Sage Journals: Discover world-class research

Abstract

The spiral of silence theory suggests that those who hold a contrary opinion to the majority tend to remain silent; however, social media may be flipping this upside down. The majority may remain silent to the hardcore vocal minority, implying a self-censoring majority. This study examines the spiral of silence theory, and the effect of political identity and ideology has on the willingness to self-censor and the fear of isolation with social media use. Through a survey distributed through a Qualtrics panel to participants that mirror the US census (N = 1,047), this study investigates political identity and ideology by examining self-censorship and fear of isolation of social media users. Results identify significant differences with self-censorship and identity politics between strong conservatives and liberals in comparison to the majority in the middle. Differing political ideologies seem to have an impact on fear of isolation between democrats and conservatives.

Keywords

spiral of silence fear of isolation self-censorship political ideology social media

In recent years, the polarization of topics has caused division not only in political discussions but also in public opinion. With the sharing of public opinion on social networking sites and the increasing view of polarized topics, there can be voices and opinions that are lost. In 2020, online opinions ranged in many different directions sharing different political outcomes and views. On social media, topics varied from masks to anti-masks, from Black Lives Matter to all lives matter, from Donald Trump to Joe Biden. These topics became rapidly politicized as they trended on Twitter and gained followings from highly outspoken people and social media influencers (Yarchi et al., 2021). However, a poll in 2020 suggested that 64% of Americans do not use social media sites like Facebook, Twitter, and Instagram to post a picture in support of a cause, and 65% of Americans do not encourage others to post for cause or issue that they deem important. Similarly, 82% report not using a hashtag related to a political or social issue on social media (Auxier, 2020). Data indicate that there is a majority of people staying silent despite new and trending topics.

The spiral of silence theory suggests that there will always be those that will silence their opinions when there is an opposing opinion. People who hide their opinions think they are in the minority because of loud vocal opinions that are seen as the perceived majority (Noelle-Neumann, 1974). However, there is also the hardcore vocal minority that will always speak out and are not afraid to tell their opinions (Noelle-Neumann, 1991). There has been plenty of research and different studies that have investigated how the spiral of silence is affecting public opinions both offline and online and the perception of opinions of the majority, the minority, and hardcore opinions (Chan, 2018; Gearhart & Zhang, 2018; Liu & Fahmy, 2011). Recent studies have viewed that the hardcore vocal minority is more willing to speak out and express unpopular opinions with different topics of climate change and racism (Chaudhry & Gruzd, 2020; Porten-Cheé & Eilders, 2015). However, none of these studies explored broadly whether individuals are self-censoring or isolating on social media through the political polarization of topics and their political identity.

With the polarization of topics, the following study will re-examine the spiral of silence theory and the perception of politicized public opinions and political identity on social media to see if the theory is changing or flipping upside down and that the hardcore vocal minority is more willing to speak out than the majority. This study seeks to understand if individuals are self-censoring online based upon their political identity and ideology.

Literature Review

Spiral of Silence

For decades, scholars of mass communication have understood that those who silence their opinions tend to be those who perceive themselves as holding a minority viewpoint. This understanding comes from the spiral of silence theory (Noelle-Neumann, 1974), which explains how individuals who think their opinion differs from the majority tend to self-censor in times of controversial opinion climates. The information about opinion climates to inform the public comes from interpersonal communication and communication sources such as news and social media sites. Previous research has suggested that those in the majority are more likely to speak their opinions in public (Noelle-Neumann 1991, 1993; Sohn, 2022). With the hardcore vocal minority opinions, it is expected that they will remain silent due to fear of isolation (FOI) or other factors. As the majority opinion becomes more and more pronounced, the minority opinion will lessen and the spiraling effect will begin (Gearhart & Zhang, 2018; Noelle-Neumann, 1991).

Hardcore Vocal Minority Opinions

There have been several studies that evaluate the spiraling effect and the differences between minority and majority opinions (Matthes et al., 2018; Noelle-Neumann, 1974). Matthes (2015) found that the spiral of silence can only exist if people believe that they are in the minority viewpoint. With social media, the silencing effect of the spiral of silence can be less severe or dominant because of the variety of opinions online (Poulakidakos et al., 2018). However, other researchers found that spiral of silence has been confirmed by people who strongly support an issue and find more people agreeing with their opinion than those who weakly support an issue (Krishna & Kim, 2020). Also, Hampton et al. (2017) found that exposure to a specific issue online is associated with their willingness to share that opinion. Minority voices are more likely to interact on social networking sites through indirect measures such as reactions with emojis rather than actively engage and comment on social media (Ordoñez & Nekmat, 2019).

Among the differing opinions, there is the hardcore minority opinions that have a more willingness to speak out than those in the majority, and their voices will not be silenced (Matthes et al., 2018; Noelle-Neumann, 1991; Porten-Cheé & Eilders, 2015). This group is created by having the defensive position on public opinion and becomes overpowered by its convictions and can sway others about their opinions of the majority (Noelle-Neumann, 1974, 1993). The hardcore minority will speak up if they find that the minority opinion is emerging even when it goes against an incongruent climate for public opinion making them a vocal hardcore minority (Gearhart & Zhang, 2018). According to Noelle-Neumann (1993), the hardcore minority is put into a defensive position for their opinions and beliefs and is no longer subject to FOI. These vocal minorities are more comfortable in sharing and expressing unpopular views of racism and climate change (Chaudhry & Gruzd, 2020; Porten-Cheé & Eilders, 2015). It has also been found that minority voices are more likely to speak out about their views such as anti-vaxxers raising their opinion and support against the issue, and those in the majority or in support of vaccination do not voice any opinion or speak out about support on the topic (McKeever et al., 2016).

