudes. The true picture is much more complicated, however. As Rutjens and his colleagues demonstrated [12, 30], science skepticism is not a homogenous phenomenon. They found that vaccine rejection correlates only moderately with other domains of science skepticism (such as climate change denial or GMO rejection), and there is much more to vaccine hesitancy than a general rejection of science.

In addition, leaders of anti-vaccine movements often bolster their authority with scientific titles or medical credentials. One should also take into account that language analysis of anti-vaccination comments published online showed that they contain linguistic markers of analytical thinking, with logically structured statements that mimic valid scientific information [31], which further indicates that vaccine rejection is not associated with general anti-scientific attitudes.

Anti-science, as well as vaccine hesitancy, have long been the subject of substantial research. Particular attention was paid to the individual differences which might predict attitudes toward science. While appreciating the value of such approach, we propose to supplement it with the perspective of social psychology—taking into account group processes, and particularly group identity.

1.4. Current research

The research presented in this paper was aimed at investigating the associations between group identity and attitudes toward science. We wanted to investigate not just pro-vaccine and anti-vaccine individuals, but specifically only those who are involved in discussion about whether vaccines are good or bad—these individuals are the most likely to have a group identity associated with their attitudes toward vaccination, and most likely to have ‘true’ strong attitudes, rather than just declarations. Primarily, we wanted to see whether the pro-vaccine and anti-vaccine groups have similar identity profiles. We also wanted to see whether active anti-vaccine group members actually are anti-scientific, and whether pro-vaccine group members support science. Finally, we wanted to investigate whether attitudes toward science are related with group identity and the level of involvement, which may indicate that attitudes toward science are an important motive for group activity in one or both of these groups. Additionally, it seems interesting to investigate how these two groups perceive each other’s knowledge about vaccines.

As indicated in the previous section, many opponents of vaccination do not necessarily reject science, though others may exhibit science skepticism. Vaccine supporters, however, seem much more homogeneous in their acceptance of science. It can therefore be assumed that a common attitude toward science may, among other variables, serve as a solid basis of group identity for vaccination supporters, but it is less likely for vaccination rejecters. As anti-vaccine attitudes arise from various different motivations and beliefs [12, 13, 30, 32], it can be hypothesized that the social identity of vaccination rejecters will be weaker than that of vaccination supporters.

An additional question worth investigating from the perspective of group processes is how both groups evaluate each other’s knowledge about vaccines and vaccination. Basic rules of social cognition allow assuming that both anti-vaccine and pro-vaccine groups would have a low opinion of the knowledge of "the other" (since these groups have antagonistic beliefs), but the underlying reasoning may differ. Pro-vaccine individuals are expected to believe that anti-vaccine individuals derive information (or rather misinformation) from unscientific and unreliable sources and that they are subject to cognitive error [33, 34]. In turn, vaccine rejecters are expected to be convinced that pro-vaccine individuals base their knowledge on manipulated and biased data, also making it unreliable [35].

In summary, the above considerations allow us to state four main hypotheses. We predicted that supporters of vaccination would have a significantly more positive attitude toward science than members of the anti-vaccine group. The assumption that trust in science is in line with supporting vaccination (a scientifically proven medical preventative measure)—and therefore being a member of the pro-vaccine group, is the basis for the second hypothesis, that in the pro-vaccine group there would be a stronger positive correlation between the levels of group identity and acceptance of science than in the anti-vaccine group. On the other hand, research indicates the motives for belonging to the anti-vaccine group may be more diverse. Thus, we expected that in the pro-vaccine group, the level of social identity would be higher than in the anti-vaccine group. Finally, in-group favoritism and outgroup negativity [36] may serve as support for the assumption that both the pro-vaccine and anti-vaccine groups would provide low evaluations of each other’s knowledge about vaccines and vaccination.

2. Method

The study was conducted using the Ariadna Nationwide Research Panel, a Polish counterpart of mTurk—a company specialized in polling of large samples for the purpose of research. The panel enables random selection of the sample from among 100,000 registered and verified users. Additionally, it has been awarded certificates issued by recognized organizations associated with social research companies (including ESOMAR). For participation in the survey, respondents received credit points that they could exchange for gifts. Study data has not been made publicly available. We sought to survey individuals with strong pro-vaccine or anti-vaccine beliefs who also actively participate in various forms of discussion about vaccines, meaning that they have contact with other similar people and are therefore members of a group rather than lone individuals with particular beliefs. Research was approved by the Ethics Commitee of the Institute of Applied Psychology at the Jagiellonian University. Electronic consent was obtained from all participants.

2.1. Participants

Three hundred and fifty people (203 women and 147 men), aged 18–76 years (M = 41.63, SD = 14.56), were recruited from the general population. This targeted sample was preselected from a general, representative population sample of N = 11579 based on two selection criteria: (1) are pro-vaccine or anti-vaccine and (2) are actively involved in the discussion about vaccination.

The first criterion was based on two questions: (a) What is your opinion about vaccination (possible answers: You should vaccinate; You should not vaccinate; and I am not sure/I don’t care);(b)Would you get vaccinated if there were a vaccine available for a new dangerous disease? (answers ranging from 0 –Definitely not to 10 –Definitely yes). Participants who answered (a)You should vaccinate and (b) at least 7 = yes were considered pro-vaccine. Participants who answered (a) You should not vaccinate and (b) no more than 3 = no were considered anti-vaccine. All others were dropped from the study.

The second criterion was based on the question (c) Do you take active part in the discussion about the need to vaccinate or the consequences of vaccination? (yes or no), which was followed by (d) a list of various types of active discussion contexts, from which the participants could select multiple responses: Internet forums, social media, conferences, pro-vaccine/anti-vaccine societies, discussion with acquaintances and other. Participants who indicated that they took part in discussions in any of these ways were included in the sample, and all other participants were omitted.

In the final sample (N = 350), one hundred ninety-three of the participants indicated that they were pro-vaccine (i.e., they answered You should vaccinate to the question What is your opinion on vaccination), and the remaining 157 were anti-vaccine (i.e., they answered You should not vaccinate; please note that these participants were selected from a larger representative sample to directly investigate pro-vaccine and anti-vaccine groups). When asked an additional question, ‘Would you get vaccinated against a new dangerous disease?’(rated on a scale from 0 to 10, with 0 = strongly disagree and 10 = strongly agree), pro-vaccine individuals strongly agreed (M = 9.26, SD = 1.32) and anti-vaccine individuals disagreed (M = 2.92, SD = 3.08). These groups did not differ in terms of education level (χ2(5, N = 350) = 3.204, p = .669, Cramer’s V = .096), but the pro-vaccine group was more gender-equal (97 women and 96 men vs. 106 women and 51 men in the anti-vaccine group, χ2(1, N = 350) = 10.584, p < .001, Cramer’s V = .174) and had more participants residing in rural areas (pro-vaccine: 69 rural, 65 town, 59 city vs. anti-vaccine: 27 rural, 60 town, 70 city, χ2(2, N = 350) = 15.979, p < .001, Cramer’s V = .226).

