Conspiracy Thinking in Europe and America: A Comparative Study

Abstract

What explains conspiracy thinking in Europe and America? This is the first and largest comparative study of conspiracy thinking to date, presenting findings using a representative sample of 11,523 respondents in nine countries. First, it shows that the overall level of conspiracy thinking in Europe is equal to or slightly lower than the United States, contradicting the notion that conspiracy theories is an especially American phenomenon. Second, people more inclined to conspiracy thinking position themselves towards the right of the political spectrum, engage in magical thinking, feel distrust towards public officials and reject the political system. Finally, we find that – surprisingly – the country context in which respondents reside has hardly any effect as predictor of levels of conspiracy thinking or as a moderator of individual-level determinants. Heterogeneity in conspiratorial thinking seems to be largely a function of individual traits.

Keywords

conspiracy theory conspiracy thinking misinformation Europe United States

Introduction

From JFK to the X-Files to 9/11, and now with a president who famously launched his political career with the ‘birther’ movement (Drochon, 2018), conspiracy theories are often perceived to be an especially American affliction, and this alleged exceptionalism has been reinforced by the field of cultural and American studies (Fenster, 2008; Knight, 2001, 2002; Melley, 2000). But Europeans, at least going back to the Protocols of the Elders of Zion, have their own tradition. So, who is more likely to believe in conspiracy theories?

With the current rise of fake news and misinformation, the study of conspiracy theories is gaining increased attention. And far from being an amusing peccadillo, belief in conspiracy theories is linked to health risks, negative attitudes and prejudices towards groups in society, political radicalization, political violence, political disengagement and diminishing support for public policies, such as measures against climate change and vaccination (Douglas et al., 2019; van Prooijen and Douglas, 2018). Following Smallpage et al. (2020), we define a Conspiracy Theory as ‘a proposed explanation of events or circumstances (past, present or future) which cites as the primary cause a conspiracy. Conspiracy theories could be either true or false, and, in most circumstances, contradict the proclamations of epistemological authorities’. Conspiracy Beliefs, therefore, are those specific conspiracy theories that individuals will say are true. Finally, Conspiracy Thinking, what is sometimes also referred to as ‘conspiracy ideation’ or ‘conspiracy mentality’, is a stable predisposition that drives individuals to see events as the product of a conspiracy.

People are not equally prone to conspiracy thinking, so what explains people’s propensity to conspiracy thinking? Most work on conspiracy beliefs examines a single context, primarily the United States (Enders and Smallpage, 2018, for non-US single country studies, see, for example, Hogg et al., 2017; Jasinskaja-Lahti and Jetten, 2019; Krouwel et al., 2017; Mancosu et al., 2017; Siddiqui, 2020; Swami, 2012). However, even when studies are conducted in different contexts, many are based on different concepts, operationalizations and analytical approaches, thereby making systematic comparison problematic (Nyhan and Zeitzoff, 2018 is the exception). The lack of comparative work raises important questions: To what extent are findings context-specific, and how does context affect people’s belief in conspiracy theories? To develop a general theory of conspiracy thinking, we need work examining conspiracy belief in different contexts, not only to verify the general applicability of findings but also to formulate and test theories on how context characteristics interact with individuals’ characteristics affecting their conspiracy thinking (Douglas et al., 2019; van Prooijen and Douglas, 2018).

In this article, we present the first cross-national representative sample survey of conspiracy beliefs across eight European countries (Sweden, France, Germany, Great Britain, Portugal, Italy, Poland, and Hungary) and the United States, analysing data of 11,523 respondents collected by YouGov in August 2018. This study contributes to the field in various ways. To begin, it is the first and largest integrated comparative study of conspiracy beliefs in the United States and Europe to date. Second, it assesses the ways in which these beliefs are interconnected and consequently the structure of conspiracy belief systems. Third, it tests for the first time the universality of individual-level predictors explaining conspiracy thinking, thereby testing the external validity of findings of previous single country studies. Finally, it offers the first exploration of the importance of country-level and cross-level predictors for explaining conspiracy thinking. Taken together, these contributions are a fundamental step towards developing a broader theory of conspiracy thinking.

