What are the predictors of political incivility perceptions?

Abstract

Incivility in public discourse has become a central concern. This research develops an approach based on citizens’ perceptions of incivility to assess predictors of such perceptions. Starting from a definition of the concept in a multidimensional sense, we identified different types of incivility that can be easily translated into operational terms (using vulgar/insulting language, referring to another politician with racist/sexist/religious epithets, lying/inventing facts to attack political opponents, and resorting to violence against political opponents). Through a survey of a representative sample of the Italian population, we found that citizens’ perception of incivility is not uniform but varies depending on the context and individual characteristics (socio-demographic variables, news consumption and relationship with politics). Overall, the importance of disentangling the concept into different types has emerged, because continuing to speak of ‘incivility’ in a broad sense does not help to clarify the phenomenon nor to identify the consequences for citizens’ relationship with politics.

Keywords

Political incivility predictors social media desensitisation citizens perceptions

Introduction

Concern about the increase in political incivility is not only evident among scholars but also among citizens: in Italy, for example, 75.8% of respondents in a representative sample of the population believe that politics has recently become more uncivil (i.e. characterised by a lack of respect for others and/or democratic mechanisms), while 81.9% state that civility is important for the functioning of democracy.¹ In the US, 93% of respondents in recent polling believe that incivility is a problem, 75% that it is worsening, and 80% that it creates serious problems for society (Weber Shandwick, 2019). There is little doubt, therefore, that a broad awareness of an uncivil drift in politics exists. There are, however, doubts as to what exactly is meant by incivility. Or rather, what citizens mean when they report that it is increasing and is a significant problem in democratic societies.

The need to define incivility more precisely has been clear to scholars since Stryker et al. (2016: 574) defined it as a three-dimensional construct of ‘Utterance Incivility, Discursive Incivility, and Deception’ and Muddiman (2017) introduced the distinction between personal and public incivility. Indeed, the multidimensionality of the concept has become a shared heritage among scholars (Bormann, 2022; Hopp, 2019; Muddiman, 2019, 2021; Stryker et al., 2016). This structuring into different dimensions that refer to different ‘types’ of incivility not only helps us clear the field of ambiguities and opaque interpretations but also allows us to shed light on the variables that most influence citizens’ perceptions. If it is true that incivility is in the eye of the beholder (Herbst, 2010), knowing what citizens see will help us better understand the nature and evolution of the phenomenon.

The goal of this work is to measure citizens’ perceptions in relation to each type of uncivil behaviour by political actors and identify predictors of these perceptions. To do so, we have adopted a simple definition of political incivility that can be easily translated into operational terms. Based on the literature that considers not only the traditional dimension of impoliteness but also those dimensions that represent a threat to the ‘collective face’ – understood both as the set of institutional actors and as functioning mechanisms – we have identified three main dimensions of the concept (disrespect for others, for democratic values and for democratic institutions) each of which is linked to specific types of incivility: discursive, vulgar, informational, violent, and discriminatory. In this study, first, we recorded respondents’ perceptions of different types of incivility adopted by politicians to understand which were perceived more clearly; secondly, we identified predictors of perception using regression models.

Definition of the concept, main dimensions, and types of political incivility

Research about political incivility has recently grown worldwide, becoming an increasingly central field of investigation not only in the US but in other countries too (Walter, 2021). This expansion, however, poses new challenges to researchers, in relation to both the definition of the concept and its empirical detection. Standard American measures of incivility do not always work in other countries (Walter, 2021), and different ways of defining incivility lead to heterogeneous results that are difficult to compare (Van’t Riet and Van Stekelenburg, 2022).

Furthermore, since our study builds on previous research on citizens’ perceptions of political incivility, we should acknowledge that what is perceived to be uncivil is inherently subjective (Bormann, 2022) and changes according to communication norms that vary over time and place (Strachan and Wolf, 2012). However, Stryker et al. (2024), after surveying scholarly definitions and research on citizens’ perceptions of political incivility (Bentivegna and Rega, 2022; Bormann, 2022; Gervais, 2017; Hopp, 2019; Laden, 2019; Muddiman, 2017, 2019; Stryker et al., 2016), argue that, amidst the diversity of definitions, methods and findings, political incivility has a common conceptual core, that is, disrespect, towards one's political interlocutors or audience, and/or democracy itself.

In the wake of this study, we too are convinced that disrespect is at the core of the concept of incivility and that it can manifest in three main dimensions: disrespect for others, disrespect for democratic values, and disrespect for democratic institutions. Table 1, adapted from Stryker et al. (2024), summarises previous multidimensional treatments of incivility having disrespect at their core. The conceptual definition of incivility is shown in the first column, and the others show the analytic dimensions included within the political incivility construct, reflecting a core of disrespect towards interlocutors, democratic values, and/or democratic institutions. As illustrated in Table 1, Bormann et al. (2022) trace incivility to five norms. The information norm refers to information that is relevant, high quality and sufficient to enable cooperation; the modality norm involves assuring mutual comprehension through clear, concise and orderly speech; the process norm concerns ensuring connectivity between discursive contributions; the relation norm encompasses politeness, deference and appreciation to assure mutual respect; the context norm includes violating constitutional, liberal or democratic principles through behaviours such as excluding oppositional voices, slurs against minorities and threats of physical harm. Muddiman (2017, 2019) and Stryker et al. (2016) identify dimensions of political incivility for Americans, and Bentivegna and Rega (2022) do the same for Italians. The latter focused on incivility adopted by political actors, using confirmatory factor analysis (CFA) to empirically confirm that the concept has the following three inter-correlated dimensions. First, violation of politeness norms (disrespect for others), including empirical indicators such as shouting, interrupting, insulting, belittling, eye-rolling, teasing, and so on. Secondly, individual delegitimization, behaviours that deny democratic principles such as inclusion, equal rights, respect for different cultures, religions, and races; here the empirical indicators used were discriminatory stereotypes, racial slurs, excluding or misleading the other, and so on. Thirdly, disrespect for democratic institutions, which included indicators such as denigrating national symbols, encouraging harm, physical confrontation in an institutional arena, and so on. The CFA showed that not all indicators were significant in determining the dimension (i.e., ‘eye-rolling’ and ‘teasing’ recorded low values and thus were marginal in determining the dimension of impoliteness) (Bentivegna and Rega, 2022).