Fear of Isolation and Willingness to Self-Censor

The theory of the spiral of silence has been tested in a variety of ways, and different factors can stimulate the reasons behind silencing motivators to create the spiral of silence. Factors that have contributed to the silencing effect are the FOI or the fear of social isolation and a person’s WTSC. The FOI occurs when a person is afraid to share their opinions because they will be isolated and rejected from their fellow peers and publics (Chan, 2018; Hayes et al., 2013; Noelle-Neumann, 1991). The WTSC is defined as hiding or withholding one’s opinion when it is perceived that the audience will disagree with the opinion. It is attributed to why people choose not to speak out their own opinions and focuses on the lack of opinion and why that opinion may be kept hidden or censored (Hayes et al., 2005a; Kwon et al., 2015). The WTSC can be an explanation or an effect for FOI which can be attributed to people’s need to align opinions and find opinion congruency due to public self-awareness and public expectations (Marder et al., 2016).

FOI and WTSC can be manifested in people’s behavior and how public opinions can shape what people share in public. Research has found that when FOI was already present, that there was greater WTSC (Fox & Holt, 2018). FOI can be present in one person which could isolate the minority view because people may stay silent out of fear of another person’s reaction rather than the issue itself (Matthes et al., 2018). From this, the spiral of silence is more likely to occur in local and small communities rather than large global populations (Sohn, 2022).

Other studies have looked at how the spiral of silence, FOI, and WTSC affect public opinions online. Chan (2018) found that FOI had an indirect negative effect on self-censorship in both an online and offline communication setting. They found in-person interactions that have a high level of disagreements are less likely to show support over communication that shares the same view. Beyond online and offline communication settings, FOI can also negatively impact anonymous and non-anonymous settings on social media (Kushin et al., 2019). However, the expectation of being personally attacked online allows those individuals to be hesitant to share opinions online, but not offline where in-person civility can be found (Neubaum & Krämer, 2018; Powers et al., 2019). The WTSC can constrain and enhance limiting behavior; this behavior can be displayed not only by the lack of communication online but also in the form of deleting pictures and posts. This form of self-censorship can limit expression for an opposing viewpoint, but not necessarily a supporting opinion (Chen, 2018). Research has also found that on social media, users are reluctant to publicly comment, however they are actively engaged in passively observing online discussion (Gearhart & Zhang, 2014).

Beyond the factors of WTSC and FOI, media opinions on social media can lead to individual’s social preferences where people rely more on their social circles than mass media (Scheufele et al., 2001; Sohn, 2022). Other research on the spiral of silence and social media found that political attitudes and information is reinforced through news reinforcement exposure (Song & Boomgaarden, 2017). Yet, other influences of social media public opinions are spread through information consumption by algorithms and echo chambers online (Kitchens et al., 2020). While individual perceptions will not always lie with mass media information (Noelle-Neumann, 1993), people can be exposed to congruent perceptions and attitude opinions that will allow them to underestimate the general population opinions (Wojcieszak & Rojas, 2011). With the content of social media sites such as Facebook being governed by users and algorithms (Pariser, 2011), users may perceive opinion climates to be diverse as well as preferences allow users to block different points of view based upon hostility (Fox & Warber, 2015), which can possibly affect social media climates (Fox & Holt, 2018).

Reasons for self-censoring comes from threats to perception of self-identity online and the act of receiving negative responses that might cause online friendship to dissolve with being unfriended or unfollowed (Powers et al., 2019). Social networking sites have a negative relationship with communication in offline settings; this reduces users of social media to engage with political conversations in offline settings (Hampton et al., 2017). However, exposure to diverse opinions has been shown to negate the effects of WTSC and allows people to post more politically online (Kwon et al., 2015).

Political Climates

Differences in opinions and the sharing of opinions can be dependent on the political climate and opposing views. Individuals share more of their political opinions for partisan media, especially if they perceive the support to be there in their public circles (Wang et al., 2017). Within political communication, the formation of public opinion is developed by quality of shared opinion and information as well as responsiveness within a democratic system (Druckman, 2014). Political interests and opinions on social media can increase an individual’s willingness to speak out (Hoffmann & Lutz, 2017). Other research suggests that diverse political communication on social media does not always yield more or enhance political discussion (Kwon et al., 2015).

Opinion climates and perceptions for political communication on social media may foster and promote political polarization (Tsfati et al., 2014). Tsfati and others found that one of the strongest forces to predicting political climate was one’s political predisposition of liberals or conservatives and the exposure of that type of polarized media. The existence of opposing ideologies does not specifically create the polarized climate, but the presence of a single ideology can create extreme opinions (Kashima et al., 2021). These extreme opinions suggest that the margins are becoming more vocal, and that partisanship in political communication influences value support at these margins rather than those who are in the majority opinion depending on values of either the Democratic or Republican party (Goren et al., 2009).