2.2. Materials and procedure

The survey was conducted online and consisted of three main parts. Part one collected demographic data (age, sex, education, residence) and responses to the above mentioned selection questions. Part two used Roccas’ group identity questionnaire [17], which consists of 16 questions measuring Importance, Commitment, Superiority and Deference with respect to the participants’ own group (pro-vaccine or anti-vaccine, respectively). Sample items include Belonging to this group is an important part of my identity (Importance); I like to help this group (Commitment); Compared to other groups of this kind, this group is particularly good (Superiority) and It is disloyal to criticize this group (Deference). Please see Roccas’ paper for full details on the questionnaire items. The total Cronbach’s alpha reliability score was α = .92. Cronbach’s alphas for the four subscales were α = .91 for Importance, α = .92 for Commitment, α = .89 for Superiority and α = .86 for Deference. Similarly to the original research by Roccas [17], the modes of identification with groups strongly correlated with each other, ranging from r = .79 for Importance x Superiority to r = .89 for Importance x Commitment. Correlations of the four modes with the total identity measure ranged from r = .92 to r = .94 (all at p < .001). In fact, the observed correlations were higher than in Roccas’ research, indicating that the group identities of active pro-vaccine and anti-vaccine individuals may be equally loaded by all modes.

We then ran a confirmatory factor analysis (AMOS, Maximum Likelihood estimation) to test whether the factorial structure of Roccas’ group identity scale holds in our sample. The model fit was acceptable with GFI = .902, CFI = .966, RMSEA = .07 (with p < .001), and χ2/df = 2.78, therefore corroborating the original factorial structure. We also ran a CFA using a two-factor model proposed by Roccas, wherein Importance and Commitment constituted one factor, and Superiority and Deference were the second factor. The model fit was worse with GFI = .887, CFI = .957, RMSEA = .08 (p < .001) and χ2/df = 3.16. A single-factor model was still worse with GFI = .837, CFI = .932, RMSEA = .10 (p = .001) and χ2/df = 4.35. Comparing the three models using the Bayesian Information Criterion yielded BIC = 639.89 for the 1-factor model, BIC = 518.77 for the two-factor model and BIC = 495.26 for the four-factor model, indicating that the four-factor model proposed by Roccas had the best fit in our sample.

Finally, we asked participants to evaluate two statements (on a scale of 0–10 from Strongly disagree to Strongly agree) measuring their attitude toward science: We can only rationally believe in what is scientifically provable and Science is the most efficient means of attaining the truth [32]. The Cronbach’s alpha reliability score was .82.

The perceived scientific basis of the outgroup’s knowledge of vaccines was measured with two statements rated on a 0–10 scale from Strongly disagree to Strongly agree: [they] base their knowledge about vaccines on scientifically confirmed information and [their] knowledge about vaccines, their content, usage and side effects, is significant. The Cronbach’s alpha reliability score was .88.

3. Results

3.1. Pro-vaccine and anti-vaccine group identities

We compared the group identities of the pro-vaccine and anti-vaccine groups using MANOVA. The pro-vaccine group had a stronger identity than the anti-vaccine group in all respects. All analyses were conducted in SPSS. The results are presented in Table 1 and Fig 1.

Table 1

Group identity levels of the pro-vaccine and anti-vaccine groups.

Identity	Pro-vaccine	Anti-vaccine	F(1,348)	P
Importance	M = 5.40, SD = 1.11	M = 4.64, SD = 1.35	32.94	< .001
Commitment	M = 5.51, SD = 1.09	M = 4.74, SD = 1.27	51.44	< .001
Superiority	M = 5.23, SD = 1.19	M = 4.70, SD = 1.29	25.11	< .001
Deference	M = 5.19, SD = 1.16	M = 4.61, SD = 1.34	29.28	< .001
Total	M = 5.33, SD = 1.06	M = 4.67, SD = 1.22	29.45	< .001

Fig 1

Group identity levels of the pro-vaccine and anti-vaccine groups.

We also found that group identity positively correlated with attitudes toward science in both the pro-vaccine and anti-vaccine groups (!). In the pro-vaccine group, Superiority, Deference and the Total score also correlated with the perceived basis of outgroup knowledge about vaccines, while in the anti-vaccine group, there were no significant correlations in this regard. The results are presented in Table 2. Additionally, attitude toward science did not significantly correlate with perceived basis of outgroup knowledge about vaccines (τ = -.06, p = .246).

Table 2

Group identity correlations with attitudes toward science and perceived outgroup knowledge.

Identity	Attitude toward science		Perception of outgroup knowledge
	Pro-vaccine	Anti-vaccine	Pro-vaccine	Anti-vaccine
Importance	.289**	.160**	.055	-.022
Commitment	.270**	.179**	.048	-.080
Superiority	.191**	.247**	.152**	-.070
Deference	.173**	.145*	.148**	-.063
Total	.240**	.214**	.104*	-.059

Note:

** indicates correlations significant at p < .005,

* indicates correlations significant at p < .05

We then ran a multiple regression analysis to investigate which of these four identity factors (Importance, Commitment, Superiority and Deference) are most strongly associated with attitudes toward science in the pro-vaccine and anti-vaccine groups. For the pro-vaccine group, the four factors explained 9% of the variance (R2 = .09, F(4,188) = 4.662, p = .001) and the only significant factor was Importance (β = .418, p = .032). For the anti-vaccine group, the four factors explained 8.6% of the variance (R2 = .086, F(4,152) = 3.575, p = .008) and the only significant factor was Superiority (β = .510, p = .001). When analyzing both groups within a single multiple regression, it turned out that the four factors explained 9% of the variance (R2 = .090, F(4,345) = 8.480, p < .001), and the only significant factor was still Superiority (β = .216, p = .036).

3.2. Involvement in discussion and group identity

The types of involvement in discussion did not significantly differ between the pro-vaccine and anti-vaccine group members; the pro-vaccine group members participated, on average, in M = 1.65, SD = .87 active forms of discussion, while the anti-vaccine members participated in M = 1.59, SD = .99; t(348) = .578. The analytical results for the individual types of active involvement are presented in Table 3; the only significant differences concerned pro-/anti-vaccine societies and discussions with acquaintances. The two groups did, however, differ in their declared level of involvement in the discussion—the pro-vaccine group reported higher involvement (M = 8.17, SD = 1.36 vs. M = 6.91, SD = 1.91; t(273,741) = 6.94, p < .001).

Table 3

Types of active involvement of pro-vaccine and anti-vaccine group members.

Type of involvement	Pro-vaccine N	Anti-vaccine N	X2	P
Internet forum	84	58	1.56	.21
Social media	58	37	1.84	.18
Societies	15	36	15.98	< .001
Conferences	15	18	1.38	.24
Discussing with acquaintances	146	101	5.34	.021
Other	2	4	n/a	n/a

The declared level of involvement significantly positively correlated with all modes of group identity. We used Kendall’s tau-b correlation coefficients due to violations of normality in the data; however, Pearson correlations indicated the same effects. The correlations are presented in Table 4.

Table 4

Correlations between declared level of involvement and group identity.

Involvement correlation with	Group identity τ
	Total sample	Pro-vaccine	Anti-vaccine
Importance	.396**	.440**	.280**
Commitment	.466**	.452**	.399**
Superiority	.398**	.446**	.307**
Deference	.396**	.415**	.328**
Total	.433**	.454**	.350**

Note:

** indicates correlations significant at p < .005,

* indicates correlations significant at p < .02.

3.3. Attitudes toward science in the pro-vaccine and anti-vaccine groups

A comparison of attitudes toward science between the groups indicated that members of the pro-vaccine group were more pro-scientific than members of the anti-vaccine group (M = 8.02, SD = 1.85 vs. M = 6.21, SD = 2.80, t(259,721) = 6.964, p< .001). Both groups presented a pro-scientific attitude (i.e., above the 5.5 neutral value), with t(192) = 18.92, p < .001, in the pro-vaccine group and t(156) = 3.16, p = .002, in the anti-vaccine group. Results are presented in Fig 2.