Table 1.

Political incivility: Definitions and dimensions of the concept.

Author(s)	Conceptual definitions of incivility	Disrespect for the others	Disrespect for democratic principles/values	Disrespect for democratic institutions
Bentivegna and Rega (2022)	‘As a lack of respect for the social and cultural norms that govern personal interaction as well as those governing the functioning of democratic systems’ (p. 4)	Impoliteness	Some aspects of individual delegitimation include intentional falsehoods and discriminatory stereotypes.	Institutional delegitimation, including denigrating national symbols and occupying Parliament.
Bormann et al. (2022)	‘Communication in public political debate that participants disapprove of as severely violating communication norms of information, modality, process, relation, or context’ (p. 16).	Information norm, modality norm, relation norm, process norm	Information norm, modality norm, process norm	Context norms
Hopp (2019)	Violations of speech-based and inclusion-based norms	Profane language, intentional disrespect, attribution of negative motives, etc.	Persuasive deception	Persuasive deception, excluding opponents, undermining faith in democracy, etc.
Muddiman (2017, 2019)	Violations of politeness (personal-level) and deliberation-based (public-level) norms.	Personal-level incivility	Deception aspect of persona-level incivility and public-level incivility.	Public-level incivility
Rossini (2022)	Uncivil discourse, in which people express themselves with foul language or harsh tone, is a rhetorical act (…)	Incivility is a matter of tone, not substance, of discourse (name-calling, vulgarities, etc.)	Some aspects of intolerant speech	Intolerant speech: discriminatory or exclusionary speech, limiting people's rights, etc.
Stryker et al. (2016)	Violations of politeness, ongoing deliberation, and truthfulness.	Insulting utterances, discursive	Deception; use of racial, ethnic or religious slurs; threatening/encouraging harm	Deception

Source: Our adaptation of Stryker et al. (2024).

Against this background, we adopt a definition of incivility encompassing all three interrelated dimensions, conceiving it as a lack of respect for the social and cultural norms that govern personal interactions as much as those that govern the functioning of democratic systems. In particular, the types of incivility linked to the three dimensions which we focus on were those found most significant in Bentivegna and Rega's study (2022): impoliteness dimension or disrespect for others (linked to ‘discursive’ and ‘vulgar’ incivility), disrespect for democratic values (linked to ‘informational’ and ‘discriminatory’ incivility), and disrespect for democracy (linked to ‘violent’ incivility), to capture citizens’ perceptions in relation to each type and subsequently identify their predictors.

Discursive incivility refers to a lack of communicative reciprocity (Bormann, 2022; Hopp, 2019; Stryker et al., 2016) and manifests through repeatedly interrupting the interlocutor, shouting to drown out their voice, and extreme forms of name-calling (‘nazi’, ‘communist’, ‘Taliban’, etc.) that eliminate any possibility of interaction. Already present in the literature, this aspect of incivility is traditionally detected in relation to the refusal to dialogue with others (Hopp, 2019; Stryker et al., 2016). Our choice to also include extreme forms of name-calling is based on the observation of how widespread this has become in public debate today.

Vulgar incivility refers to vulgar language (Coe et al., 2014; Kenski et al., 2019, 2020; Massaro and Stryker, 2012) used against individual opponents/parties. Regardless of whether this takes the form of a gesture (e.g. the middle finger) or a verbal externality, we intended to investigate the perception of vulgarities in reference to both the individual and collective spheres (within parliament or another institutional venue).

As far as informational incivility is concerned, we have included spreading false or inaccurate news to strengthen one's own position and participation in defamation campaigns against opponents. This is a highly corrosive type of incivility for democratic systems, whose empirical detection has frequently been traced to lying accusations, and misleading or deceptive activities (Hopp, 2019; Kenski et al., 2019, 2020; Stryker et al., 2016).

Another mode of alteration of the democratic game concerns violent incivility, which refers to extremely aggressive political behaviour, for example, the use of physical violence during a discussion in institutional venues or the threat to use force against people one disagrees with. This type of incivility has increasingly attracted the attention of researchers recently (Bormann, 2022) and it finds important parallels with ‘immoral–criminal incivility’ (Muddiman, 2021).

Finally, discriminatory incivility concerns the stigmatisation of subjects/groups (immigrants, LGBTQ+, religious minorities, etc.), denying them the right to speak, and the use of racist, sexist, or religious epithets (Lugosi-Schimpf and Thorlakson, 2021; Papacharissi, 2004; Rossini, 2022). Again, this type of incivility delegitimises basic principles of democracy such as pluralism, equal rights, and inclusiveness.

These five types of political incivility allow us to analyse how perceptions change across different kinds of uncivil discourse and measure the predictors of perception.

Hypotheses

Starting from the recognition that ‘in a democracy, regular citizens – not just political elites or scholars – are crucial arbiters of what constitutes incivility and whether there is too much of it in a public discourse’ (Kenski et al., 2020: 798), our study investigates citizens’ evaluation of uncivil behaviours by political actors. Since the five types of incivility selected for analysis, and their relative indicators (items/statements) were perceived as uncivil and/or extremely uncivil in previous studies (Bentivegna and Rega, 2022; Muddiman, 2017; Stryker et al., 2016), we first hypothesize that all 10 indicators of these five types are perceived as uncivil by the majority of respondents. However, we should also consider some possible context-specific diversifications. For example, referring to Muddiman’s (2017) study, it would appear that American citizens perceive personal-level incivility (e.g. name-calling, pejorative speech, etc.) more clearly than public-level forms (e.g. racial slurs, misleading, etc.). Despite using indicators that are not perfectly homogeneous, Kenski et al. (2020) found similar results, showing that people in the US perceive vulgarity and name-calling as more uncivil than aspersions and lying accusations. However, recent surveys from Europe suggest different results (Bormann, 2022). Our adoption of indicators referring to discursive incivility allows us to verify their robustness in a non-US context, at a time when the rise of populism has made bad manners so pervasive in Italian political communication as to be almost normalised. Indeed, the spread of a populist communicative style – first with the Lega party and then with the Five Star Movement – may have increased people's familiarity with personal attacks and the stigmatisation and demonisation of other individuals/groups (e.g. immigrants, homosexuals, etc.) to the point of desensitisation. The above considerations led to the following hypotheses:

H1a: Most respondents find many or all of the 10 items of incivility to be uncivil, or extremely so.