With partisan political communication, opinions are shared through political identity and/or ideology (Tappan, 2005). Political identity is usually self-proclaimed and shared through voter means of political party such as Republican and Democrat. Whereas political ideologies are shared upon platforms and beliefs in how the government should be controlled or should act (Connolly, 2002). Along with social polarization, audiences move to like-minded sources of information within political preferences continuing the spiral of silence that is accepted by their political ideologies (Dvir-Gvirsman, 2017). With opposing ideologies and political communication, further research is needed to explore how political communication and polarization are affecting the spiral of silence on social media, and how political margins might be silencing the majority.

Current Study

From the different ways, the spiral of silence, FOI, and self-censorship have been studied, this study seeks to identify self-censorship and FOI on social media within a political context. The hardcore minority is viewed as vocal through their engagement online and is on the edges of the political spectrum considered extremely conservative or solid liberals with the majority considered more in the middle for political beliefs. The vocal hardcore minority openly expresses opinions on divided social issues, this study seeks to understand if the majority is self-censoring because of political margins and minorities. Based off the literature this study hypothesizes the following:

H1. The majority in the middle are more likely to self-censor on social media than the hardcore minority.

H2. The majority in the middle are more likely to report higher levels of FOI on social media than the hardcore minority.

Method

Population

With permission from the authors’ institutional review board, data were collected from 1,047 participants who took a multiple section self-report survey. This study used a Qualtrics nationwide panel to sample a population that mirrored the US census. The distribution of the survey allowed for a stratified random sample, allowing for a diverse population with a focus on an equal distribution of people who identify as Republican or Democrat. Demographical information of age and race were asked. Other information about news consumption and how often participants used social media were used for demographic information. All participants consumed some type of news online and reacted on social media.

Participants

The sample had participants varied in range for age with samples from age groups consisting of 18- to 24-year-olds (6.7%), 25- to 34-year-olds (22.7%), 35- to 44-year-olds (19.3%), 45- to 54-year-olds (13.7%), and 55-year-old and above (37.3%). The participants were predominantly White (84.2%) with other participants identifying as Black (8.3%), Hispanic (5.9%), Asian (5.1%), Native American (1.8%), and other (1.8%). The sample was fairly split and representative of US political parties of Democrat (45%), Republican (44.6%), and Independent (10%). Participants were also asked to self-report where they stand politically as strong conservatives (19.9%), moderate conservative (22.1%), neither (19.7%), moderate liberal (21.9%), and strong liberal (16.3%). Moderate conservatives, neither, and moderate liberals were grouped together as the majority in the middle for analysis for FOI and WTSC with strong conservatives and liberals as the minority in comparison in numbers.

Measures

Social Media Use

Participants were what type of social media platform they use such as Facebook (78%), Instagram (43.9%), Twitter (17.3%), LinkedIn (16.3%), Snapchat (20.1%), and TikTok (12.6%). Further questions asked how often participants use their social media and participate with reactions (like, heart, other emojis), comment, post, or share and if they used it most of the time, about half the time, or rarely. Social media use was viewed as vocal through participants’ answers to comment, post, or share and was then analyzed as a whole measurement.

Fear of Isolation

The factor of FOI was measured from the scale developed by Hayes et al. (2013). These scale items were adapted to reflect context with political and social media constructs. Participants responded to five items that were based off a 5-point Likert-type scale from 1 (strongly disagree) to 5 (strongly agree) with items such as “One of the worst things that could happen to me on social media is that my friends block me because of my opinions about politics” and “It is scary to think about not being connected on social media with my friends because of political differences” (Cronbach α = .86).

Willingness to Self-Censor

Looking at participants’ WTSC on social media and terms of political situations, this study adapted Hayes et al. (2005b) 8-item scale WTSC scale. Using a 5-point Likert-type scale with 1 (strongly disagree) and 5 (strongly agree), the items were adapted in terms of censorship of political social media posts. With examples of items being: “It is difficult for me to express my opinion about political topics on social media if I think others won’t agree with what I say” and “When I disagree with others’ political viewpoints on social media, I’d rather stay quiet than post or comment about it.” Some items needed to be recoded to match agreement and items were summed into an index (Cronbach α = .85).

Political Typology

Questions asked participants to self-identify their political affiliations such as Democrat, Liberal, Republican, Conservative, or Independent. These responses were representative of how participants identify with a political party and where on a left and right scale they would put their political affiliations.

In addition, participants were asked about their political ideology typology through 17 statements of politicizing social issues. These questions, and the political typology categories that participants were grouped into based on their responses, were taken from Pew Research Center. Pew looked beyond the simple partisanship of Democrats and Republicans to gain a better understanding of people’s political leanings and politics in the United States (Doherty et al., 2017). Participants were asked to choose between two options on political social issues and select which item that was closer to their political views. Examples of issues asked are as follows: “Government regulation of business is necessary to protect the public interest” and “Government regulation of business usually does more harm than good.” Other topics in the survey include poverty, government intervention, environmental laws, foreign policy, and government spending, and so on. The political typology questions from Pew help individuals become aware of where their political ideologies lie on a left to right spectrum that fits between liberal and conservative. Because the questions are about vexing social issues, the Pew typology goes beyond political self-identification (Republican and Democrat) and political partisanship. Pew’s own analysis of the 17 social issue statements resulted in a statistically sound eight-category typology. The different typology groups are as follows: solid liberal, opportunity democrats, disaffected democrats, devout and diverse, new era enterprisers, market skeptics republicans, country first conservative, and core conservatives. These groups are separated in half with solid liberals, opportunity democrats, disaffected democrats, and devout and diverse with Democrat/liberal leanings; and new era enterprisers, market skeptics republicans, and country first conservatives, and core conservatives with Republican/conservative leanings. The different typologies demonstrate how participants feel about political issues clarifying where they lie on within political ideologies along the spectrum of democrats and conservative beliefs.