We find that to explain people’s propensity to conspiracy thinking, we first have to look at individual-level predictors as the effects of country-level and cross-level predictors are surprisingly limited. As such, the propensity to conspiracy thinking almost appears like an anthropological constant. This appears to replicate psychological work that takes this view as its premise, but contradicts a basic ontological assumption of the social sciences that people’s attitudes and behaviour are affected by the context that they live in. We demonstrate people’s proneness to conspiracy thinking to be particularly conditional on ideological position, magical thinking, trust in public figures and attitude towards the political system. Moreover, contrary to both the popular view and the one professed by American and cultural studies, we find that these eight European countries have similar or slightly lower overall levels of conspiracy thinking compared to the United States: Europeans and Americans are as likely to believe in conspiracy theories as each other.

First, we discuss the literature on conspiracy thinking and individual-level, country-level and cross-level predictors, which leads to our hypotheses. Second, we present our data, analytical design and operationalization of our variables. We then present and discuss the results, and end by drawing conclusions and implications for further study of conspiracy thinking.

Who Believes in Conspiracy Theories?

What explains people’s tendency towards conspiracy thinking in Europe and America? Conspiracy thinking is a property that varies from individual to individual. The notion of conspiracy thinking is based on the finding that people who already believe in a particular conspiracy theory are more likely to believe in others that are not necessarily related or may even be contradictory (Douglas et al., 2019; Swami et al., 2011; Swami and Furnham, 2012; Wood et al., 2012).

Numerous studies have been conducted identifying characteristics and attitudes that explain belief in specific conspiracy theories and/or a general tendency towards conspiracy thinking (Douglas et al., 2019). These studies have looked at personality traits (e.g. Darwin et al., 2011; Swami et al., 2011), sociodemographic factors (e.g. Freeman and Bentall, 2017; Hogg et al., 2017; Stempel et al., 2007), political ideology (e.g. Galliford and Furnham, 2017; Krouwel et al., 2017; Uscinski and Parent, 2014; van Prooijen et al., 2015), religious and supernatural beliefs (e.g. Galliford and Furnham, 2017; Jasinskaja-Lahti and Jetten, 2019; Oliver and Wood, 2014), media exposure (e.g. Einstein and Glick, 2015; Mulligan and Habel, 2013; Warner and Neville-Shepard, 2014) and politico-economic exclusion (e.g. Drochon, 2018; Hofstadter, 1996; Uscinski and Parent, 2014). We will test various forms of these theories in a comparative perspective.

First of all, we look at the role of sociodemographic variables. Some studies find that conspiracy believers are more likely to be male (Freeman and Bentall, 2017; Galliford and Furnham, 2017; Hogg et al., 2017), other studies claim that conspiracy believers are more likely to be female (Mancosu et al., 2017; Stempel et al., 2007), and still others find that gender does not matter at all (Enders and Smallpage, 2019; Uscinski and Parent, 2014). Research suggests that those who are unmarried (Freeman and Bentall, 2017), unemployed (Freeman and Bentall, 2017; Hogg et al., 2017), younger (Galliford and Furnham, 2017; Goertzel, 1994; Stempel et al., 2007) and/or poorly educated are more likely to engage in conspiracy thinking (Bogart and Bird, 2003; Douglas et al., 2016; Freeman and Bentall, 2017; Galliford and Furnham, 2017; Hollander, 2018; Mancosu et al., 2017; Oliver and Wood, 2014; Uscinski and Parent, 2014; van Prooijen, 2017), although again not all studies find a link between education and conspiracy thinking (Enders and Smallpage, 2019; Galliford and Furnham, 2017; Goertzel, 1994).