H1b: Discursive incivility is rated as less uncivil than the other types (based on its gradual spread in contemporary political communication).

In addition to perceptions of political incivility, our study concerns the individual characteristics of citizens associated with more or less clear perceptions of the phenomenon. In other words, if ‘incivility is in the eye of the beholder’ (Herbst, 2010), what are the characteristics of those who perceive incivility more clearly?

The literature on the subject is contradictory regarding socio-demographic characteristics. Women appear to be more sensitive than men to incivility, especially if it is attributable to impoliteness (Kenski et al., 2020), but less sensitive to information manipulation (Conway and Stryker, 2021). Another frequently investigated socio-demographic characteristic is age, which previous studies have shown to be inversely correlated with tolerance towards incivility. Older adults, growing up when uncivil content was less prevalent, perceive incivility more clearly, while the young, socialised from the beginning to digital platforms, perceive it less clearly (Fridkin and Kenney, 2019). Younger generations’ increasing use of social media (SM) – with its well-known traits of informality, directness, and accessibility to information from the most diverse sources – certainly facilitates habituation to expressions of incivility, as demonstrated by recent studies concerning more extreme forms, such as incitement to hatred (Gubitz, 2022). Finally, the education variable must be considered, to identify the contribution of citizens’ cultural level to their perception of incivility. This remains a poorly explored issue and, even considering areas other than political communication, very few studies show an impact of education on perception. However, there is some indicative evidence showing that students with highly educated parents are more sensitive to incivility than those whose parents are less well-educated (Aul, 2017). As the author speculates, the latter, being more exposed to harassing comments, perceive them as normal and acceptable. Consistent with this explanation, it has also been found that uncivil behaviour is more prevalent among undergraduate students than among those in graduate courses (Wahler and Badger, 2017). In relation to socio-demographic characteristics, we have therefore formulated the following hypothesis:

H2: Women, the elderly, and people with a higher level of education perceive political incivility more clearly.

After these socio-demographics, we consider news consumption habits that can contribute to explaining variations in incivility perception. In general, strong news consumers should have a greater interest in public and collective life. However, media discourse often contains extreme forms of incivility (Sobieraj and Berry, 2011), so a significant part of the exposure to incivility is through the consumption of news (print, TV or web) or TV programmes featuring politicians. This is especially the case with talk shows, which have not surprisingly been identified as predictors of the negative characterization of the ‘overall tone and level of civility in politics today’ (York, 2013: 117). Indeed, these shows have long been the focus of scholarly attention due to the numerous episodes of political incivility present (Berry and Sobieraj, 2013; Sobieraj and Berry, 2011), reported in political news coverage and shared/commented on SM. Based on these findings, we think that the consumption of information which often contains incivility can help create the conditions for that desensitisation process already recognised in studies on media violence and indirectly among consumers of uncivil media (Kenski et al., 2020). Individuals who use such sources tend not only to perceive incivility less clearly, but also to express themselves in a less civil manner (Gervais, 2014). In short, heavy news consumers may have more opportunities to encounter incivility, and their perception of it is thus weaker. Therefore, our third hypothesis:

H3: People with greater information consumption perceive political incivility less clearly.

The final investigated area concerns variables related to politics. The literature shows that the perception of incivility varies in relation to the partisanship of subjects. Not only do people tend to perceive incivility that comes from co-partisan subjects less strongly (Muddiman, 2017; Mutz, 2015), but some studies have shown differences in perceptions between people aligned to right or left-wing positions. Kenski et al. (2020) showed that conservatives perceive incivility, especially lying accusations, less than liberals, especially regarding lying accusations. This finding seems coherent with what has been said so far in relation to the desensitisation effect. If we consider the prevalence of uncivil content in pro-Republican news media (Berry and Sobieraj, 2013), as well as the more pronounced use of incivility by conservative candidates and leaders (Heseltine and Dorsey, 2022), it is plausible that voters of this persuasion perceive incivility as a regular practice of public debate.² In the same arena, we finally considered the frequency and quality of consumption of political content on SM. When using SM, people can encounter different types of political content (news, memes, parodies, and discussions), which can activate minimal or more engaging forms of involvement (e.g. simply consuming communicative content, or commenting on it). Here, rather than the declination of these activities as new forms of political participation, we want to detect habituation to incivility on the part of individuals who regularly interact with problematic/uncivil online content, which might lead to campaigns of disinformation, hatred, and defamation (Giglietto et al., 2020). Given this, we think that people using SM frequently for political activities are more likely to encounter uncivil content and, consequently, perceive it less clearly. Thus, the following hypotheses:

H4a: People positioning themselves on the left perceive political incivility more clearly.

H4b: People with a high level of engagement on SM for political activities tend not to recognise political incivility distinctly.

Data and methods

To answer these questions, we conducted a survey of a representative sample of the Italian population (1000 respondents) during the second week of January 2022. Alongside socio-demographic data and political positioning, respondents were asked to indicate their information sources and frequency of use, SM usage practices in relation to political topics and interest in politics. To measure citizens’ perception of political incivility, we operationalized the concept by taking it back to five main types, each of which was empirically linked to two items/statements describing utterances of politicians in different settings (10 uncivil items plus two civil items). Respondents were given descriptions of politicians’ statements and behaviour in news, TV programmes or online posts. For each statement, respondents were asked to rate the degree of civility/incivility using a scale of 1 to 5.

