You Are Wrong Because I Am Right! The Perceived Causes and Ideological Biases of Misinformation Beliefs

Abstract

In the context of increasing concerns about false and deceptive information in public opinion, this research note explores which causes and sources news users associate with the dissemination of misinformation. Based on representative survey data collected in the Netherlands (N = 1,994), we found that news users distinguish unintentional causes related to uncertain evidence and lacking expert knowledge from politically or financially motivated falsehoods. People on the left-wing of the political spectrum associate falsehoods more with the radical-right, whereas those on the right-wing tend to associate misinformation with the radical-left. Right-wing participants, however, are most likely to perceive misinformation as driven by a deliberate attempt to hide reality. Our findings point to an ideological bias in information credibility that could foster polarization along epistemic lines.

The dissemination of false and misleading content in digital information settings has been described as a severe challenge for democracy (e.g., Freelon & Wells, 2020; Marwick & Lewis, 2017; Wardle, 2017). Especially in times of crisis, such as the COVID-19 pandemic, news users perceived high levels of misinformation in their information environment (Nielsen et al., 2020). As cynical perceptions related to the goal-directed manipulation of information (i.e., conspiracy beliefs) can have more severe consequences than skeptical beliefs related to the lack of expert knowledge (i.e., misinformation beliefs), it is important to explore the perceived sources, origins, and causes of mis- and disinformation resonating in public opinion. In this research note, we therefore map the audience’s credibility perceptions in a communication context associated with increasing concerns on mis- and disinformation, and the weaponization of the term Fake News among politicians and citizens (Tong et al., 2020).

As key theoretical contribution, we assess how the different unintentional and intentional causes associated with mis- and disinformation on the supply side of the information ecology (see e.g., Bennett & Livingston, 2018; Wardle & Derakhshan, 2017) resonate with public perceptions. With our study, we aim to explore the demand side of mis- and disinformation as a perceptual crisis, moving beyond most research that has looked at false information as an informational phenomenon. This endeavor is relevant considering the relatively rare occurrence of misinformation as an informational crisis vis-à-vis the salience of misinformation beliefs and discussions about Fake News and other forms of falsehoods in society. Extant research found that less than one percent of U.S. citizens’ media exposure consists of misinformation (Allen et al., 2020). At the same time, however, citizens across the globe are very concerned about false and misleading information. The Reuters Digital News Report (2021), for example, indicates that 82% of news users in Brazil are concerned about false information. Even in the “least concerned” nation, Germany, 37% of all news users are worried about misinformation. As the perceptual crisis surrounding misinformation and the weaponized term “Fake News” seems more prevalent than the actual proportion of misinformation in people’s newsfeed, it is crucial to explore the nature and perceived causes of beliefs in misinformation.

In line with this, perceptions of false information are found to have real consequences. Such beliefs can, for example, coincide with reduced willingness to comply with the authorities in the context of COVID-19 (Hameleers et al., 2020). Distrusting news users are also more likely to find misinformation credible (Zimmermann & Kohring, 2020), and approach alternative media outlets that perpetrate “Fake News” accusations (Müller & Schulz, 2021). Even though extant research has increasingly focused on the user side of mis- and disinformation by mapping users’ susceptibility to false information (e.g., Schaewitz et al., 2020; Zimmermann & Kohring, 2020) or perceptions of the “Fake News” term (Tong et al., 2020), we know little about which causes citizens perceive to be driving false information. To more comprehensively understand how news users perceive misinformation, and how these perceptions differ across ideological divides, we rely on representative survey data collected in the Netherlands and map citizens’ (1) misinformation perceptions related to different offline and online information channels; (2) the specific causes they perceive to be the driving forces of falsehoods and; (3) the ideological bias of misinformation beliefs in public opinion.

Perceived Untruthfulness: Intentional Deception or Unmotivated Errors?