How gender, employment, marital status and age are linked to conspiracy thinking is somewhat undertheorized. Substantially more theory can be found on how education is linked to conspiracy thinking. First of all, people with low levels of education might have lower analytical skills (van Prooijen, 2017). Analytical thinking is negatively related to belief in conspiracy theories, as analytical skills enable people to evaluate conspiracy theories more critically, in particular conspiracy theories that defy logic and are not supported by evidence (Swami et al., 2014). An alternative explanation is that those with lower educational qualifications would be more inclined to attribute agency and intentionality in circumstances when it does not exist and thus have a stronger tendency to conspiracy thinking (Douglas et al., 2016). Finally, highly educated people feel more in control of their environment and are therefore less likely to engage in conspiracy thinking (van Prooijen, 2017). On the basis of these findings, we formulate the following hypothesis:

Education Hypothesis (H1): We expect conspiracy thinking to be dependent on citizens’ level of education, whereby the higher the education level, the lower people’s propensity to conspiracy thinking.

After examining the link between conspiracy thinking and various sociodemographic variables, we look at how people’s attitudes matter for conspiracy thinking. Conspiracy thinking is often thought to be connected to people’s ideology, although not all studies find that link (Dieguez et al., 2015). Several studies have found conspiracy thinking to be stronger at the political extremes (Krouwel et al., 2017; van Prooijen et al., 2015). Although all kinds of people engage in conspiracy thinking, conspiracy thinking is according to some stronger on the far right than the far left (Galliford and Furnham, 2017; Krouwel et al., 2017; van der Linden et al. 2020; van Prooijen et al., 2015). The reason for this might be that people on the right are more likely to have personality predispositions that stimulate conspiracy thinking, such as low tolerance for uncertainty and a desire for simple solutions and order (Jost et al., 2003; van Prooijen et al., 2015). The link between ideological extremity and conspiracy thinking is contested by some US studies that argue that conspiracy thinking is stronger among individuals who consider themselves independents or associate with third parties (non-Democrats and Republicans) (Uscinski et al., 2016; Uscinski and Parent, 2014). Subsequently, we formulate the following hypotheses:

Right-wing Hypothesis (H2): We expect people’s propensity to conspiracy thinking to be conditional on their ideological position, whereby the more people perceive themselves as positioned towards the right of the political spectrum, the higher their propensity to conspiracy thinking.

Ideological Extremity Hypothesis (H3): We expect people’s propensity to conspiracy thinking to be conditional on the extremity of their ideological position, whereby people who perceive themselves as positioned on the ideological extremes of the political spectrum have a higher propensity to conspiracy thinking than respondents who perceive themselves in the centre of the political spectrum.

Conspiracy thinking has been found to be associated with supernatural beliefs such as religious beliefs (e.g. Galliford and Furnham, 2017; Jasinskaja-Lahti and Jetten, 2019; Mancosu et al., 2017; Oliver and Wood, 2014) and belief in the paranormal (e.g. Swami et al., 2011; Swami and Furnham, 2012), but there are exceptions (Freeman and Bentall, 2017). People who hold supernatural beliefs tend to be more superstitious and conspiracy thinking shows similarities with belief in an entity that is all-powerful (Jasinskaja-Lahti and Jetten, 2019). Conspiracy thinking is therefore associated with so-called magical thinking, which is the willingness to believe in ‘unseen, intentional and malevolent forces to explain an event, instead of directly seen or observable ones’ (Oliver and Wood, 2014: 953). Consequently, we formulate the following hypothesis:

Magical Thinking Hypothesis (H4): We expect people’s propensity to conspiracy thinking to be conditional on their degree of magical thinking, whereby people who have a high level of magical thinking have a greater tendency to conspiracy thinking.