Fig 2

Attitudes toward science in the pro-vaccine and anti-vaccine groups.

We also investigated how declared active involvement correlated with attitudes toward science. In the pro-vaccine group, involvement positively correlated with pro-scientific beliefs (τ = .20, p = .001), while in the anti-vaccine group, there was no significant correlation (τ = .10, p = .091).

3.4. Perceived vaccine-related knowledge basis of outgroups

We also wanted to investigate how the pro0vaccine and anti-vaccine groups perceive each other’s knowledge about vaccines, i.e. whether it is perceived as based on science or not. We found no significant differences in how these two groups perceived each other’s knowledge (t(348) = .399, p = .69). These perceptions of the basis of outgroup knowledge about vaccines were below the neutral value of 5.5 (on a scale from 1 –nonscientific to 10 –scientific): t(192) = -5.93, p< .001, for the pro-vaccine group’s responses and t(156) = -4.79, p< .001, for the anti-vaccine group’s responses. Results are presented in Fig 3.

Fig 3

Perceived outgroup vaccine-related knowledge bases (higher = more scientific).

4. Discussion

Studies on the attitudes of opponents of vaccination and the motives driving them to not vaccinate their own children have so far been analyzed in a way detached from the perspective of group processes [19, 37, 38]. Our attempt sought to account for this context. Vaccination opponents and supporters interact with each other both within and between groups (usually online) and influence each other. Our understanding of the attitudes and motives of opponents of vaccination can therefore become more complete when both groups (i.e., opponents and supporters of vaccination) are analyzed simultaneously.

Let us remember that being actively involved in the discussion about vaccination was the criterion for selecting people into the study groups (in addition to being pro-vaccine or anti-vaccine). Involvement in discussions on vaccination positively correlated with group identity both in the pro-vaccine group and the anti-vaccine group. This means that active participation in such discussions was a good criterion for selecting people who can be considered members of the group.

In line with our assumptions, the pro-vaccine group had a stronger overall identity than the anti-vaccine group. It should be stressed that all modes of identification (Superiority, Commitment, Importance and Deference) were significantly stronger in the pro-vaccine group. It turned out that in the group of vaccination supporters, group identity quite strongly positively correlated with attitudes toward science. Interestingly, a positive relationship between level of identification and acceptance of science also occurred in the anti-vaccine group. It should be noted that in the pro-vaccine and anti-vaccine groups, the correlation between positive attitude toward science and the overall level of social identity was similar (τ = .240 vs. τ = .214, respectively).

The results of our research show that pro-vaccine individuals hold a strongly positive attitude toward science and are significantly more pro-scientific than anti-vaccine individuals. The result for vaccination supporters is consistent with our assumption that their activity is at least partially based on belief in science. However, the fact that anti-vaccination group members have less positive attitudes toward science than pro-vaccination group members does not mean that these attitudes are clearly negative. On the contrary, most participants who declared that they are against vaccination also agreed with the statements that it is rational to believe only what is scientifically proven and that science is the most effective means of finding truth. This result supports the assumption that vaccine rejection does not necessarily mean rejection of science, and there are different reasons for being skeptical about vaccines [12, 30]. This pattern of results challenges the previous literature in which vaccine rejection has often been treated as a manifestation of anti-science [3, 20] because although rejecting vaccines is a form of rejecting something scientifically proven, it is not equivalent to rejecting science in general. A very important result is that while positive attitudes toward science were most strongly associated with their group’s Importance, the anti-vaccine group exhibited the strongest association between pro-science and Superiority. This may indicate that pro-vaccine individuals are pro-scientific because they feel that their pro-scientific group is important to them, or they feel that being pro-vaccine is important because it fits their pro-scientific attitudes. In case of the anti-vaccine group, it seems that they exhibit pro-scientific attitudes because they believe that being anti-vaccine is actually scientifically correct—the anti-vaccine group is, in their eyes, superior in understanding science. This is in line with existing research indicating that anti-vaccine groups often use excerpts from scientific papers as support for their ideas. Anti-vaccinationists can build their social identity on the belief that they possess unique, hard-earned knowledge about the true nature of vaccines—knowledge that is unavailable to the majority of the public, who are ‘misled by pharmaceutical companies and their bribed physicians’. As Motta, Callaghan and Sylvester noted [39], this overconfidence is linked to opposition to the policy of mandatory vaccination.

We also compared attitudes toward science based on the extent to which individuals in the two groups considered themselves active in discussions about vaccination. The results indicated that in the pro-vaccine group, involvement positively correlated with pro-scientific beliefs, while in the anti-vaccine group, there was no significant correlation. This result supports our assumptions that in the case of vaccination supporters, a positive attitude toward science is one of the motivations for activity, and this activity (involvement in discussions) goes hand in hand with a sense of identification with the group. In the case of opponents of vaccination, the attitude toward science, as we assumed, is not the main motive for engaging in group activity; hence, there are no significant correlations between attitude toward science and group identity.

It seems that the motive driving supporters of vaccination to actively engage in the group is consistent and is connected with, among other things, support for science and its discoveries; with trust in modern, conventional medicine; and with the perception of threats to the population due to infections carried by non-vaccinated people. In contrast, the factors motivating group membership among those who oppose vaccination are more diverse, therefore they have a weaker group identity. The minority group usually has a stronger group identity than a majority group [27]; especially with regard to uncontrollable characteristics (e.g. skin color, nationality). It turned out that this effect did not apply to the anti-vaccine group. This may be because people rarely think of themselves in terms of being pro-vaccine vs. anti-vaccine. If it happens to them (e.g. in the course of internet discussions), pro-vaccinationists see the homogeneity of their group (everyone believes in science) which strengthens the sense of group identity, while anti-vaccinationists see the heterogeneity of their group (some of them believe in science, others do not; some are afraid of epidemics, others are not; some believe in conspiracies, others do not).

We assumed (and it is consistent with the presented results) that a clearly and unequivocally positive attitude toward science is a common and important factor of support for vaccines and that the group of vaccine rejecters is more heterogeneous in this respect. Some vaccine rejecters support science, others are skeptical of it, and still others reject it [31]. However, in our research we did not find a factor which would clearly unite this group. These results are in line with existing research on vaccine hesitancy. For instance Kahan [40] showed that anti-vaccine attitudes have a global nature, without a single reason behind them, nor are they associated with particular political views, religiosity, science comprehension and their anti-scientific beliefs (p. 47). Atwell et al. [41] and Reich [42] indicate that vaccine rejecters simply believe there to be a ‘healthier alternative’ in form of natural medicine or similar practices, not necessarily linked to anti-scientific beliefs per se.

The current study also assessed how both groups evaluate outgroup knowledge about vaccination. The results revealed that anti-vaccine individuals evaluated the knowledge of pro-vaccine individuals negatively, and vice versa. Apart from ingroup favoritism and outgroup negativity, these results can be explained by a kind of lay epistemology that Ross and Ward [43] call naïve realism. According to them, naïve realism is based on an individual’s belief that his or her opinions and preferences (here, about vaccination) stem from an objective, impartial review of facts and evidence. If someone disagrees with these (here, pro-or anti-vaccine) opinions, it means that he or she has no access to the true facts, is unable to evaluate them objectively, or is biased and blinded by his or her own interests or ideology.