Table 2 shows the correspondence between the dimensions of the concept previously illustrated and the empirical indicators. The table represents an attempt to better clarify our approach (the conceptualization of political incivility, and its dimensions, types, and empirical indicators) to allow its replication in other countries. Only by doing so is it possible to generalize the Italian case and discuss the similarities and differences emerging from other contexts.

Table 2.

Correspondence between indicators and dimensions of the concept of incivility.

Empirical indicators	Types of incivility	Dimensions
1. Interrupting, shouting and/or speaking over others during a discussion between members of different parties, preventing the debate from taking place.	Discursive incivility	Disrespect for others
2. Publicly describing a political opponent as a ‘traitor to the homeland’, ‘Taliban’, ‘Nazi’, etc.	Discursive incivility	Disrespect for others
3. Using vulgar and insulting language to speak about a political opponent (‘that ass**** is still talking’).	Vulgar incivility	Disrespect for others
4. Ostentatiously showing the middle finger or using bad language to refer to the opposition (‘those sons of *******’) during a parliamentary debate in the Chamber.	Vulgar incivility	Disrespect for others
5. Referring to another politician with racist, sexist, religious, etc.. epithets.	Discriminatory incivility	Disrespect for democratic values
6. Publicly denying minorities/groups, such as immigrants, LGBT, Muslims, etc, the right to speak.	Discriminatory incivility	Disrespect for democratic values
7. Lying/inventing facts and/or distorting interpretations to attack political opponents (e.g., ‘all cancelled ballots were in our favour’).	Informational incivility	Disrespect for democratic values
8. Actively participating in a defamatory campaign – by means of slander and unsubstantiated accusations –against a political opponent (‘I'm telling you, that guy has weird sexual preferences’).	Informational incivility	Disrespect for democratic values
9. Physically attacking other politicians during a particularly heated debate in Parliament.	Violent incivility	Disrespect for democratic institutions
10. Physically threatening an interlocutor with whom one disagrees during a public debate.	Violent incivility	Disrespect for democratic institutions
11. Seeking an agreement in Parliament with other political forces for the swift approval of a bill.	Civil	Civil
12. Publicly disagreeing with what was claimed by a member of another party.	Civil	Civil

‘General incivility’ was calculated as the mean of the mean ratings of all five types. In addition to 10 statements containing elements of incivility, two in which no elements of incivility were present were included as controls. The 12 statements (Tables 2 and 3) are the result of a reworking of the list from a previous survey.

Table 3.

Items used to evaluate the different types of incivility (%).

How do you judge the behaviour of political actors described below?	Incivility types	Extremely civil	Somewhat civil	Neither civil nor uncivil	Uncivil	Extremely uncivil
Lying/inventing facts and/or distorting interpretations to attack political opponents (e.g. ‘all cancelled ballots were in our favour’).	Informational	3.5	8.1	14.2	22.7	51.5
Actively participating in a defamatory campaign –by means of slander and unsubstantiated accusations – against a political opponent (‘I'm telling you, that guy has weird sexual preferences’).	Informational	3	3.3	12.8	27.3	53.6
Interrupting, shouting and/or speaking over others during a discussion between members of different parties, preventing the debate from taking place.	Discursive	3.7	5.5	13.8	29.2	47.8
Publicly describing a political opponent as a ‘traitor to the homeland’, ‘Taliban’, ‘Nazi’, etc.	Discursive	3.1	5	15.6	32.3	44
Ostentatiously showing the middle finger or using bad language to refer to the opposition (‘those sons of *******’) during a parliamentary debate in the Chamber.	Vulgar	2.8	5.7	12	17.2	62.3
Using vulgar and insulting language to speak about a political opponent (‘that ass**** is still talking’).	Vulgar	3	4.6	11.7	28	52.7
Physically attacking other politicians during a particularly heated debate in Parliament.	Violent	2.9	5.8	12.2	9.9	69.2
Physically threatening an interlocutor with whom one disagrees during a public debate.	Violent	2,1	4.5	11.7	20.5	61.2
Referring to another politician with racist, sexist, religious, etc. epithets.	Discriminatory	2.8	5.3	12.8	18.2	60.9
Publicly denying minorities/groups, such as immigrants, LGBT, Muslims, etc., the right to speak.	Discriminatory	2.5	6.7	14.4	25.7	50.7
Seeking an agreement in Parliament with other political forces for the swift approval of a bill.	Not uncivil	23.3	31	30.8	10	4.9
Publicly disagreeing with what was claimed by a member of another party.	Not uncivil	18.1	17.2	36.1	17.2	11.4

To assess the congruence within each pair of items, Cronbach's alpha was calculated (Table 4).

Table 4.

Mean ratings for different types of incivility.

Type of incivility	α	M	SD
Informational incivility	0.77	4.18	0.96
Discursive incivility	0.74	4.10	0.94
Vulgar incivility	0.84	4.26	0.97
Violent incivility	0.86	4.35	0.97
Discriminatory incivility	0.86	4.22	0.99
General incivility	0.96	4.22	0.90
Civility (control statements)	0.74	2.65	0.90

To confirm the appropriateness of our choice to respect and reproduce the multidimensionality of the concept, we conducted a series of exploratory factor analyses (Table 5). First, we applied a separate factor analysis to each pair of items, which provided a one-factor solution that explained between 80% and 90% of the variance in the data, and saved the resulting factorial scores; then, a factor analysis was applied again to the five factors obtained, accounting for over 85% of the new variance. It is worth mentioning that where two-thirds, three-fourths, or nine-tenths of the variance is accounted for, the claim of a strong inference will be remarkably plausible and adequately founded (Di Franco and Marradi, 2013). As a further test, we extracted the general factorial structure from the whole basket of 10 items submitted for analysis (Table 6). In this case, the quota of variance reproduced by the first component was >70%.

Table 5.

Total variance explained by the different types of incivility (%): Two-stage principal component analysis (PCA).

Type of incivility	λ
Informational incivility^a	81.3
Discursive incivility^a	79.2
Vulgar incivility^a	86.3
Violent incivility^a	87.7
Discriminatory incivility^a	87.3
General incivility^b	86.0

Results of separate PCAs for each type of incivility (first stage).