It is also argued that conspiracy thinking is associated with people who feel powerless and excluded (Drochon, 2018; Hofstadter, 1996; Uscinski and Parent, 2014); however, others dispute this view and argue that any ordinary citizen can become a conspiracy theorist (e.g. Radnitz and Underwood, 2017). Uscinski and Parent (2014) argue that in particular people who do not feel that their interests are represented within the political system are likely to believe in conspiracy theories. Consequently, we formulate the following hypothesis:

Representation Within the Political System Hypothesis: We expect people’s tendency to conspiracy thinking to be dependent on the extent that they feel represented within the political system, whereby people who do not feel represented within the political system are more inclined to conspiracy thinking than people who do feel represented within the political system.

Another indicator of political exclusion than not feeling represented within the political system is overall distrust in political institutions and their representatives. Various studies have found political distrust to be associated with a stronger propensity to conspiracy thinking (Imhoff and Bruder, 2014) – conspiracy theorists tend to reject the ‘official line’ (Wood et al., 2012) – and political trust with a lower propensity to conspiracy thinking (Einstein and Glick, 2015; Jasinskaja-Lahti and Jetten, 2019; Swami et al., 2010, 2011), although not all studies agree (Mancosu et al., 2017). We formulate the following hypothesis:

Institutional Trust Hypothesis: We expect people’s tendency to conspiracy thinking to be dependent on their level of institutional trust, whereby people who have low levels of institutional trust are more inclined to conspiracy thinking than people who have high levels of institutional trust.¹

Drochon (2018) argues that it is not about representation within the political system itself, but that conspiracy theorists reject the political system as a whole. We formulate the following hypothesis:

Political System Rejection Hypothesis: We expect people’s tendency to conspiracy thinking to be dependent on their attitudes towards the political system, whereby people who reject the political system as a whole are more inclined to conspiracy thinking than those who do not reject the political system.²

Drochon (2018) also argues that we should not solely look at political exclusion, but also economic exclusion. There is work linking economic insecurity to belief in conspiracy theories (Goertzel, 1994), but findings are inconclusive (Hollander, 2018). We formulate the following hypothesis:

Economic Insecurity Hypothesis: We expect people’s tendency to conspiracy thinking to be affected by their economic security, whereby people who feel insecure about their economic situation have a higher propensity to conspiracy thinking.

Various studies find that the media also play an important role in conspiracy thinking, as exposure to media messages propagating conspiracy theories leads to greater belief in conspiracy theories not only in the short term (Einstein and Glick, 2015; Mulligan and Habel, 2013; Warner and Neville-Shepard, 2014), but in the long term too (Kim and Cao, 2016). Hollanders (2018) did not find an effect of news exposure but did find that exposure to Fox News is positively related to belief in conspiracy theories; he therefore argues against the use of generic media measures in studies. In particular, studies find that exposure to nonmainstream media that are more likely to contain misinformation, such as blogs, tabloids and YouTube news channels, stimulates conspiracy thinking (Stempel et al., 2007).

Nonmainstream News Hypothesis: We expect people’s media use to affect their tendency to conspiracy thinking, the more people expose themselves to nonmainstream media news, the more prone they are to conspiracy thinking.

We know that people respond differently to similar phenomena in different contexts and thus this raises the question whether people’s proneness to conspiracy thinking is context dependent. We do not assume that people across countries are intrinsically different, but that the way in which their orientations and motivations can manifest themselves differ. For social scientists including political scientists, context is an important explanatory factor for studying political attitudes and behaviour across countries and regions. For instance, people’s level of political knowledge is not just a reflection of individual characteristics, such as their level of education and political interest, but also their context, as people’s access to high-quality information about political processes is dependent on the their media environment (Granberg and Holmberg, 1988). Another example is that a voter’s propensity to vote cannot just be explained with individual characteristics such as sense of civic duty or political interest, but it is also important to take context characteristics into account, such as the electoral system and the presence of compulsory voting rules (Singh, 2011; Stockemer, 2017).