Based on our research and referring the paper by Rutjens et al. [30, p. 23], we believe that the statement “science does not speak with a single voice” can be recast as “vaccine skepticism does not speak with a single voice” to reflect attitudes toward science. In other words, there may be various types of anti-vaccine individuals—those who reject science in general and those who reject vaccines for other reasons while accepting the scientific method and published results [12, 30]. These ‘other’ motivations may include an egocentric strategy of relying on collective immunity without vaccinating, to protect one’s family from even the slightest side effects (or even pecuniary costs) of vaccines.

The presented research has some limitations, stemming both from theory (or lack thereof) and difficulties reaching active anti-vaccine individuals. Primarily, the sample could be larger, however the 350 people in our study were the most we could find in a representative group of 11 000 participants. Particularly hard to find were active anti-vaccine individuals, which stems from the rather low percentage of anti-vaccine individuals in the population. Another serious limitation is that the associations between group identity and vaccine rejection are strongly based on attitudes toward science, which is only one of many possible reasons to become a pro-vaccine or anti-vaccine activist. This, however, is representative of the fact that we know relatively little about the reasons behind vaccine rejection, far from any unified theory, and increasingly more evidence shows that vaccine rejection may have different causes than other anti-scientific beliefs.

Summing up the results of our research, we would like to state that an approach based on the analysis of group processes allowed us to identify several important things. First, as we have mentioned above, vaccination supporters believe in science more than opponents of vaccination, but the latter group is not as homogeneous as is usually perceived. Second, contrary to widespread opinion, the group identity of vaccination opponents turned out to be weaker than that of vaccination supporters. These results show the prospective empirical value of adding an approach based on group processes to the analysis of anti-vaccine attitudes. It is, for example, extremely interesting what role online discussion with people who have different opinions about vaccination plays in shaping group identity. Similarly, it seems important to investigate whether such discussion strengthens pre-existing attitudes on both sides of the dispute or causes a more careful analysis of the respective arguments. These are only examples of questions worth answering by conducting research on anti-vaccine attitudes and behavior based on the group processes perspective.

(SAV)

Click here for additional data file.

19 Apr 2021

PONE-D-21-02562

Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science.

PLOS ONE

Dear Dr. Polak

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

I have read the ms and the feedback provided by three expert reviewers. All Rs found merit in this research but they also raise important issues that prevent me from accepting the ms in its present state. A particular opportunity here is that R2 is a qualitative researcher and this is an opportunity for you to make sure that the concepts and language is accessible even to those who are not well versed in some aspects of your study. R1 and R2 are concerned that the rationale is under-developed and does nor motivate the hypotheses. R2 also points out that the abstract needs to better represent the study. R3 on the other hand invites you to consider the role of group membership and the role it plays more prominently.

In addition to these I would like you to consider:

1. Language and style needs some work

2. Please elaborate on Ariadna Nationwide Research Panel, e.g. where, which country, public daya or not?

3. Provide full list of items as supplement if you have not done so already.

4. Detail the software you used

5. Figures appeared to be directly copied and pasted from SPSS output? While this is not a major issue it gives the manuscript a rahter naive (for lack of a better description). Consider editing the format. You may find other prreviosuly published papers useful for ideas.

6. More importantly, consider running additional analyses to determine which aspect of group membership is more influential on attittudues toward science and vaccination. I would even consider a more complex model where dimensions of identity predict attitudues toward vaccination via attitudues toward vaccination.

7. Please elaborate on the limitations of your research. This is important.

I would like to reiterate that my decision on the revised version of the ms will depend on how well you would be able to adress issues raised by the Reviewers and my own observations I detail above.

Please submit your revised manuscript by May 23 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Huseyin Cakal

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

2. Please amend your manuscript to include your abstract after the title page.

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In the manuscript entitled “Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science”, the authors make use of a sub-sample of 350 participants to examine attitudes towards science, social identification, and outgroup attitudes among people active in discussions pro- or against vaccines. It provides several interesting correlations between these variables, shedding light on the similarities and differences between vocal supporters and opponents of vaccines.

I found the overall argumentation and structure of the article rather difficult to follow. Although several aims and hypotheses were described, I did not find a clear research question in the manuscript. Additionally, there were some inconsistencies between hypotheses, measures, and the reported findings; most notably around the concept of social identity. I will elaborate on these concerns below.

Regarding the use of social identity in the current manuscript, some things did not become clear to me. My main concern is that the manuscript does not provide a clear reason why and how group identity/social identity should play a role among vaccine supporters and opponents. I am sure there are plenty of good reasons to study group processes here, but without a clear reason how it needs to be studied, I found it difficult to assess whether the methods were appropriate to do so. Could, for example, being personally pro- or anti vaccines not explain the same patterns in support for science?

The methods to assess social identification match the useful distinction between the four aspects of group identity (p.4). However, two things are unclear or inconsistent here. First, the hypotheses mention “level of group identity” as a singular measure, whereas the correlations are presented for each of the four aspects of group identity. Moreover, the single reliability coefficient in the measures section suggests that indeed all 16 items are considered as a single group identity scale. When considered separately, I think a factor analysis would add to the understanding of these four sub-scales. Second, the methods section lacks sample items for the identity scale. Based on the data set, I could not fully assess what the group identity exactly pertains clearly to in the items. I think this is especially important given my earlier point that it is unclear what the group identity adds in explanatory power over and above personally being pro- or anti vaccines. In the theory, I also would have liked to a see a specific discussion on this: what does a pro- or anti vaccine identity look like exactly?

I think the sub-sample drawn from the larger representative sample is fitting to test the hypotheses. The selection criteria are well-argued in the methods section. Perhaps the introduction could have been clearer that the current study concerns active/involved opponents and supporters of vaccines. Based on vaccination rates, vaccine supporters can either be an overwhelming majority group or form a minority and this can influence identification processes. Some more background information on this would help argue for the current hypotheses: how do the opponents and supporters compare to the general representative sample that they were drawn from? And what can be said beforehand about pro- and anti-vaccine groups in the current context, both practically and theoretically?

The methods and results sections contained some other puzzling information. I found the types of involvement very useful in order to get a better understanding of who the participants in the study were. However, the type of involvement was not mentioned in the introduction, hypotheses, nor methods and therefore came as a surprise to me. In the results and discussion, involvement seems to be discussed as a predictor. Additionally, the “outgroup identity” in Table 2 was not clear to me. I did not find this measure in the methods section.

For the general methodological approach of the manuscript, I think the language in the abstract, introduction, and hypotheses is not always consistent. For example, the abstract mentions “mechanisms”, suggesting group identity to potentially play a mediating role that is not test. On page 2, “understanding the predictors of […] attitudes towards vaccination” is mentioned, suggesting at least some kind of inference about predictors rather than correlations. The correlational analyses are correct as far as I can tell, but I think in some cases a regression analysis would be helpful to examine the unique contribution of, for example, the different aspects of identity in certain relations.

Finally, I would suggest to consider addressing some conceptual and theoretical concerns in the introduction and discussion. In the introduction, I would suggest to address in more depth what the expectations are for the identification processes of opponents of vaccines. Based on the current overview of the literature, it seems that it is already clear that anti-science is not central to their ingroup or beliefs. This would make answering the question, like it is done in the current study, superfluous. Another thing I think would be important for the discussion, in what “belief in science” or “trust in science” would mean for either group. This might be hard to answer based on the measures in the study, but I could imagine that each group has “their own science” to trust, meaning that the absolute level of trust in science is less relevant than which scientific findings are believed by either group.