Results of a new PCA on the extracted first components (second stage).

Table 6.

Component loadings and communalities reproduced by the dimension of ‘general incivility’: One-stage principal component analysis (PCA).^a

Items	r	h²
They intentionally spread false or inaccurate news to reinforce their political positions (e.g. ‘all canceled ballots were in our favor’).	0.78	0.68
They actively participate in a defamatory campaign – by means of slander and unsubstantiated accusations – against a political opponent (‘I'm telling you. that guy has weird sexual preferences’).	0.88	0.80
During a confrontation between members of different parties, they repeatedly interrupt, shout and/or speak over others, preventing the debate from taking place.	0.85	0.75
They publicly describe a political opponent as a ‘traitor to the homeland’, ‘Taliban’, ‘Nazi’, etc.	0.77	0.84
During a parliamentary debate in the Chamber, they ostentatiously show the middle finger or use bad language referring to the opposition forces (‘those sons of a *****’).	0.89	0.81
They publicly use vulgar and insulting language against other politicians (‘that ass**** is still talking’).	0.86	0.79
During a particularly heated debate in Parliament, they physically attack other politicians.	0.90	0.85
They physically threaten an interlocutor with whom they disagree during a public debate.	0.87	0.78
They refer to another politician with racist, sexist, religious, etc. epithets.	0.90	0.82
They publicly deny the right to speak to minorities/groups, such as immigrants, LGBT, Muslims, etc.	0.83	0.72

Results relative to the first component extracted.

The mean of the mean ratings for different types of incivility, which we defined as ‘general incivility’, was 4.22 (SD = 0.90). We found that people perceived violent incivility as the most uncivil form (4.35) and discursive incivility as the least (4.10). Already these initial data provide interesting insights into the different degrees of sensitivity expressed by respondents for each type of incivility. However, before commenting on this data, it is important to illustrate the variables used as predictors of respondents’ perception of political incivility. The demographic variables used were gender, age and education transformed into dummy-coded variables. Media information consumption, on the other hand, was structured into five items that recorded the frequency of reading newspapers (both print and digital), exposure to TV newscasts and radio and TV talk shows, reading news online only, and acquiring information through SM. Interest in politics was recorded on a scale from 0 to 10. Turning our attention to the final variable, the SM experience was investigated by means of five items that recorded the frequency over the past week (from ‘never’ to ‘as often as I log on’) of participation in online discussions, posting political content, comments or reactions to political posts (we then categorized this variable according to the distinction between ‘low engagement’ and ‘high engagement’ on SM for political activities). To test the validity of our hypotheses regarding the variables predicting greater/lesser sensitivity to political incivility, six regression models were constructed.

Findings

The first element of interest is that most respondents rated as uncivil or extremely uncivil all the items provided to empirically measure incivility (confirming H1a). Furthermore, while the average score for general incivility is 4.22, among types of incivility discursive scores lowest (4.10), followed by informational (4.18), discriminatory (4.22), vulgar (4.26), and violent (4.35). Although the differences are only slight, we can assume that H1b is also confirmed. We will return to this aspect in our discussion because this data can be better understood when linked to the Italian context.

Concerning individual characteristics associated with the perception of ‘general incivility’, regression models were tested. First, a model was constructed for incivility as a whole, then five others were constructed, one for each type of incivility. The results of the regression models are reported in Tables 7 and 8.

Table 7.

Results of the regression model predicting the perception of ‘general incivility’.

	Measure of perceived incivility
Predictors	β
Sex: female	0.2
Age: 18–24	−0.4
Age >55	0.10***
Education: low	−0.10***
Newspaper (print and website)	−0.10****
TV newscast	0.07*
Social Media	0.01
Online News	0.02
Political TV–radio talk show	−0.09***
Interest in politics: 0–10	0.05
Political collocation: no collocation	0.01
Political collocation: right	0.03
Political collocation: left	0.12*
Low engagement on social media for political activities	0.09**
High engagement on social media for political activities	−0.46****
R	0.46
R ²	0.21
R² adapted	0.20
Standard error of the estimate	0.82

*p < 0.10; **p < 0.05; ***p < 0.01; ****p < 0.001.

Table 8.

Results of the regression model predicting perceptions of the different types of incivility.

	Informational incivility	Discursive incivility	Vulgar incivility	Violent incivility	Discriminatory incivility
Predictors	β	β	β	β	β
Sex: female	−0.03	0.06	0.06	0.00	0.01
Age: 18–24	−0.02	−0.06	−0.03	−0.05	−0.03
Age: >55	0.12****	0.08**	0.11***	0.10***	0.05
Education: low	−0.10****	−0.10***	−0.10***	−0.08*	−0.10**
Newspaper (print and website)	−0.07*	−0.09***	−0.10****	−0.09***	−0.08**
TV news	0.06	0.05	0.08*	0.07*	0.07*
Social Media	0.02	0.01	0.02	−0.01	0.01
Online News	0.04	0.02	0.00	0.03	0.00
Political TV–radio talk show	−0.07*	−0.07	−0.08*	−0.08*	−0.11****
Interest in politics: 0–10	0.04	0.05	0.04	0.04	0.06
Political Collocation: no collocation	0.06	−0.03	−0.01	0.02	0.03
Political collocation: right	0.10	−0.02	−0.06	0.07	0.03
Political Collocation: left	0.15**	−0.04	0.08	0.14**	0.14***
Low engagement on social media for political activities	0.09*	0.07	0.06	0.11**	0.11**
High engagement on social media for political activities	−0.46****	−0.42****	−0.41****	−0.44****	−0.42****
R	0.44	0.42	0.43	0.44	0.42
R ²	0.19	0.18	0.19	0.19	0.18
R² adapted	0.18	0.16	0.17	0.18	0.17
Standard error of the estimate	0.86	0.85	0.880	0.87	0.90

*p < 0.10; **p < 0.05; ***p < 0.01; ****p < 0.001.