Participants in the current study were sorted into one of the eight Pew-categories based on their responses through an algorithm created and shared with the researchers by Pew. For analysis, the typologies were analyzed as separate groups in comparison to FOI and WTSC on social media. The political typology is a distinct variable from the self-identification measure of political affiliation because political typologies clarify the political beliefs of a person or participant who may not know where they actually lie on the political spectrum based on their opinion of issues. On the other hand, political identification demonstrates how participants self-identify in terms of partisan politics.

Results

The following results assess the difference between political identity and political typology between the factors of WTSC and FOI. Different analysis of variance (ANOVA) tests were run with results from factors about WTSC and total social media use as well as FOI and social media use and how it was displayed through political identity and typology.

Before running the ANOVA tests, FOI and WTSC were examined in terms of their relationship and dimensionality as dependent variables to determine whether ANOVA or MANOVA should be used in the analysis. In terms of relationship, FOI and WTSC had a significant but weak Pearson correlation coefficient (r = .22, p < .001). Furthermore, to determine dimensionality, an EFA (principal components with varimax rotation) was conducted with the indicators for both variables. The EFA showed that both variables clearly loaded on separate factors. Due to the weak correlation and the separate factor loadings of the two variables, this study looks at FOI and WTSC as separate dependent variables.

Self-Censorship

H1 sought to understand the relationship of self-censorship with social media use and political typology and identity. Differences between political typology and identity were found through different ANOVA tests.

A one-way ANOVA was conducted to examine the effect of ideological political differences on people’s self-reported, self-censorship on social media. ANOVA results were not significant, F(7, 999) = 0.99, p = .44. Table 1 represents the averages of self-censorship with political typology. The differences in the means are shown in Figure 1.

Table 1.

Means, Standard Deviations, and One-Way Analysis of Variance of Self-Censorship by Pew Political Typology.

	Mean	SD
Solid liberal	3.19	0.90
Opportunity democrats	3.42	0.92
Disaffected democrats	3.26	0.92
Devout and diverse	3.24	0.65
New era	3.39	0.83
Market skeptics	3.45	0.87
Country first	3.22	0.91
Core conservatives	3.28	0.97

F(7, 1,001) = 1.79, p = .085, partial η² = .01.

Figure 1.

Comparison of Self-Censorship Means for Pew Typology Categories.

A one-way ANOVA was conducted to examine the effect of political identity differences on people’s self-reported, self-censorship on social media. Results of the ANOVA, F(2, 1038) = 8.43, p < .001, partial η² = 0.02, showed a significant main effect of self-reported political identity on self-censorship on social media. Bonferroni post hoc tests revealed that strong liberals (M = 3.15, SD = 0.93) differed significantly from the majority in the middle (M = 3.40, SD = 0.83; M_diff = 0.24, SD = 0.07, p = .001). Similarly, strong conservatives (M = 3.19, SD = 0.90) differed significantly from the majority in the middle (M = 3.40, SD = 0.83; M_diff = 0.20, SD = 0.07, p = .02). There was not a significant difference between strong liberals and strong conservatives. Table 2 displays the averages between strong liberals, strong conservatives, and the majority in the middle with Figure 2 displaying the differences in the means.

Table 2.

Means, Standard Deviations, and One-Way Analysis of Variance of Self-Censorship by Self-Identified Political Identity.

Strong liberal		Majority in the middle		M_diff	SD	p
Mean	SD	Mean	SD	M_diff	SD	p
3.18	1.04	3.37	0.81	0.19	0.08	.04
Majority in the middle		Strong conservative		M_diff	SD	p
Mean	SD	Mean	SD	M_diff	SD	p
3.37	0.81	3.28	1.00	0.09	0.08	.52
Strong liberal		Strong conservative		M_diff	SD	p
Mean	SD	Mean	SD	M_diff	SD	p
3.18	1.04	3.28	1.00	0.10	0.11	.61

Figure 2.

Comparison of Self-Censorship Means for Self-Described Political Identity.

Willingness to Self-Censor

H2 sought to understand the relationship between political typology and identity with social media and the FOI.

A one-way ANOVA was conducted to examine the effect of ideological political differences on people’s self-reported, FOI on social media. ANOVA results were not significant, F(2, 1,040) = 1.18, p = .31, partial η² = 0.002. Table 3 displays the averages for FOI and political identity on social media. The differences between the means are shown in Figure 3.

Table 3.

Means, Standard Deviations, and One-Way Analysis of Variance of Fear of Isolation by Self-Identified Political Identity.