Minor points

The methods section mentions that the questionnaire consists of three parts, but only two are explained

I think there is a slight mismatch in the number of participants between abstract and methods section.

The second hypothesis is not formally tested: it states that the correlation between level of identity and acceptance of science is stronger in the pro-vaccine group. The findings are presented in Table 4, but no conclusion is drawn about the difference between the pro- and anti-vaccine groups.

The writing is generally clear. Some expressions or phrases need to be made more consistent or clear, such as the “predictors”, “mechanisms” that I mentioned before, but also a phrase like “another question that arises” (p.5) needs some substantiation: where does this arise from exactly? Other than that, I thought that the language and writing was clear.

Reviewer #2: This well-written article explores group identity in pro- and anti- vaccination individuals who self-identify as being active in conversations about this issue. It finds that pro-vaccination participants have stronger group identity, both groups identify as pro-science more than not, and both take a dim view of the other’s scientific basis. With some adjustments, I think this article would be a good addition to the literature.

I note that the abstract does not really pull out the most interesting findings and urge the authors to edit it to play to the article’s strengths.

The authors build from a basis of thinking about vaccinators and non-vaccinators as people who engage in (non)scientific thinking, but at line 97-99 this appears somewhat jarring as a basis for the study, even though the authors have offered some basis for why we might think this. I think the reason it doesn’t feel right here is that pro-vax supporters are assumed to be pro-science and that this is a causal factor for their support for vaccination. I will return to this topic later. For now, I think these lines can be improved by offering evidence as to why we might believe that pro-vaxxers are also passionately pro-science. There may be academic articles that cover this point, and the authors should look for some. Failing that, they could rest on the evidence that many skeptic organisations (who are first and foremost science lovers!) engage quite heavily with the vaccination issue, including through podcasts, articles etc, and cite some examples. I feel that this evidence would make the authors’ assertion that science views matter to pro-vax views more convincing, and hence make the study’s premise more convincing.

I note that I’m a qualitative researcher so I cannot comment on the article’s quant methods. It did strike me that the N is small, and I’d like the authors to defend in the article whether the study is sufficiently powered.

Because I’m not a psychologist, I prefer explanations of results that non-quant people can understand. So, for example, at lines 212, I’d like the sentence about significant differences to actually explain who thought what. Likewise at line 221, please explain what this means in concrete language about the people in your study. At line 260, please add something that demonstrates what it means (you talk about attitudes toward science but not in which direction) Eg “meaning that pro-science people had stronger group identities, whether they were pro or anti vaccine.” This is the kind of concrete language I’m asking for throughout. At line 262, again please explain in concrete terms what the first half of this sentence is saying about your population. Including a basic explanation of what your findings actually mean in concrete terms will make your article more readable and probably more likely to be cited by others outside your discipline.

At line 290 , once again we revisit the idea that pro science motivates pro- vax behaviour / attitude / action. I’d like the authors to consider, and include in their limitations, the idea that it might swing the other way. It could also be that pro-vax people get socialised into being strong supporters of science! I raise this point because this was my journey; it might well be that of other people. Consider that causation runs the other way.

Line 313, you find that there was not a factor that clearly unites vaccine rejecters. This has been discussed in other literature – see

Kahan, D. M. 2014. Vaccine Risk Perceptions and Ad Hoc Risk Communication: An Empirical Assessment. In CCP Risk Perception Studies: Yale Law School; Harvard University - Edmond J. Safra Center for Ethics.

And then see critiques that he, too, might not have looked for the right thing:

Attwell, K., and Smith, D. T. 2017. Parenting as Politics: Social Identity Theory and Vaccine Hesitant Communities. International Journal of Health Governance 22 (3):183-198.

For your reflections on how the groups view each other’s scientific knowledge at lines 326-332, some other studies may be useful:

Attwell, K., Smith, D. T., and Ward, P. R. 2018. ‘The Unhealthy Other’: How vaccine rejecting parents construct the vaccinating mainstream. Vaccine 36 (12):1621-1626. doi: https://doi.org/10.1016/j.vaccine.2018.01.076.

Rozbroj, T., Lyons, A., and Lucke, J. 2019. The mad leading the blind: Perceptions of the vaccine-refusal movement among Australians who support vaccination. Vaccine 37 (40):5986-5993. doi: https://doi.org/10.1016/j.vaccine.2019.08.023.

And for your discussion of the motivations of anti-vaccinators lines 338-340, these may be useful:

Ward, P. R., Attwell, K., Meyer, S. B., Rokkas, P. R., and Leask, J. 2017. Understanding the perceived logic of care by vaccine-hesitant and vaccine-refusing parents: A qualitative study in Australia. PLoS One 12 (10).

Reich, J. 2014. Neoliberal Mothering and Vaccine Refusal: Imagined Gated Communities and the Privilege of Choice. Gender & Society 28 (5):679-704. doi: 10.1177/0891243214532711.

All suggestions are made for the authors’ benefits with no expectations that they cite any of them unless they consider them useful for their framing and wider analysis.

Reviewer #3: The paper provides a helpful insights to an important topic, which hopefully will have a positive impact in both practical and theoretical aspects. I was thrilled by the decision of the researcher to adopt a group dynamic approach to the topic, which does provide more depth of understanding. I have no major comments on the manuscript, except one concern in regard how the researchers talk about group identity, as following:

The researchers seem to base their approach on assumption that science is “the” fundamental force behind the position toward vaccines and activism around it. Thus, they measured attitude towards science and science-based social identity, and used them in their analysis. Although the authors believe that “any domain (value, symbol, attitude) can become the basis for group identity, provided that it is accepted and considered valid by the members of the group” (P.4, Line 91) they seemed to be over focused on science alone as base for group membership, and ignoring other possible sources, which might explain some possible flaws in the analysis and conclusions. Other possible sources for group identity can be the individual liberal rights (resisting the tyranny of state), which is known to be a common part of anti-vaccine debates. It is true that the paper makes an excellent argument that anti-vaccine groups are not necessarily have a common negative attitude toward science , and show in the result that such attitude seems not to be the base for their group identity; However, I’m concerned about framing such results as denying the role of social identity, in general. So, if the authors aims to explore specifically the role of science-based social identity, then I would suggest making that clear, wherever needed, to avoid misleading conclusions such as “social identity of vaccination rejecters is weaker than that of vaccination supporters”.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Review.docx

Click here for additional data file.

24 Jun 2021

Dear Editor and Reviewers,

Thank you for the insightful reviews. Below are our responses to each comment.

Editor’s comments:

R1 and R2 are concerned that the rationale is under-developed and does nor motivate the hypotheses.

� We extensively modified the introductory part of the manuscript, including a separate part with hypotheses and their justification.

R2 also points out that the abstract needs to better represent the study

� We re-wrote the abstract to better indicate what the research is about.

Language and style needs some work

� We changed the style in many parts of the manuscript and hope that the language is better now; extensive changes requested by the Reviewers forced us to rewrite parts of the manuscript from scratch and we hope these parts are also more readable.

Please elaborate on Ariadna Nationwide Research Panel, e.g. where, which country, public daya or not?

� We provided a detailed explanation in the Method section

Provide full list of items as supplement if you have not done so already.

� We added a reference to Roccas’ social identity scale in the text. Since this is not our own scale, we don’t feel it would be ethical to reproduce it in full. All other questions are presented in full within the Methods section.

Detail the software you used

� We added information about using SPSS for analyses into the Results section

Figures appeared to be directly copied and pasted from SPSS output?