Looking at Table 7, some results immediately stand out. Firstly, contrary to findings in other studies, the gender variable does not influence the perception of general incivility. Regarding the age variable, however, being older is a predictor of a greater perception of incivility. Finally, a lower educational level is a predictor of a lower perception of incivility. Overall, H2 is partially confirmed. Concerning news consumption, our data supports H3, which held that higher consumption was a predictor of a lower perception of incivility. This applies, in the first place, to newspapers and TV/radio talk shows. Consumption of television newscasts, on the other hand, bucks the trend (albeit with a low level of significance), confirming the superficial and distracted consumption traditionally attributed to the users of such news. In short, the intentional news consumption typical of reading a newspaper (printed or online) and/or exposure to a political talk show offers so many opportunities for contact with uncivil episodes (or narratives of these) that a sort of desensitisation results.

In terms of politics-related variables, self-positioning on the left is a predictor of a clearer perception of incivility (H4a is supported), while high engagement on SM for political activities emerges as the clearest predictor of a lower perception of incivility (H4b is fully supported).

The results of the regression models for each type of incivility (Table 8) confirm the centrality of the previously identified predictors, while introducing some interesting nuances. The most significant predictor of incivility perception – in a negative sense – is high engagement on SM for political activities: those most active around politics on SM tend not to perceive incivility clearly. This is particularly evident in relation to informational and violent incivility, which are both widespread on SM (for misleading content see Hindman et al., 2022 and Valenzuela et al., 2019; for violent/aggressive content, see Phillips and Milner, 2017). In addition to SM engagement, education, older age and newspaper reading are also confirmed as predictors of altered perception. It is interesting to note that respondents’ self-placement on the left politically is a predictor of a greater perception of informational, violent and discriminatory incivility. These are types of incivility that, to varying degrees, violate the norms that govern the functioning of democratic institutions and contribute to the common ground for civil coexistence. As far as discriminatory incivility is concerned, exposure to talk shows is a predictor that dampens the perception of the seriousness of the phenomenon. Lastly, it is noteworthy that discursive incivility – which can be traced back to the dimension of impoliteness – is the least clearly perceived type, and also sees among its predictors only older age, a low level of education, reading newspapers and engagement in SM. While older age activates a greater perception of discursive incivility, a low level of education, newspaper reading and SM use for political activities go in the opposite direction, predicting lower perception.

To provide better insight into the size of group differences by type of incivility, we repeated regressions using average marginal effects (Figure 1(a) to (f)). In line with previous studies (Conway and Stryker, 2021; Kenski et al., 2020), women appear to perceive discursive and vulgar incivility more clearly than men, but this effect is absent for the other types. Regarding news consumption, the deviances between models are less significant. Although SM users tend to perceive violent incivility less clearly, people who read news online are only slightly less likely to perceive discriminatory incivility. Finally, people self-aligned to right-wing positions tend not to recognise discursive and vulgar incivility, while those self-aligned on the left perceive informational, violent, and discriminatory incivility more clearly.

Figure 1.

(a) Average marginal effects for the regression model predicting the perception of ‘general incivility’; (b) average marginal effects for the regression model predicting the perception of ‘informational incivility’; (c) average marginal effects for the regression model predicting the perception of ‘discursive incivility’; (d) average marginal effects for the regression model predicting the perception of ‘vulgar incivility’; (e) average marginal effects for the regression model predicting the perception of ‘violent incivility’; (f) average marginal effects for the regression model predicting the perception of ‘discriminatory incivility’.

Discussion and conclusion

Despite concern among scholars, political actors and journalists regarding the spread of political incivility in contemporary democracies, there is still no unambiguous definition of the concept. The nuances that accompany the definition of incivility are such that the task of clearly identifying its consequences for the relationship between citizens and politics is complex (Van’t Riet and Van Stekelenburg, 2022). However, this does not preclude operational definitions of incivility that allow its translation into empirical terms. In this study, we have traced the main dimensions of political incivility proposed in the literature to five types, which encompass behaviours showing disrespect for others, for democratic principles/values and for democratic institutions. This structuring of the concept into different types allowed us to record its perception by citizens. In this regard, the respondents’ low perception of discursive incivility should be noted, as it highlights the relevance of the context in which the study took place. In Italy, where populism has established itself at an institutional level, communicative practices based on the denial of respect for others, be they political opponents, or government or media institutions, are so widespread they are not perceived as clearly negative by citizens. The mechanism at work is similar to that described by Muddiman et al. (2022: 276), who argue: ‘it is possible that, because a person believes politicians use name-calling frequently in campaigns, the person will believe that name-calling is not descriptive norm violating’. Therefore, we are no longer scandalized when political subjects shout at, interrupt or insult each other.

Furthermore, the increased sensitivity to violent incivility signals how a progressive shift towards an interpretation of the phenomenon as ‘immoral/criminal incivility’, as defined by Muddiman (2021), is taking place. This is a process already noted in a previous study in the US (Muddiman, 2019), which recorded the definition of violent incivility given freely by the respondents. The indication from our research – done in a non-US context – confirms the public's sensitivity to episodes of political violence. It is worth emphasising that in both countries there have been episodes – albeit of varying severity – in which political incivility has resulted in calls for violence against political institutions and trade unions that led to heated and prolonged public debates, and legal proceedings. It cannot be excluded that these episodes triggered a greater perception of violent incivility by citizens.

Turning now to the predictors of incivility perception, a lack of homogeneity in the individual characteristics of the respondents emerged. Firstly, among socio-demographic variables, gender seems to be less relevant in determining citizens’ perception of incivility compared to US studies (Kenski et al., 2020; Stryker et al., 2016). However, as in the US, here too it emerged that women perceive specific types of incivility linked to the disrespect for others dimension (vulgar and discursive incivility) more clearly. Mature age was found to be a predictor of a greater perception of incivility, while a low education level was associated with a lower perception of the phenomenon. In interpreting the differences concerning demographic variables that emerge with respect to previous research, one must consider not only the cultural and social contexts of the studies, but also the characteristics of the samples. We used 1000 individuals representative of the Italian population, while other studies employed smaller samples, sometimes from a specific part of the population (e.g. university students, as in Stryker et al., 2016). This could explain the different evaluations of incivility found and highlight the need to further enrich the study of predictors from a cross-country comparative perspective.