	Mean	SD
Strong liberal	2.47	1.12
Majority in the middle	2.58	0.90
Strong conservative	2.61	1.21

F(2, 1040) = 1.18, p = .31, partial η² = 0.002.

Figure 3.

Comparaison of Fear of Isolation Means for Self-Described Political Identity.

A one-way ANOVA was conducted to examine the effect of ideological political differences on people’s self-reported, FOI on social media. Because the assumption of homogeneity of variances was violated for this test, Levene Statistic (2, 1,000) = 2.73, p = .008, the analysis was performed using the Games-Howell test. Results show significant differences between solid liberals and opportunity democrats (M_diff = 0.78, SD = 0.11, p < .001), disaffected democrats (M_diff = 0.49, SD = 0.12, p = .001), devout and diverse (M_diff = 0.81, SD = 0.12, p < .001), new era enterprisers (M_diff = 0.59, SD = 0.11, p < .001), and market skeptics (M_diff = 0.34, SD = 0.10, p = .02). On the liberal side of the typology, additional differences were found between opportunity democrats and market skeptics (M_diff = 0.43, SD = 0.11, p = .003), country first republicans (M_diff = 0.45, SD = 0.13, p = .008), and core conservatives (M_diff = 0.72, SD = 0.12, p < .001). There was also a significant difference between disaffected democrats and core conservatives (M_diff = 0.44, SD = 0.12, p = .01). Similarly, people in the devout and diverse typology group differed significantly from market skeptics (M_diff = 0.46, SD = 0.14, p = .02), country first republicans (M_diff = 0.49, SD = 0.14, p = .01), and core conservatives (M_diff = 0.76, SD = 0.13, p < .001). On the conservative side of the typology, there was a significant difference between new era enterprisers and core conservatives (M_diff = 0.54, SD = 0.13, p < .001; see Tables 4 and 5, and Figure 4 for the averages between groups).

Table 4.

Means, Standard Deviations, and One-Way Analysis of Variance of Fear of Isolation by Liberal Pew Political Typologies.

Solid liberal		Opportunity democrats		M_diff	SD	p
Mean	SD	Mean	SD	M_diff	SD	p
2.20	0.90	2.97	1.12	0.78	0.11	<.001
		Disaffected democrats		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		2.69	0.91	0.49	0.12	.001
		Devout and diverse		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		3.01	0.85	0.81	0.12	<.001
		New era enterprisers		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		2.79	0.96	0.59	0.11	<.001
		Market skeptics		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		2.55	1.07	0.34	0.10	.02
Opportunity democrats		Market skeptics		M_diff	SD	p
Mean	SD	Mean	SD	M_diff	SD	p
2.97	1.12	2.55	1.07	0.43	0.11	0.003
		Country First Republicans		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		2.52	0.85	0.45	0.13	.008
		Core conservatives		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		2.25	0.92	0.72	0.12	<.001

Table 5.

Means, Standard Deviations, and One-Way Analysis of Variance of Fear of Isolation by Conservative Pew Political Typologies.

Core conservative		Disaffected democrats		M_diff	SD	p
Mean	SD	Mean	SD	M_diff	SD	p
2.25	0.92	2.69	0.91	0.44	0.12	.01
		Devout and diverse		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		3.01	0.85	0.76	0.13	<.001
		New era enterprisers		M_diff	SD	p
		Mean	SD	M_diff	SD	p
		2.79	0.96	0.54	0.13	<.001

Figure 4.

Comparison of Fear of Isolation Means for Pew Typology Categories.

Discussion

Of the several studies that have looked at the spiral of silence, FOI, and WTSC, there has been limited research in how these factors are affected by political typology, identity, and social media usage. Many previous studies have looked at the factors of WTSC and FOI in terms of the spiral of silence with how opinions climates have been shared and changed with online or offline conversations. Notably, many studies have investigated how the hardcore vocal minority opinion speak out against the opinions they do not share (Gearhart & Zhang, 2018; McKeever et al., 2016; Porten-Cheé & Eilders, 2015). The aim of this study was to view how political factors, self-censorship, and FOI may be affecting how users interact or share on social media and if the majority opinion is becoming more silent as the hardcore minority opinion becomes louder.

As suggested from the literature, the hardcore vocal minority opinion will not be silenced (Noelle-Neumann, 1974), and they will speak up about the issues they care about (Porten-Cheé & Eilders, 2015). In terms of H1, the results demonstrated that a majority of voices are self-censoring themselves in terms of political identity. With low levels of self-censorship for strong conservatives and liberals, findings suggest that the minority opinions are not self-censoring on social media as much as the majority in terms of identity politics. This suggest that the hardcore minority opinion or the political margins may be speaking up with their peers online which agrees with the findings of Goren and others (2009) that the political margins are more vocal. In terms of political identity and evidence of the spiral of silence through self-censorship, these findings start to suggest that a shift in the spiral of silence is appearing through the vocal minority opinion and there is a presence of a silent majority.

Interestingly, findings revealed that self-censorship on social media was not significantly impacted by political typologies across the spectrum of liberal to conservative. With no major differences between the range of liberals and conservatives on typology or with how participants view political ideology with their self-censorship on social media, but differences within self-reported political identity, it is assumed that identity politics could be a motivation for self-censoring and not the political ideologies. Previous research found that a single ideology can create extreme opinions (Kashima et al., 2021) which may suggest from the findings that it is the presence of identity politics that create extreme opinions as well as the difference of political ideology where self-censorship on social media was not found significant in this study.