� We changed the figures to ones produced outside SPSS.

More importantly, consider running additional analyses to determine which aspect of group membership is more influential on attittudues toward science and vaccination

� We ran regression analyses to determine which aspects of group identity are most important for attitudes toward science, with interesting results.

Please elaborate on the limitations of your research. This is important.

� We added a paragraph about study limitations to the Discussion.

Reviewer 1’s comments:

I found the overall argumentation and structure of the article rather difficult to follow. Although several aims and hypotheses were described, I did not find a clear research question in the manuscript. Additionally, there were some inconsistencies between hypotheses, measures, and the reported findings; most notably around the concept of social identity. I will elaborate on these concerns below.

� We extensively rewrote the introduction and discussion and hope that the paper is now easier to follow.

Regarding the use of social identity in the current manuscript, some things did not become clear to me. My main concern is that the manuscript does not provide a clear reason why and how group identity/social identity should play a role among vaccine supporters and opponents. I am sure there are plenty of good reasons to study group processes here, but without a clear reason how it needs to be studied, I found it difficult to assess whether the methods were appropriate to do so. Could, for example, being personally pro- or anti vaccines not explain the same patterns in support for science ?

� We added more justification for investigating the relation between group identity and vaccine-related attitudes. The question whether being ‘personally pro/anti vaccine explain the same patterns’ is quite interesting and we cannot answer it based on our study; however we did use the well-established social identity theory, along with the social identity scale, to investigate social identity of pro-vaccine and anti-vaccine individuals, so there is very little discrepancy between theory and method. The main goal was to use this established theory to investigate the particular group of pro-vaccine and anti-vaccine people.

The methods to assess social identification match the useful distinction between the four aspects of group identity (p.4). However, two things are unclear or inconsistent here. First, the hypotheses mention “level of group identity” as a singular measure, whereas the correlations are presented for each of the four aspects of group identity. Moreover, the single reliability coefficient in the measures section suggests that indeed all 16 items are considered as a single group identity scale. When considered separately, I think a factor analysis would add to the understanding of these four sub-scales.

� We changed the wording to ‘levels’. Also, we added alpha coefficients for all subscales.

Second, the methods section lacks sample items for the identity scale. Based on the data set, I could not fully assess what the group identity exactly pertains clearly to in the items. I think this is especially important given my earlier point that it is unclear what the group identity adds in explanatory power over and above personally being pro- or anti vaccines.

� We refrained from providing a full set of items for Roccas’ scale, since it is an existing scale by another author and we don’t have the right to fully reproduce it. We did add a few sample items into the Methods section.

In the theory, I also would have liked to a see a specific discussion on this: what does a pro- or anti vaccine identity look like exactly?

�We have developed, as far as possible on the basis of our research, the characteristics of the group identity of both groups

I think the sub-sample drawn from the larger representative sample is fitting to test the hypotheses. The selection criteria are well-argued in the methods section. Perhaps the introduction could have been clearer that the current study concerns active/involved opponents and supporters of vaccines.

� We added a more clear statement that this is the case

Based on vaccination rates, vaccine supporters can either be an overwhelming majority group or form a minority and this can influence identification processes. Some more background information on this would help argue for the current hypotheses: how do the opponents and supporters compare to the general representative sample that they were drawn from? And what can be said beforehand about pro- and anti-vaccine groups in the current context, both practically and theoretically?

� We included these statistics in the description of our sample. Indeed, pro-vaccine individuals were thankfully a vast majority (nearly 75%), and anti-vaccine individuals were only 5% of the sample. The final sample size was constricted by the numer of anti-vaccine participants who were also active, which was about a fourth of the total anti-vaccine sample. While this may suggest that anti-vaccine individuals feel a stronger association with their minority group, these two groups operate mostly online in information bubbles, so anti-vaccine individuals may as well feel as if their group is larger than it actually is. This is why we did not state any theoretical assumptions regarding their identity based on just one group being larger or smaller than the other.

The methods and results sections contained some other puzzling information. I found the types of involvement very useful in order to get a better understanding of who the participants in the study were. However, the type of involvement was not mentioned in the introduction, hypotheses, nor methods and therefore came as a surprise to me. In the results and discussion, involvement seems to be discussed as a predictor. Additionally, the “outgroup identity” in Table 2 was not clear to me. I did not find this measure in the methods section.

�Thank you for this comment. Commitment was an important criterion for selecting proponents and opponents of vaccination for the study. We have now devoted more to this factor in the introduction, method and hypotheses. As for ‘outgroup identity’, we included this measure by mistake, it has now been removed from the manuscript (It was an attempt at measuring how the two groups perceive each other’s identities, but it does not fit into the theoretical framework of this paper).

For the general methodological approach of the manuscript, I think the language in the abstract, introduction, and hypotheses is not always consistent. For example, the abstract mentions “mechanisms”, suggesting group identity to potentially play a mediating role that is not test. On page 2, “understanding the predictors of […] attitudes towards vaccination” is mentioned, suggesting at least some kind of inference about predictors rather than correlations.

� We thoroughly rewrote the manuscript to avoid this type of language, as we meant ‘mechanisms’ in a theoretical context and not as a mediator; we also conducted regression analyses which allow us to speak of ‘predictors’ now, but this assumption still feels too strong.

The correlational analyses are correct as far as I can tell, but I think in some cases a regression analysis would be helpful to examine the unique contribution of, for example, the different aspects of identity in certain relations.

� We conducted regression analyses and added them to the manuscript, as their results are indeed very interesting.

Finally, I would suggest to consider addressing some conceptual and theoretical concerns in the introduction and discussion. In the introduction, I would suggest to address in more depth what the expectations are for the identification processes of opponents of vaccines. Based on the current overview of the literature, it seems that it is already clear that anti-science is not central to their ingroup or beliefs. This would make answering the question, like it is done in the current study, superfluous. Another thing I think would be important for the discussion, in what “belief in science” or “trust in science” would mean for either group. This might be hard to answer based on the measures in the study, but I could imagine that each group has “their own science” to trust, meaning that the absolute level of trust in science is less relevant than which scientific findings are believed by either group.

� We added more commentary on this in the Discussion, however it does not seem possible to clearly answer these issues with our data.

The methods section mentions that the questionnaire consists of three parts, but only two are explained

� We added a clear indication that ‘attitudes toward science’ and ‘perceived scientific knowledge’ were the third part.

I think there is a slight mismatch in the number of participants between abstract and methods section.

� Thank you for noticing this, we amended the mistake

The second hypothesis is not formally tested: it states that the correlation between level of identity and acceptance of science is stronger in the pro-vaccine group. The findings are presented in Table 4, but no conclusion is drawn about the difference between the pro- and anti-vaccine groups.

� We modified the hypothesis to reflect what we initially meant more precisely.

a phrase like “another question that arises” (p.5) needs some substantiation: where does this arise from exactly?

� We changed this wording so it does not create confusion.

Reviewer 2’s comments:

I note that the abstract does not really pull out the most interesting findings and urge the authors to edit it to play to the article’s strengths.

� We thoroughly rewrote the abstract to make it reflect the study better

I think the reason it doesn’t feel right here is that pro-vax supporters are assumed to be pro-science and that this is a causal factor for their support for vaccination.

� Thank you for noticing this; the relation between pro-science and vaccination support can go either way, we changed the wording to reflect it.

It did strike me that the N is small, and I’d like the authors to defend in the article whether the study is sufficiently powered.