Misinformation can generally be defined as erroneous information that is not necessarily disseminated with the intention to mislead or deceive the public (e.g., Wardle, 2017). Misinformation may be seen as an overarching concept that refers to all sorts of untruthful information that is deemed false in light of relevant empirical evidence and expert knowledge. Disinformation refers to the intentional deception, manipulation, or doctoring of information that is spread to achieve a certain political goal (Bennett & Livingston, 2018; Freelon & Wells, 2020; Marwick & Lewis, 2017). False information may also be motivated by strategical intentions (i.e., gaining attention and clicks in a context of increased competition), financial gains (i.e., advertising revenues), or the ideological de-legitimization of opposed views (i.e., the Fake News label). Intention may thus distinguish mis- from disinformation—a notion that has become central in many conceptualizations of communicative untruthfulness (e.g., Karlova & Fisher, 2013). Yet, we lack an understanding of how these different intentions are perceived by the public: Do news users also distinguish “honest mistakes” from goal-directed deception?

As misinformation as a perceptual crisis is likely to be more prevalent than the actual dissemination of falsehoods, and as people with stronger concerns about politically motivated disinformation are shown to demonstrate a lower level of trust in the media whilst engaging in less pro-social behaviors amidst a global health crisis (Hameleers et al., 2020), it is crucial to assess the causes people associate with misinformation. Just like disinformation may be driven by different goals, perceived causes might differ in their degree of intentionality and could relate to perceived inconclusive evidence or a lack of expert knowledge (resulting in unintentional errors) or, at the other end of the spectrum, beliefs in goal-directed political manipulation, or even hiding reality from the public (resulting in deceptive disinformation). The core question in this regard is whether news users attribute false information to unintentional causes or deliberate deception driven by specific (political or financial) goals. To explore to what extent and how news users’ own perceptions of the origins and causes of false information resonate with supply-side conceptualizations of mis- and disinformation, we forward the following research question (RQ₁): What causes and origins do news consumers associate with the dissemination of false information?

The Ideological Resonance of Perceived Untruthfulness

Literature on the hostile media bias postulates that strong partisans tend to perceive information on issues they deem important to be biased against their views (Choi et al., 2009; Vallone et al., 1985). Empirical evidence in this regard shows that exactly the same content is seen as hostile when it comes from an opposed media source, whereas it is seen as substantially less hostile when it comes from a supported or ideologically similar media source (Arpan & Raney, 2003). These hostile and friendly media biases can be understood as a consequence of motivated reasoning (Festinger, 1957). When media outlets are seen as supportive of an individual’s (ideological) views on highly involved issues, this can be thought of as a cue of trustworthiness and favoritism (Goldman & Mutz, 2011). Based on this, misinformation may be seen as most likely when it comes from the other side. People on the left-wing of the ideological spectrum may therefore perceive misinformation as coming mostly from the right, whereas news users with right-wing orientations may associate misinformation mostly with left-wing sources. We therefore raise the following hypothesis on the ideological bias of mis- and disinformation perceptions: Mis- and disinformation is most likely to be associated with sources that are opposed to people’s ideological views (H1).

Method

Respondents were recruited by the panel company Ipsos in the Netherlands through various online channels. They were rewarded for their participation through the company’s credit system. Quotas were used to achieve a representative sample regarding age, region, and gender interlocked with education. The survey was distributed between November 19, 2020 and December 7, 2020. A total of 21,861 invitations were sent out; 7,846 responded. However, not all these respondents were included in the sample, for example, because quota had already been met, they did not complete the questionnaire, or failed both of the two attention checks. The final sample consisted of 1,994 complete and valid responses. Details on the composition of the sample and the recruiting process can be found on the Open Science Framework at https://osf.io/yu64r/.