FOI was viewed in H2 as a motivating factor for WTSC. From the literature, research suggest that that FOI could already be present to allow people to self-censor or isolate their opinions (Matthes et al., 2018) or increases a person’s WTSC (Fox & Holt, 2018). These study findings indicate no significant data or correlation with FOI in relation to political identity and WTSC on social media. This suggests that although self-censorship was found among political typologies for the majority in the middle, FOI was not the motivating factor.

While FOI yields no significance in political identity and social media usage, variance of FOI was found in the differences between political typologies. Opportunity democrats, disaffected democrats, devout and diverse, new era enterprisers, and market skeptics all had higher levels of FOI than solid liberals. This suggests that these groups of voices fear speaking out whereas the solid liberals have very low FOI on social media compared to the other groups. These findings suggest that the margin on the liberal sides does not fear an isolating opinion on social media as much as other democrat ideologies. With the significant differences to the other democrat groups, that the findings suggest that solid liberals may have lower FOI, and that the other liberal groups are isolating from solid democrats. The findings may suggest that solid liberals may not face threats to perception of identity or are more recognized in the hardcore minority opinion grouping (Noelle-Neumann, 1993; Powers et al., 2019).

The results found differences between other groups of political ideologies for levels of FOI on social media usage in comparison to other groups. Opportunity democrats have high levels of FOI in comparison to conservative ideology suggesting that conservatives do not fear isolation on social media as much as opportunity democrats. Conservative ideologies within FOI on social media result are not as disruptive or diverse as differences between democrat ideologies. The findings suggest that the margins for conservatives such as core conservatives are less likely to show FOI, much like strong democrats. This is also in line with the vocal hardcore minority opinion such as strong liberals and core conservatives refusing to silence their voice or take part in the spiral of silence (Noelle-Neumann, 1993). With WTSC being low for political identities from strong conservatives and democrats, along with core ideologies of solid liberals and core conservatives demonstrating low levels of FOI, this study suggests that the hardcore vocal minority opinion is not affected by WTSC or FOI and will speak their opinion in any condition.

Limitations and Future Research

Limitations that are found in this study may be due to the sample. There was a majority percentage of White participants, and the study may be lacking in more diverse voices especially in terms of political ideology and identity. This is because the study chose to mirror the US census rather than a demographic represented on for social media users. Future research could look for a more diverse and larger range of participants that are representative of active social media users. Other limitations may also be due to the percentages of differences in participants for the pew political ideology typology. This study had a higher percentage of participants as solid liberals than other categories of the typology. There are also limitations from using Pew’s political typology because there is a possibility of imperfection using their measures and algorithms. Within a constantly changing political spectrum, the social issue questions used to construct the typology may become outdated. The survey to participants was self-reported which could display error in answers for social media usage and political identity. This study did not look at a singular platform for social media, but there was a large percentage of participation for Facebook which can lead to limitations on how answer may vary depending different platforms.

This analysis of this study suggests that it is the theory of the spiral of silence that explains political identities are self-silencing. The spiral of silence suggest that one group silences another group, future research should look at the different variables of FOI and WTSC to look at variables of why people may be self-censoring. Social media responses did not focus on a specific social media platform, and more information about FOI and WTSC may be found on specific platforms such as Twitter or Facebook. Because this study found that FOI was not a motivating factor for social media self-censoring for political ideologies, future studies can investigate what other variables may be causing self-censorship. This study took a quantitative approach through a survey, other research could approach this study through in-depth interviews to understand why people may self-censor or demonstrate FOI online politically. Information was gathered about how often participants consume news for demographic purposes, future research could investigate how news consumption on social media may affect political factors online and self-censoring and isolating motivations. Other factors of social media use could also be explored for amount of time spent on social media and official social activities for participants with WTSC and FOI use. Researchers can also investigate other factors of WTSC and FOI beyond political conversation on social media and view other variables that may affect social media self-censorship. An in-depth analysis on the differences of political ideologies and typology in comparison to political identity and their interaction on a specific social media platform could also be beneficial to understanding FOI and WTSC.

This study sought to understand self-censorship and FOI on social media to see if the spiral of silence was being flipped from a perceived majority to a vocal hardcore minority opinion (Noelle-Neumann, 1974). This study found evidence for the spiral of silence to be censoring the majority in the middle, or the moderate conservatives, independents, and moderate liberals are self-censoring on social media creating a silencing effect of the majority. FOI was not the moderating factor for this self-censorship on social media, but it is contributing factor to differences between political ideologies. With the continuing change on social media and the rise of polarization online, more attention will need to focus on how the spiral of silence is changing and what factors may be causing people to self-censor online.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Brigham Young University Ashton Grant.

ORCID iD

Alycia Burnett

Author Biographies

Alycia Burnett (MA) is a recent graduate student of Mass Communications from Brigham Young University. Her research interests include social media influencers, film studies, and media effects.

Devin Knighton (PhD) is an assistant professor of public relations at Brigham Young University in the School of Communications. His research interests include social networking influences, and public relations management and strategy.