� The small N is a result of the anti-vaccine group being, thankfully, only a small percentage of the population (in our research it was only 5% of the primary sample, we added these statistics to the text) and an even smaller percentage are active anti-vaccine individuals (about 25% of the entire anti-vaccine group). The base sample for the study was over 10500 participants, and the low final Ns show how few active anti-vaccine people there actually are.

Because I’m not a psychologist, I prefer explanations of results that non-quant people can understand.

� While we are used to presenting results in standard APA style, we tried our best to explain results in concrete terms wherever possible. However, we felt that it is necessary to keep the standard wording alongside, not to draw critique from ‘quant people’.

we revisit the idea that pro science motivates pro- vax behaviour / attitude / action. I’d like the authors to consider, and include in their limitations, the idea that it might swing the other way. It could also be that pro-vax people get socialised into being strong supporters of science!

� Of course this argument is very valid; the tendency to assume that pro-science causes vaccine support seems to stem from existing research on ‘anti-scientific beliefs’, however we agree that especially for vaccine rejection and support the relation may go the other way – people are known to reject vaccines and at the same time support science, even though this science becomes very distorted with time. We changed our wording to less causal.

Line 313, you find that there was not a factor that clearly unites vaccine rejecters. This has been discussed in other literature

� Thank you for the valuable literature suggestions. We used some of the cited papers, mainly to support the idea that vaccine rejection may have various causes.

Reviewer 3’s comments:

The researchers seem to base their approach on assumption that science is “the” fundamental force behind the position toward vaccines and activism around it. Thus, they measured attitude towards science and science-based social identity, and used them in their analysis. Although the authors believe that “any domain (value, symbol, attitude) can become the basis for group identity, provided that it is accepted and considered valid by the members of the group” (P.4, Line 91) they seemed to be over focused on science alone as base for group membership, and ignoring other possible sources, which might explain some possible flaws in the analysis and conclusions.

� It seems that we may have written some sections of the article in a way that caused a misunderstanding. We did not assume that pro-science or anti-science is the only cause for vaccine support or rejection. Especially we did not assume that attitudes toward science are the basis of group identity. Our main assumption was in fact that being pro-vaccine or anti-vaccine causes one to perceive themselves as part of a larger group, and identifying themselves with it. Group identity of pro-vaccine and anti-vaccine group members was our main focus of research. Whether these groups exhibit pro-scientific or anti-scientific beliefs was more of a side question which could explain people being pro-vaccine/anti-vaccine, as well as identifying themselves with each group in different ways. We changed the order of most of the paper to better indicate it, and also indicated some other possible sources of anti-vaccine beliefs.

It is true that the paper makes an excellent argument that anti-vaccine groups are not necessarily have a common negative attitude toward science , and show in the result that such attitude seems not to be the base for their group identity; However, I’m concerned about framing such results as denying the role of social identity, in general. So, if the authors aims to explore specifically the role of science-based social identity, then I would suggest making that clear, wherever needed, to avoid misleading conclusions such as “social identity of vaccination rejecters is weaker than that of vaccination supporters

� We aimed to explore the group identity (or group identification) of pro-vaccine and anti-vaccine people. It could be said that we wanted to specifically explore vaccine-based identity, attitudes toward science being one of its particular aspects (to test the hypothesis that vaccine support and vaccine rejection are caused by science support and science rejection, respectively, which turned out to be untrue. Thank you for pointing out that our paper can be interpreted in this way; we changed the wording of statements which we feel may have suggested that science is the only or main basis for group identity of vaccine supporters and rejecters.

Submitted filename: response to reviewers.docx

Click here for additional data file.

11 Oct 2021

PONE-D-21-02562R1

Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science.

PLOS ONE

Dear Dr. Polak,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. While one reviewer recommend accept, the other recommended substantial changes to nearly all aspects of the paper from tiny details to big picture issues.

Please submit your revised manuscript by Nov 20 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Peter Karl Jonason

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In the revised manuscript “Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science”, most of the suggestions made by reviewers seem to be addressed in one way of another. Additionally, the relevance of the topic of attitudes towards vaccination remains an important one. Nonetheless, I do think that the structure of the manuscript still contains omissions between the theoretical rationale, hypotheses, and reported findings.

One of the problems that is still in the manuscript, is how “level of involvement” is treated. First, the findings on level of involvement are reported quite extensively in section 3.2 of the results. Also, on line 198, it says that there is an aim of investigating this variable in relation to group identity. However, there is no specific theory, hypothesis, nor a clear measure for it. Moreover, because the participants are selected on this variable, I think section 3.2 can logically only be used as some kind of description of either group. The conclusion on page 19 (in the discussion) seems to indicate that this was the intended way of using level involvement as well.

Another problematic part of the manuscript remains the role of social identity and how it is operationalized. The reasons why social identity is used is not entirely clear even in the revised manuscript, especially when it comes to the four dimensions of social identity. In the expectations in the theoretical framework, but also in the hypothesis itself, there is no mention of difference between any of the four types of identification. Yet, these are measured and reported separately in the methods and results section. I find the regression analysis of the various social identity measures interesting but also would like to know how strongly they are related. Additionally, ideally there would be some kind of control variable for religiosity in these analyses in order to rule out alternative explanations for attitudes towards science. I do not know whether such additional measures are available.

Additionally, I think there is still work to be done in order to completely make the case for what the identities are that are measured; what their content could be. For example, some of the individual differences that are mentioned on page 3 include orthodox religiousness and moral purity concerns, both of which could easily be part of shared ingroup norms and not individual differences. The authors do not make clear how, then, the intergroup relations perspective adds to this line of research. One solution for this, I suggest, would be to provide context as to who individuals in either group could be, given the current sociopolitical situation in Poland. On page 6, there is some information on why the anti-vaccine group might have more diverse motivations and therefore identify less as anti-vaccine supporters. In many contexts, this might seem counter-intuitive because the majority group in most situations would identify less strongly. I urge the authors to expand on this point of view by using the national context and existing literature. The Rutjens et al article on line 168 might be useful here, but then some reflection on how the current study adds to these findings is important.

I would recommend changing the structure of the theoretical framework to some extent, in order to be able to see the main hypotheses and their argumentation sooner. For example, I find the hypothesis about perception of anti-science attitudes of the other group interesting, but there is barely any theorizing on it. The few lines around line 212 mention “basic rules of social cognition” while only when discussing the hypotheses, the theoretical foundations of outgroup bias are mentioned (line 231) but not discussed in depth.

All in all, quite a few confusing aspects of the initial manuscript are resolved in this revision and the correlates of pro- and anti-vaccine individuals are reported clearer now. Nonetheless, my points above still make it difficult sometimes to assess the necessity and purpose of some of the measures.

Minor issues

Space missing on line 69 and line 128

The reference on opinion-based groups should maybe already been mentioned on line 136.

The discussion paragraph starting at line 459 invokes all kinds of motivations for group membership, but I suggest making clearer what is known from previous literature versus what the findings of the current research are.

The discussion does not engage much with the theory: the findings are explained well, but it is not clear what it means for existing theory.

Reviewer #2: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

22 Nov 2021

Reviewer #1: In the revised manuscript “Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science”, most of the suggestions made by reviewers seem to be addressed in one way of another. Additionally, the relevance of the topic of attitudes towards vaccination remains an important one. Nonetheless, I do think that the structure of the manuscript still contains omissions between the theoretical rationale, hypotheses, and reported findings.