The survey questions used for this study were part of a larger, collaborative research project (see Araujo et al., 2020) that covered several topic areas. The survey started with demographic questions followed by political orientation, measured as follows: “In politics, people sometimes talk of ‘left’ and ‘right.’ Where would you place yourself on this scale, where 0 means the left and 10 means the right? (You are not obligated to answer and can select the option ‘I do not want to answer’).” Several unrelated questions followed, which should prevent a too strong priming effect of ideological biases affecting the misinformation beliefs tapped later. Yet, even though the Dutch setting is not as polarized regarding partisan identities as the United States, it should be acknowledged that asking about people’s ideological orientations may activate their political biases. Perceptions of false and deceptive information were measured as follows: “To what extent do you agree with the following statements on a scale from 1 (completely disagree) to 7 (completely agree)? (1) There is a lot of false information in the current information environment. (2) There is a lot of deceptive information in the current information environment.” The item order was randomized.

Perceived causes for misinformation were measured as follows: “To what extent do you agree with the following statements on a scale from 1 (completely disagree) to 7 (completely agree)? False information is mostly disseminated due to… (1) …a lack of expert knowledge. (2) …when insufficient evidence and facts are available. (3) …because of uncertainty about facts and evidence. (4) …to disrupt the societal order. (5) …to influence our political decisions. (6) …to make financial gains. (7) …to hide reality from the people.” The item order was randomized. These items were developed for this study in particular, which we regard as a first tentative exploration of underlying motivations for perceived false information. We based the formulation of items on literature on mis- and disinformation as informational phenomena, in which a distinction between unmotivated errors (misinformation) and goal-directed disinformation with financial, political, or ideological goals is often made (e.g., Wardle & Derakhshan, 2017). In addition, for the more “extreme” causes pointing to deception and hiding reality from the people, we based ourselves on the link between populist attitudes and perceptions of a hostile media bias (e.g., Schulz et al., 2020).

Sources of misinformation were assessed as follows: “To what extent do you agree with the following statements on a scale from 1 (completely disagree) to 7 (completely agree)? False information… (1) …is present most on mainstream media platforms. (2) …is present most in digital information settings (such as social media). (3) …is spread by the radical right-wing. (4) …is spread by the radical left-wing.” The item order was randomized. We decided to refer to “false information” in general, as misinformation can mean different things according to different subjective interpretations and conceptualizations. We thus allowed for different individual perceptions of what false information may imply—corresponding to the multifaced nature of mis- and disinformation as a supply-side phenomenon.

Results

Perceived Causes and Sources of Misinformation

First, we asked participants to identify the causes underlying the dissemination of misinformation, and the most likely sources of false information (RQ₁). The mean scores of the different perceived causes are displayed in Figure 1. A series of Bonferroni-corrected pairwise t tests were conducted to assess the differences between the different causes explored in our survey. Participants perceived all causes as relatively likely (the average scores are higher than the midpoint of the scale). However, they perceived more disruptive causes—such as hiding reality—to be less commonplace than general political motives and misinformation due to a lack of expert knowledge. Uncertainty was not considered differently from insufficient evidence (p = 1.00) or misinformation due to a lack of expert knowledge (p = .35). In addition, the perceived cause of financial gains was not statistically different from lack of expert knowledge (p = 1.00). All other differences are statistically significant.

The two more extreme, but also abstract motivations—hiding reality and causing disruption—are least commonly considered as causes of misinformation. Uncertainty, lack of expert knowledge, and insufficient evidence are considered more common—i.e., two motivations that are due to external factors, rather than the intentional dissemination of misinformation. Finally, financial gains and influencing politics—i.e., two more malicious, but less abstract, motivations are considered as motivators for misinformation most commonly by respondents (but note that the mean difference between lack of expert knowledge and financial gains is not statistically significant).

Regarding the perceived sources of misinformation, participants perceived misinformation as substantially less common in the mainstream media (M = 3.60, SD = 1.62) than in digital information settings (M = 5.57, SD = 1.25). This difference is significant according to a pair-wise t test (p < .001). Overall, participants think that there is more deceptive information (M = 5.61, SD = 1.17) than false information (M = 5.50, SD = 1.23). This difference is significant according to a pair-wise t test (p < .001), but is not large in size. These high levels of perceived mis- and disinformation point to significant issues with media credibility and trust—especially when applied to digital sources of information.