Christopher Wilson (PhD) is an associate professor of public relations at Brigham Young University in the School of Communications. His research interests include public relations management and strategy.

References

Auxier

(2020, August 28). Activism on social media varies by race and ethnicity, age, political party. Pew Research Center. https://www.pewresearch.org/fact-tank/2020/07/13/activism-on-social-media-varies-by-race-and-ethnicity-age-political-party/

Chan

(2018). Reluctance to talk about politics in face-to-face and Facebook settings: Examining the impact of fear of isolation, willingness to self-censor, and peer network characteristics. Mass Communication and Society, 21(1), 1–23. https://doi.org/10.1080/15205436.2017.1358819

Chaudhry

Gruzd

(2020). Expressing and challenging racist discourse on Facebook: How social media weaken the “spiral of silence” theory. Policy & Internet, 12(1), 88–108. https://doi.org/10.1002/poi3.197

Chen

H. T.

(2018). Spiral of silence on social media and the moderating role of disagreement and publicness in the network: Analyzing expressive and withdrawal behaviors. New Media & Society, 20(10), 3917–3936. https://doi.org/10.1177/1461444818763384

Connolly

W. E.

(2002). Identity, difference: Democratic negotiations of political paradox. University of Minnesota Press.

Doherty

Kiley

Johnson

(2017). Political typology reveals deep fissures on the right and left. Pew Research Center.

Druckman

J. N.

(2014). Pathologies of studying public opinion, political communication, and democratic responsiveness. Political Communication, 31(3), 467–492.

Dvir-Gvirsman

(2017). Media audience homophily: Partisan websites, audience identity and polarization processes. New Media & Society, 19(7), 1072–1091.

Fox

Holt

L. F.

(2018). Fear of isolation and perceived affordances: The spiral of silence on social networking sites regarding police discrimination. Mass Communication and Society, 21(5), 533–554. https://doi.org/10.1080/15205436.2018.1442480

10.

Fox

Warber

K. M.

(2015). Queer identity management and political self-expression on social networking sites: A co-cultural approach to the spiral of silence. Journal of Communication, 65(1), 79–100.

11.

Gearhart

Zhang

(2014). Gay bullying and online opinion expression: Testing spiral of silence in the social media environment. Social Science Computer Review, 32(1), 18–36. https://doi.org/10.1177/0894439313504261

12.

Gearhart

Zhang

(2018). Same spiral, different day? Testing the spiral of silence across issue types. Communication Research, 45(1), 34–54. https://doi.org/10.1177/0093650215616456

13.

Goren

Federico

C. M.

Kittilson

M. C.

(2009). Source cues, partisan identities, and political value expression. American Journal of Political Science, 53(4), 805–820. https://doi.org/10.1111/j.1540-5907.2009.00402.x

14.

Hampton

K. N.

Shin

(2017). Social media and political discussion: When online presence silences offline conversation. Information, Communication and Society, 20(7), 1090–1107. https://doi.org/10.1080/1369118X.2016.1218526

15.

Hayes

A. F.

Glynn

C. J.

Shanahan

(2005a). Validating the willingness to self-censor scale: Individual differences in the effect of the climate of opinion on opinion expression. International Journal of Public Opinion Research, 17(4), 443–455. https://doi.org/10.1093/ijpor/edh072

16.

Hayes

A. F.

Glynn

C. J.

Shanahan

(2005b). Willingness to self-censor: A construct and measurement tool for public opinion research. International Journal of Public Opinion Research, 17(3), 298–323. https://doi.org/10.1093/ijpor/edh073

17.

Hayes

A. F.

Matthes

Eveland

W. P.

(2013). Stimulating the quasi-statistical organ: Fear of social isolation motivates the quest for knowledge of the opinion climate. Communication Research, 40(4), 439–462. https://doi.org/10.1177/0093650211428608

18.

Hoffmann

C. P.

Lutz

(2017). Spiral of silence 2.0: Political self-censorship among young Facebook users. In Proceedings of the 8th international conference on social media & society (pp. 1–12). https://doi.org/10.1145/3097286.3097296

19.

Kashima

Perfors

Ferdinand

Pattenden

(2021). Ideology, communication, and polarization. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 376(1822), 20200133.

20.

Kitchens

Johnson

S. L.

Gray

(2020). Understanding echo chambers and filter bubbles: The impact of social media on diversification and partisan shifts in news consumption. The Mississippi Quarterly, 44(4), 1619–1649.

21.

Krishna

Kim

(2020). Understanding President Trump supporters’ behaviors about a political controversy. International Journal of Information and Communication Technology Education: An Official Publication of the Information Resources Management Association, 24(2), 119–136.

22.

Kushin

M. J.

Yamamoto

Dalisay

(2019). Societal majority, Facebook, and the spiral of silence in the 2016 US presidential election. Social Media & Society, 5(2), 1–11. https://doi.org/10.1177/2056305119855139

23.

Kwon

K. H.

Moon

S.-I.

Stefanone

M. A.

(2015). Unspeaking on Facebook? Testing network effects on self-censorship of political expressions in social network sites. Quality & Quantity, 49(4), 1417–1435.

24.

Liu

Fahmy

(2011). Exploring the spiral of silence in the virtual world: Individuals’ willingness to express personal opinions in online versus offline settings. Journal of Media and Communication Studies, 3(2), 45–57.