Due to the focus on single country studies, research on how context affects people’s conspiracy thinking has been limited and consequently the theory is somewhat underdeveloped. In the literature we find suggestions that context is important for conspiracy thinking (Uscinski and Parent, 2014) and that conspiracy thinking may be affected by the level of democratization or quality of democracy (Drochon, 2018; Smallpage, 2018), the electoral system (Uscinski and Parent, 2014) or the country’s economic wealth and economic inequality (Drochon, 2018). However, these authors are not clear on whether these effects are main effects,³ moderation effects⁴ or composition effects.⁵ Without formulating explicit hypotheses, the question of how context matters remains open. We therefore take the freedom to interpret these relationships as both main and moderation effects and explore whether they matter for conspiracy thinking. In addition, we will control in our models for composition effects, that is, countries matter because the constellation of their population is different.

Data

The data used in this article were collected as part of the Conspiracy and Democracy project at CRASSH, the University of Cambridge. They were collected online by YouGov using representative samples of the population between 14 August and 31 August 2018. The total number of observations in the data set is 11,523. The respondents reside in France (N = 1019), Sweden (N = 1009), Germany (N = 2065), Great Britain (N = 2171), Italy (N = 1012), Portugal (N = 1003), Poland (N = 1016), Hungary (N = 1005) and the United States (N = 1223). The European countries were selected from all European countries to maximize variation in terms of ‘deemed relevant’ independent factors, such as democratic quality, economic inequality and electoral system. This is a variant of the well-known ‘most different cases design’ that can be used in the absence of definitive information about their values on outcome variables (Przeworski and Teune, 1970). The United States was selected as a baseline category.

To measure conspiracy thinking, we offered respondents a number of conspiracy theory statements they could agree or disagree with. The question asked was, ‘Which, if any, of the following statements would you say are true? (Please select all that apply)’, and respondents could choose the answer options ‘Yes’ or ‘No’. See Table 1 for the surveyed conspiracy theory items; see Table A1 in the supplementary materials for descriptive statistics on these items. Table A1 shows that there is considerable variance in people’s belief in specific conspiracy theories between Europe and the United States, but also within Europe. Our data confirm the notion that conspiracy belief is widespread as the percentage of people believing in one or more conspiracy theories varies between 52.3 and 84.9 in the European countries studied. This set of conspiracy theory items constitutes the basis of our dependent variable Conspiracy Thinking (see ‘Methods’ section for more details). The selected conspiracy theory statements are generic in nature and are not strongly connected to the specific societal, economic or political systems of the examined countries; we therefore do not believe that our selection introduces bias in our estimation; however, one can never exclude this.

Table 1.

Survey Items Conspiracy Theories.

Item
A	Even though we live in what’s called a democracy a few people will always run things in this country anyway.
B	Secret plots that harm the nation are more common in this country than in other countries.
C	Humans have made contact with aliens and this fact has been deliberately hidden from the public.
D	The AIDS virus was created and spread around the world on purpose by a secret group or organization.
E	Regardless of who is officially in charge of governments and other organizations, there is a single group of people who secretly control events and rule the world together.
F	The idea of man-made global warming is a hoax that was invented to deceive people.
G	The government is deliberately hiding the truth about how many immigrants really live in this country.
H	The truth about the harmful effects of vaccines is being deliberately hidden from the public.
I	Muslim immigration to this country is part of a bigger plan to make Muslims a majority of this country’s population.
J	The official account of the Nazi Holocaust is a lie and the number of Jews killed by the Nazis during World War II has been exaggerated on purpose.

Note: Question asked, ‘Which, if any, of the following statements would you say are true? (Please select all that apply)’.

Methods

Constructing the Conspiracy Thinking Dependent Variable

For the construction of the dependent variable, we conduct a common factor analysis on the basis of tetrachoric correlations as the conspiracy theory items are binary. From the common factor analysis, we exclude item J as this item was not asked in all countries (see Table A1 in the supplementary materials). With the remaining eight items we test the underlying dimensionality of the factor space. Scree plots indicate unambiguously that there is no more than a single underlying factor in all countries studied, with the exception of Hungary. In Hungary one item in particular is responsible for a possible second factor and that is item G (see Table 1). After excluding this item from the analysis, we find in all countries that the dimensionality of the underlying factor space is one (see Figure 1 in the supplementary materials). On that basis a factor analysis has been conducted, extracting one factor, and factor scores for each individual on that factor have been calculated. We interpret this factor as generic conspiracy thinking and the factor scores can be interpreted as individual scores on the tendency of conspiracy thinking.