In the area of information consumption, both the reading of newspapers (in print and/or digital format) and exposure to radio and television talk shows act to decrease the perception of incivility. In contrast, the consumption of television newscasts seems to help individuals perceive the presence of the phenomenon, particularly the vulgar, violent and discriminatory types. A possible explanation is the short, fast format of television newscasts combined with their equally rapid consumption, aimed at a concise update on the ‘latest facts’. On the contrary, reading newspapers and talk show exposure is associated with stronger interest and more available time. Reading detailed reconstructions of uncivil episodes featuring politicians produces desensitisation towards the phenomenon. Long studied by scholars, the shift from news to talk (Meltzer, 2019) brings an increased circulation of incivility in the information supply (Coe et al., 2014; Mutz, 2015; Sobieraj and Berry, 2011; York, 2013). The strategic use of incivility made by journalists and media companies in recent decades, aimed at capturing audience shares in a competitive market (Webster, 2014) by presenting emotionally engaging narratives of politics marked by dramatisation and conflict, is therefore an important contextual element. Thus, one cannot ignore the roles of the journalistic universe and media companies (Berry and Sobieraj, 2013; Mutz, 2015) in creating a climate in which witnessing incivility ‘can be a form of entertainment, as good as many extreme sports’ (Mutz, 2012: 79).

Finally, among variables related to politics, respondents’ self-positioning on the left tends to heighten their perception of incivility, confirming the relevance of left-wing versus right-wing partisanship (Kenski et al., 2020). In this area, however, the most significant predictor of low incivility perception is the intense use of SM for the consumption/production of politically related content. In this regard, politics seems to play an almost ancillary role – political interest emerges as a marginal variable – while it is SM use itself, functioning like a powerful ‘anaesthetic’, which reduces the perception of both general incivility and the various types identified. In other words, the engagement does not seem to be political, but related to digital folklore repertoires (De Seta, 2020) in which internet culture is intertwined with trivial content and meme production, trolling, and so on. This is not unexpected, given the established literature on the circulation of incivility on SM (Anderson and Huntington, 2017; Oz et al., 2018; Phillips and Milner, 2017). However, it is worth noting the clearly demonstrated positive correlation to a low perception of political incivility linking SM engagement and news and talk show consumption, which represents another indicator of the progressive hybridization of the media system (Chadwick, 2017). This contributes, albeit indirectly, to pushing back on attributing to SM a preponderant role in the spread of incivility, showing that its circulation in the public debate can be read as the result of the activation of different actors and the dynamics of message circulation that are triggered within an interconnected ecosystem between digital platforms and traditional media.

Overall, the data confirm the need to disentangle the concept of incivility into different types and record citizens’ evaluations of each. As has emerged from the most recent debate in this field (Masullo Chen et al., 2019), continuing to speak of ‘incivility’ in a broad sense is unhelpful when seeking to clarify its nature, identify its consequences on citizens’ relationships with politics (in terms of trust and participation) and, above all, plan measures to counteract its spread.

Likewise, the data confirm that incivility perception is not uniform, and the characteristics of the beholder must be considered. With this study, we wanted to contribute in this direction by taking the empirical approach beyond the US, where most research has been done. The divergences that emerged in relation to the assessment of the types of incivility and the different roles of the predictors can be attributed both to the specificities of the Italian context and the time when the surveys were conducted (up to eight years ago). Clearly, the desensitisation process can be the result of particular political, social and cultural conditions that may significantly change perception levels over even a limited time span.

One of the main limitations of our work is the failure to differentiate the media environments in which incivility occurs. As previous research has shown, the channel and structure of the media platform and the roles of the actors involved influence subjects’ perceptions (Bormann, 2022; Gubitz, 2022). Nevertheless, we believe that this study offers an operationalization approach to incivility that can be replicated in other countries, therefore highlighting the importance of enriching the literature on the predictors of the phenomenon in comparative contexts, to better identify the relevance of the single variables.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Sara Bentivegna

Notes

References

Anderson

Huntington

(2017) Social media, science, and attack discourse: How Twitter discussions of climate change use sarcasm and incivility. Science Communication 39(5): 598–620.

Aul

(2017) Who’s uncivil to who? Perceptions of incivility in pre-licensure nursing programs. Nurse Education in Practice 27: 36–44.

Bentivegna

Rega

(2022) Searching for the Dimensions of Today’s Political Incivility. Social Media + Society 8(3): 1–12. DOI: https://doi.org/10.1177/20563051221114430.

Berry

Sobieraj

(2013) The Outrage Industry: Political Opinion Media and the New Incivility. Oxford: Oxford University Press.

Bormann

(2022) Perceptions and evaluations of incivility in public online discussions – insights from focus groups with different online actors. Frontiers in Political Science 4: 1–17.

Bormann

Tranow

Vowe

, et al. (2022) Incivility as a violation of communication norms: A typology based on normative expectations toward political communication. Communication Theory 32(3): 332–362.

Chadwick

(2017) The Hybrid Media System: Politics and Power. Oxford: Oxford University Press.

Coe

Kenski

Rains

(2014) Online and uncivil? Patterns and determinants of incivility in newspaper website comments. Journal of Communication 64(4): 658–679.

Conway

Stryker

(2021) Does a speaker’s (in) formal role in news media shape perceptions of political incivility? Journal of Broadcasting and Electronic Media 65(1): 24–45.

10.

De Seta

(2020) Digital folklore. In: Hunsinger

Allen

Klastrup

(eds) Second International Handbook of Internet Research. Dordrecht and Heidelberg: Springer, 167–183.

11.

Di Franco

Marradi

(2013) Factor Analysis and Principal Component Analysis. Milan: Franco Angeli.

12.

Fridkin

Kenney

(2019) Taking Aim at Attack Advertising: Understanding the Impact of Negative Campaigning in U.S. Senate Races. Oxford: Oxford University Press.

13.