The Ideological Bias of Misinformation Beliefs and Causes

In the next step, we tested the expectation of an ideological or hostile media bias in perceived mis- and disinformation (H1). As can be seen in Figure 2, the plotted interaction effects between left-right self-placement and ideological biases of mis- and disinformation support our hypothesis. In addition, two simple correlation tests show that a more right-wing political orientation is associated with lower perceived levels of misinformation from the radical right (r = −.17, p < .001, df = 1843), whereas it is associated with higher perceived levels of misinformation from the radical left (r = .25, p < .001, df = 1843). Figure 2 shows this association in more detail: People who identify as left-wing are significantly and substantially more likely to perceive that false information is disseminated by the radical right (compared to the radical left). The same hostile/friendly bias applies to people identifying as right-wing: They are more likely to perceive that misinformation is disseminated by the radical left. Yet, there are some nuances to report: More moderate participants (scoring between 5 and 6 on the ideological self-placement scale) do not perceive an ideological bias in sources of misinformation (the confidence intervals are intersecting and there are no significant differences in this region). There is also some disbalance between left- and right-wing participants: The nonoverlapping area is bigger for people on the left (0–5) than the right (6–10). This means that, overall, left-wing participants perceive an ideological gap between ideological sources of misinformation in a more extensive region than right-wing participants. In support of H1, there is a strong ideological bias in the allocation of misinformation to the opposite side of the political spectrum.

If we look at how the perceived causes (RQ₁) are experienced differently by left- and right-wing-oriented participants (Figure 3), we see that less severe perceived causes and unmotivated errors are associated with misinformation to similar extents by left- and right-wing participants (i.e., misinformation driven by uncertainty). However, participants who identify more with the right-wing of the ideological spectrum are significantly more likely to associate misinformation with the motive of deliberately hiding reality from the people than participants on the left-wing. Hence, the most severe or extreme perceived cause identified in this study is substantially more likely to be expressed by right-wing than left-wing participants.

Discussion

This research note explored how commonplace the issue of misinformation is among news users, and which sources and causes they perceive to be driving its dissemination. Theoretically, we can discern three overarching explanations for misinformation’s spread identified by news users: (1) Unmotivated errors due to a lack of factual knowledge and expert agreement; (2) perceptions related to the motivated dissemination of falsehoods to achieve financial or political goals, and; (3) more extreme beliefs related to conspiracies: The goal-directed manipulation of information to hide reality or disrupt the societal order. Importantly, this more extreme cluster is occupied more by right-wing than left-wing participants, which shows that hostile perceptions of the news media as an enemy of the people are more salient on the right-wing.

Within the first cluster of unintentionally false information, three perceived causes for the spread of false information are considered similarly common: False information due to uncertainty, a lack of expert knowledge, or insufficient evidence. For intentional deception, there is a difference between more abstract and more concrete motivations. People are less likely to associate falsehoods with the goal of creating societal disruption and hiding reality from the people than with the goal of political and financial gains. More extreme and abstract motivations, that resemble conspiracy beliefs, are thus seen as less common than the more specific motivations of political and financial gains.

As a crucial implication of these findings, we show how misinformation is experienced as a perceptual crisis by news users. In line with previous research indicating that beliefs about motivated untruthfulness have more severe democratic implications than beliefs about honest mistakes (Hameleers et al., 2020), we believe that our findings are useful for research that aims to map the (political) consequences of misinformation beliefs. Adding to research that identified a relationship between populist attitudes and “Fake News” accusations as a more extreme form of disinformation beliefs (e.g., Fawzi, 2019), we offer a more refined assessment of the attitudinal components of misinformation beliefs that resonate with the distinctions between unmotivated and goal-directed falsehoods identified on the supply side of mis- and disinformation (Wardle & Derakhshan, 2017).