25.

Marder

Houghton

Joinson

Shankar

Bull

(2016). Understanding the psychological process of avoidance-based self-regulation on Facebook. Cyberpsychology, Behavior and Social Networking, 19(5), 321–327. https://doi.org/10.1089/cyber.2015.0564

26.

Matthes

(2015). Observing the “spiral” in the spiral of silence. International Journal of Public Opinion Research, 27(2), 155–176. https://doi.org/10.1093/ijpor/edu032

27.

Matthes

Knoll

von Sikorski

(2018). The “spiral of silence” revisited: A meta-analysis on the relationship between perceptions of opinion support and political opinion expression. Communication Research, 45(1), 3–33. https://doi.org/10.1177/0093650217745429

28.

McKeever

B. W.

McKeever

Holton

A. E.

J.-Y.

(2016). Silent majority: Childhood vaccinations and antecedents to communicative action. Mass Communication and Society, 19(4), 476–498. https://doi.org/10.1080/15205436.2016.1148172

29.

Neubaum

Krämer

N. C.

(2018). What do we fear? Expected sanctions for expressing minority opinions in offline and online communication. Communication Research, 45(2), 139–164. https://doi.org/10.1177/0093650215623837

30.

Noelle-Neumann

(1974). The spiral of silence a theory of public opinion. The Journal of Communication, 24(2), 43–51. https://doi.org/10.1111/j.1460-2466.1974.tb00367.x

31.

Noelle-Neumann

(1991). The theory of public opinion: The concept of the spiral of silence. Annals of the International Communication Association, 14(1), 256–287. https://doi.org/10.1080/23808985.1991.11678790

32.

Noelle-Neumann

(1993). The spiral of silence: Public opinion, our social skin. University of Chicago Press.

33.

Ordoñez

M. A. M.

Nekmat

(2019). “Tipping point” in the SoS? Minority-supportive opinion climate proportion and perceived hostility in uncivil online discussion. New Media & Society, 21(11–12), 2483–2504.

34.

Pariser

(2011). The filter bubble: How the new personalized web is changing what we read and how we think. Penguin.

35.

Porten-Cheé

Eilders

(2015). Spiral of silence online: How online communication affects opinion climate perception and opinion expression regarding the climate change debate. Studies in Communication Sciences, 15(1), 143–150. https://doi.org/10.1016/j.scoms.2015.03.002

36.

Poulakidakos

Veneti

Fangonikolopoulos

(2018). Post-truth, propaganda and the transformation of the spiral of silence. International Journal of Media & Cultural Politics, 14(3), 367–382. https://doi.org/10.1386/macp.14.3.367_1

37.

Powers

Koliska

Guha

(2019). “Shouting matches and echo chambers”: Perceived identity threats and political self-censorship on social media. International Journal of Communication, 13(1), 3630–3649.

38.

Scheufele

D. A.

Shanahan

Lee

(2001). Manipulating the dependent variable in the spiral of silence research. Communication Research, 28, 304–324. https://doi.org/10.1177/009365001028003003

39.

Sohn

(2022). Spiral of silence in the social media era: A simulation approach to the interplay between social networks and mass media. Communication Research, 49, 139–166.

40.

Song

Boomgaarden

H. G.

(2017). Dynamic spirals put to test: An agent-based model of reinforcing spirals between selective exposure, interpersonal networks, and attitude polarization. The Journal of Communication, 67(2), 256–281.

41.

Tappan

M. B.

(2005). Domination, subordination and the dialogical self: Identity development and the politics of “Ideological Becoming.” Culture & Psychology, 11(1), 47–75.

42.

Tsfati

Stroud

N. J.

Chotiner

(2014). Exposure to ideological news and perceived opinion climate: Testing the media effects component of spiral-of-silence in a fragmented media landscape. The International Journal of Press/Politics, 19(1), 3–23. https://doi.org/10.1177/1940161213508206

43.

Wang

M. Y.

Hmielowski

J. D.

Hutchens

M. J.

Beam

M. A.

(2017). Extending the spiral of silence: Partisan media, perceived support, and sharing opinions online. Journal of Information Technology & Politics, 14(3), 248–262. https://doi.org/10.1080/19331681.2017.1338980

44.

Wojcieszak

Rojas

(2011). Hostile public effect: Communication diversity and the projection of personal opinions onto others. Journal of Broadcasting & Electronic Media, 55(4), 543–562.

45.

Yarchi

Baden

Kligler-Vilenchik

(2021). Political polarization on the digital sphere: A cross-platform, over-time analysis of interactional, positional, and affective polarization on social media. Political Communication, 38, 98–139. https://doi.org/10.1080/10584609.2020.1785067

The Self-Censoring Majority: How Political Identity and Ideology Impacts Willingness to Self-Censor and Fear of Isolation in the United States

Abstract

Keywords

Literature Review

Spiral of Silence

Hardcore Vocal Minority Opinions

Fear of Isolation and Willingness to Self-Censor

Political Climates

Current Study

Method

Population

Participants

Measures

Social Media Use

Fear of Isolation

Willingness to Self-Censor

Political Typology

Results

Self-Censorship

Willingness to Self-Censor

Discussion

Limitations and Future Research

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Author Biographies

References