One of the problems that is still in the manuscript, is how “level of involvement” is treated. First, the findings on level of involvement are reported quite extensively in section 3.2 of the results. Also, on line 198, it says that there is an aim of investigating this variable in relation to group identity. However, there is no specific theory, hypothesis, nor a clear measure for it. Moreover, because the participants are selected on this variable, I think section 3.2 can logically only be used as some kind of description of either group. The conclusion on page 19 (in the discussion) seems to indicate that this was the intended way of using level involvement as well.

Thank you for pointing out that our description of of „level of involvement” may be unclear. The basis for referring to this construct is the contemporary approach to group formation, in which involvement is treated as a fundamental process (Moreland and Levine, 1994). In addition to the question of defining oneself in terms of belonging to a group, contemporary analyses place a strongly emphasis on different forms of involvement in group functioning (Hogg, 1992). Therefore, in our study, the level of involvement in various activities was the criterion for selecting people into the study groups (in addition to strong pro- and anti-vaccine beliefs). Whereas in the paragraph concerning the aims of the study (on lines 189-190) we stated that the level of involvement was one of the study variables, declaring that the aim of our study was, among other things, to test its relationship with group identity and attitudes towards science. In fact, it was not a variable, but a selection criterion so the word ‘relationship’ was misleading. In the current manuscript, we have removed this phrase and also referenced the above view on the importance of involvement in group functioning.

Another problematic part of the manuscript remains the role of social identity and how it is operationalized. The reasons why social identity is used is not entirely clear even in the revised manuscript, especially when it comes to the four dimensions of social identity. In the expectations in the theoretical framework, but also in the hypothesis itself, there is no mention of difference between any of the four types of identification. Yet, these are measured and reported separately in the methods and results section. I find the regression analysis of the various social identity measures interesting but also would like to know how strongly they are related.

We have tried to better clarify the reasons for using social identity according to Roccas, and we have added more information in the manuscript about the four dimensions of social identity and the possible differences and relationships between them. Based on the results of our study, we performed additional analyses on the relationship between these dimensions of social identity (correlations between identity modes and expanded CFA).

Additionally, ideally there would be some kind of control variable for religiosity in these analyses in order to rule out alternative explanations for attitudes towards science. I do not know whether such additional measures are available.

Unfortunately, religiosity of the participants in our study was not measured. While we understand that measuring religiosity would add a possible additional level of explanation, so would measuring other variables known to relate to vaccine rejection and/or anti-science, which however would require a much larger research project, whereas we focused on group identity from the beginning.

Additionally, I think there is still work to be done in order to completely make the case for what the identities are that are measured; what their content could be. For example, some of the individual differences that are mentioned on page 3 include orthodox religiousness and moral purity concerns, both of which could easily be part of shared ingroup norms and not individual differences. The authors do not make clear how, then, the intergroup relations perspective adds to this line of research. One solution for this, I suggest, would be to provide context as to who individuals in either group could be, given the current sociopolitical situation in Poland. On page 6, there is some information on why the anti-vaccine group might have more diverse motivations and therefore identify less as anti-vaccine supporters.

As stated above, we have added more information about the four modes of group identity. Thank you for pointing out that the individual variables mentioned can be linked to norms, which are an important part of the structure of a group.

We understand that the study of group processes covers an extremely wide range of phenomena. One can focus, among other, on the individual cognitive and motivational processes leading to group behavior, on the interconnection of individual, interpersonal and social processes, on the cohesiveness of the group and its identity, etc. (Moreland & Levine, 1994).

In the manuscript, we mentioned traditional research on seeing which individual variables in a population might be predictors of attitudes toward science, toward vaccination, or toward climate change (e.g., whether climate change denial is associated with political conservatism). In contrast, we purposively recruited two groups for the study that had to meet not only the criterion of acceptance or rejection of vaccinations but also the criterion of engagement in activity related to these attitudes - to describe and compare an important group process such as group identity. In short – while existing research investigates individual differences which may be associated with group norms or group behavior (which is an assumption worth researching by itself), we investigated how much and in what manner the active pro- and anti-vaccine individuals identify with the groups they belong to – a key factor to understand whether their attitudes and beliefs are individual in nature or whether they truly identify with their respective groups

The proposition to include the current sociopolitical situation in Poland is interesting, as polls indicate that being anti-vaccine is common in supporters of the ruling Law and Justice party (a conservative, populist party), and that there is a general distrust in the government typical for the Eastern Bloc, which fuels vaccine hesitancy. However, we refrained from discussing this in our manuscript, as we did not directly measure political attitudes, so it would be pure speculation. In truth, we are not aware of any research directly describing the ‘active’ vaccine supporters and rejecters in Poland, and these groups may be different from the general anti-vaccine and pro-vaccine demographics.

In many contexts, this might seem counter-intuitive because the majority group in most situations would identify less strongly. I urge the authors to expand on this point of view by using the national context and existing literature. The Rutjens et al article on line 168 might be useful here, but then some reflection on how the current study adds to these findings is important.

We confirm the above remark. The minority group usually has a stronger group identity than a majority group; especially with regard to uncontrollable characteristics (e.g. skin color, nationality). But this does not apply to the anti-vaccine group, which is probably because people rarely think of themselves in terms of pro-vaccinationist vs. anti-vaccinationist. If it happens to them (e.g. in the course of internet discussions), pro-vaccinationists see the homogeneity of their group (everyone believes in science) which strengthens the sense of group identity, while anti-vaccinationists see the heterogeneity of their group (some of them believe in science, others do not; some are afraid of epidemics, others do not; some believe in conspiracy, others do not). We added this reasoning to the discussion of our research findings.

I would recommend changing the structure of the theoretical framework to some extent, in order to be able to see the main hypotheses and their argumentation sooner. For example, I find the hypothesis about perception of anti-science attitudes of the other group interesting, but there is barely any theorizing on it. The few lines around line 212 mention “basic rules of social cognition” while only when discussing the hypotheses, the theoretical foundations of outgroup bias are mentioned (line 231) but not discussed in depth.

We changed the structure of the paper and added more theoretical background.

All in all, quite a few confusing aspects of the initial manuscript are resolved in this revision and the correlates of pro- and anti-vaccine individuals are reported clearer now. Nonetheless, my points above still make it difficult sometimes to assess the necessity and purpose of some of the measures.

Minor issues

Space missing on line 69 and line 128

The reference on opinion-based groups should maybe already been mentioned on line 136.

The discussion paragraph starting at line 459 invokes all kinds of motivations for group membership, but I suggest making clearer what is known from previous literature versus what the findings of the current research are.

The discussion does not engage much with the theory: the findings are explained well, but it is not clear what it means for existing theory.

We spell-checked the entire manuscript, and hope to have solved all problems including missing spaces. As regards the discussion, unfortunately there is very little theory and the research is very exploratory in nature; to our knowledge, there has been no research on the identity of pro- and anti-vaccine groups, let alone strictly active ones. We added clearer statements that most of the discussion refers directly to our research. We also added some suggestions on how to merge our research with existing theory, and especially how the results of our research differ from what would be expected based on earlier research on group identity, for instance that the minority may not have a stronger identity than the majority.

Submitted filename: Answer to reviewers.docx

Click here for additional data file.

9 Dec 2021

Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science.

PONE-D-21-02562R2

Dear Dr. Polak,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Peter Karl Jonason

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

13 Dec 2021

PONE-D-21-02562R2

Active pro-vaccine and anti-vaccine groups: their group identities and attitudes toward science.

Dear Dr. Polak:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Peter Karl Jonason

Academic Editor

PLOS ONE