In line with this, it can be argued that the first cluster of perceived ‘honest mistakes’ may have more positive ramifications for society and democracy than beliefs related to conspiracies and politically motivated disinformation. When citizens believe that information is inaccurate due to a lack of expert consensus, for example, they may be more motivated to critically verify content, sources, and their own biases—which is in line with the principles of conducive news media literacy behaviors (Jones-Jang et al., 2021). However, when disinformation is seen as driven by goal-directed manipulation or financial aims, trust in established sources may decline, whereas citizens may approach more alternative sources that amplify their distrust (Hameleers et al., 2020). We leave it up to future research to use our typology and explore the (behavioral) consequences of the differentiated beliefs.

Our findings offer strong support for an ideological bias in mis- and disinformation beliefs. On the left-wing, people accuse the right-wing of spreading false information. People who identify more with the right-wing, in contrast, accuse the left-wing of spreading false information. This corresponds to a hostile media bias in news credibility ratings. As an implication, confirmation biases and polarization may be fostered by people’s media credibility ratings: People are more likely to accept information from their in-group, whereas they are likely to associate opposed sources with mis- and disinformation. Less openness to information from opposed ideological sources could result in “truth polarization” over time: People may selectively approach information in line with their political beliefs and label incongruent or other information as false. Although extant literature has mainly regarded the weaponization of Fake News as a (radical) right-wing phenomenon (Egelhofer & Lecheler, 2019), our findings indicate that (at least on the citizen level) people on both the left- and right-wing of the political spectrum are likely to view the “other side” as sources of misinformation. Yet, people on the right-wing are more likely to associate falsehoods with conspiracies—which supports the notion that the “Fake News” label and perceiving established information outlets as an enemy of the people are more likely to be found among people clinging on to right-wing (populist) worldviews (e.g., Fawzi, 2019).

Our study has limitations that can be addressed in future research. First of all, we rely on rather crude measures of self-perceived mis- and disinformation, causes, and sources. We did not differentiate between specific political motivations, or a variety of online and offline sources. In addition, we did not take into account that the general term “false information” may mean different things to different people, which could bias our findings. Although it reaches beyond the scope of this research note, it could be argued that a more encompassing perception of false information (i.e., information that is completely out of touch with reality) corresponds to more far-reaching perceived causes than false information that gets the context wrong. We suggest future research to (qualitatively) explore individuals’ unprimed perceptions of false information and connect these to perceived causes and consequences. Second, we collected data in just one country in the context of heightened concerns on mis- and disinformation related to a pandemic that dominated news agendas (Nielsen et al., 2020). Future research needs to assess the transferability to other settings in terms of timing, issue salience, and national context.

Here, we note that, according to the Reuters Digital News Report (2021), news users in the Netherlands are—just like news users in many Northern and Western European countries—substantially less concerned about false information than news users in more polarized countries (i.e., the United Kingdom and the United States) and Southern European countries (i.e., Spain and Greece). Considering that more polarized bi-partisan settings, such as the United States, may offer a more favorable opportunity structure for the perception of ideological biases, we mainly regard our findings as transferrable to similar European countries with comparable levels of misinformation beliefs and news trust. Yet, we believe that the differentiation between different causes is robust across settings: Although the level of misinformation beliefs may differ depending on indicators as overall trust, polarization, and press freedom, the distinction between unmotivated and deceptive causes should hold across settings.

Funding

This study was made possible with financial support of the Digital Society initiative of the Amsterdam School of Communication Research.

Conflicts of interest: None declared.

References

Dr. Michael Hameleers (PhD, University of Amsterdam) is Assistant Professor in Political Communication at the Amsterdam School of Communication Research (ASCoR), Amsterdam, The Netherlands. His research interests include populism, disinformation, framing, (affective) polarization, and the role of social identity in media effects.

Dr. Anna Brosius (PhD, University of Amsterdam), Amsterdam School of Communication Research (ASCoR), University of Amsterdam.