Gervais

(2014) Following the news? Reception of uncivil partisan media and the use of incivility in political expression. Political Communication 31(4): 564–583.

14.

Gervais

(2017) More than mimicry? The role of anger in uncivil reactions to elite political incivility. International Journal of Public Opinion Research 29(3): 384–405.

15.

Giglietto

Righetti

Rossi

, et al. (2020) It takes a village to manipulate the media: Coordinated link sharing behavior during 2018 and 2019 Italian elections. Information, Communication and Society 23(6): 867–891.

16.

Gubitz

(2022) Race, gender, and the politics of incivility: How identity moderates perceptions of uncivil discourse. The Journal of Race, Ethnicity, and Politics 7(3): 526–543.

17.

Herbst

(2010) Rude Democracy: Civility and Incivility in American Politics. Philadelphia: Temple University Press.

18.

Heseltine

Dorsey

(2022) Online incivility in the 2020 congressional elections. Political Research Quarterly 75(2): 512–526.

19.

Hindman

Lubin

Davis

(2022, February 10) Facebook has a superuser-supremacy problem. The Atlantic. https://www.theatlantic.com/technology/archive/2022/02/facebook-hate-speech-misinformation-superusers/621617/

20.

Hopp

(2019) A network analysis of political incivility dimensions. Communication and the Public 4(3): 204–223.

21.

Kenski

Coe

Rains

(2019) Perceptions of incivility in public discourse. In: Boatright

Shaffer

Sobieraj

Young

(eds) A Crisis of Civility?: Political Discourse and its Discontents. New York: Routledge, 45–60.

22.

Kenski

Coe

Rains

(2020) Perceptions of uncivil discourse online: An examination of types and predictors. Communication Research 47(6): 795–814.

23.

Laden

(2019) Two concepts of civility. In: Boatright

Shaffer

Sobieraj

Young

(eds) A Crisis of Civility?: Political Discourse and its Discontents. New York: Routledge, 9–30.

24.

Lugosi-Schimpf

Thorlakson

(2021) Words that hurt democracy: ‘The sticks and stones of anti-democratic discourse in Hungary and Germany’. In: Walter

(ed.) Political Incivility in the Parliamentary, Electoral and Media Arena. London: Routledge, 35–55.

25.

Massaro

Stryker

(2012) Freedom of speech, liberal democracy, and emerging evidence on civility and effective democratic engagement. Arizona Law Review 54: 375–441.

26.

Masullo Chen

Muddiman

Wilner

, et al. (2019) We should not get rid of incivility online. Social Media+Society 5(3): 1–5.

27.

Meltzer

(2019) From News to Talk: The Expansion of Opinion and Commentary in US Journalism. Albany: Suny Press.

28.

Muddiman

(2017) Personal and public levels of political incivility. International Journal of Communication 11: 3182–3202.

29.

Muddiman

(2019) How people perceive political incivility. In: Boatright

Shaffer

Sobieraj

Young

(eds) A Crisis of Civility?: Political Discourse and its Discontents. New York: Routledge, 31–44.

30.

Muddiman

(2021) Conservatives and incivility. In: Jarvis

(ed.) Conservative Political Communication. New York: Routledge, 119–136.

31.

Muddiman

Flores

Boyce

(2022) Descriptive and injunctive incivility norms in political campaigns: Differences across behavior type, candidate gender, and candidate party position. American Behavioral Scientist 66(3): 274–291.

32.

Mutz

(2012) Television and uncivil political discourse. In: Shea

Fiorina

(eds) Can We Talk?: The Rise of Rude, Nasty, Stubborn Politics. New York: Pearson Higher Education, 67–91.

33.

Mutz

(2015) In-Your-Face Politics. Princeton: Princeton University Press.

34.

Zheng

Chen

(2018) Twitter versus Facebook: Comparing incivility, impoliteness, and deliberative attributes. New Media and Society 20(9): 3400–3419.

35.

Papacharissi

(2004) Democracy online: Civility, politeness, and the democratic potential of online political discussion groups. New Media and Society 6(2): 259–283.

36.

Phillips

Milner

(2017) The Ambivalent Internet: Mischief, Oddity, and Antagonism Online. Cambridge Malden: Polity.

37.

Rossini

(2022) Beyond incivility: Understanding patterns of uncivil and intolerant discourse in online political talk. Communication Research 49(3): 399–425.

38.

Sobieraj

Berry

(2011) From incivility to outrage: Political discourse in blogs, talk radio, and cable news. Political Communication 28(1): 19–41.

39.

Strachan

Wolf

(2012) Political civility: Introduction to political civility. PS: Political Science & Politics 45(3): 401–404.

40.

Stryker

Conway

Bauldry

, et al. (2024) Emotional Markers of Disrespect: A Fourth Dimension of Perceived Political Incivility? Communication Research. DOI: 10.1177/00936502231221926.

41.

Stryker

Conway

Danielson

(2016) What is political incivility? Communication Monographs 83(4): 535–556.

42.

Valenzuela

Halpern

Katz

, et al. (2019) The paradox of participation versus misinformation: Social media, political engagement, and the spread of misinformation. Digital Journalism 7(6): 802–823.

43.

Van’t Riet

Van Stekelenburg

(2022) The effects of political incivility on political trust and political participation: A meta-analysis of experimental research. Human Communication Research 48(2): 203–229.

44.

Wahler

Badger

(2017) Exploring faculty perceptions of student incivility in social work: Results from a national survey. Advances in Social Work 17(2): 340–354.

45.

Walter

(2021) Introduction: The study of political incivility and its challenges. In: Walter

(ed.) Political Incivility in the Parliamentary, Electoral and Media Arena. London: Routledge, 1–16.

46.

Weber Shandwick (2019, June 26) ‘Civility in America 2019: Solutions for Tomorrow’. Report.

47.

Webster

(2014) The Marketplace of Attention: How Audiences Take Shape in a Digital Age. Cambridge, MA: MIT Press.

48.

York

(2013) Cultivating political incivility: Cable news, network news, and public perceptions. Electronic News 7(3): 107–125.