By conducting common factor analysis, we assume that part of the variance of the separate conspiracy theory items can be accounted for by their shared relevance for conspiracy thinking. This does not mean that all variance in the separate conspiracy items reflect only such a generic tendency of conspiracy thinking. In addition, each of the items has also unique meaning elements, which are not reflected in the common factor (which reflects a generic tendency of conspiracy thinking). With this approach we capture from the set of conspiracy theory items the part that expresses this generic shared conspiracy thinking and use this as a measurement instrument.⁶ This shared conspiracy thinking is captured within the factor loadings of the common factor analysis (see Figure 2 in the supplementary materials for the frequency distribution of conspiracy thinking per country). Robustness analyses showed that the dependent variable constructed is not sensitive to the items selected (see Table A8 and Figure 4 in the supplementary materials).

Method of Analysis

Given the hierarchical nature of the data – citizens (level 1) residing within countries (level 2) – the obvious choice would be to estimate a multilevel regression model. However, in this study, we opt for running fixed effects ordinary least squares (OLS) regression models on pooled data instead (see Allison, 2009). We do so as we are only dealing with nine countries and because such a small sample size at level 2 leads to biased estimates of second-level standard errors when using multilevel modelling (e.g. Bryan and Jenkins, 2016; Maas and Hox, 2004). Some might wonder why we do not conduct nine separate regression analyses and subsequently compare the coefficients derived from each of these separate regression analyses. This strategy, however, would lead to coefficients that are not comparable because of the differences of distribution of the dependent and independent variables in the separate analyses. Moreover, it would ignore Przeworski and Teune’s (1970) dictum that the use of proper names as explanatory factors (in this case, names of countries) is theoretically sterile and uninformative. Instead, we want to analyse the information about conspiracy thinking in all countries simultaneously. We structure the regression models in a block recursive manner to take into account the differences in assumed causal antecedence of individual variables. In our fixed effects OLS regression models, country dummies are inserted to control for the dependence of observations at the individual level. In all models the United States serves as the baseline. The statistical analyses are run on 6926 observations due to missing observations on our dependent and independent variables.

Operationalizing the Independent Variables

First, we include in all models a set of demographic control variables measuring the respondent’s sex, age, marital status, employment status and level of education. These demographic controls are operationalized as follows:

Male is a dummy variable measuring the sex of the respondent (1 = Male; 0 = Female).

Age is grouped and measured with a set of dummy variables, namely, Under 24, Between 25 and 35, Between 35 and 45, Between 45 and 55 and Over 55 (1 = Yes; 0 = No). Below 24 is the reference category.

Unmarried is a dummy variable measuring the respondent’s marital status (1 = Unmarried; 0 = Married).

Unemployed is a dummy variable measuring the respondent’s employment status (1 = Unemployed; 0 = Employed).

Respondents’ level of education is measured as a set of dummy variables, respectively, Primary Education, Secondary Education and Tertiary Education (1 = Yes; 0 = No). Primary is the reference category.

The other individual variables included in our models are operationalized as follows:

Respondents’ ideological position is measured with a set of dummy variables on the basis of people’s self-placement in the political spectrum, respectively, Left, Centre and Right (1 = Yes; 0 = No). Respondents who stated to be left wing or slightly left of centre were all labelled left. Respondents who stated to be right wing or slightly right of centre were all labelled right. Left is the reference category in all models.

Ideological Extremity is a dummy variable measuring whether people position themselves at the extremes of the political spectrum (1 = Ideologically Extreme; 0 = Not Ideologically Extreme).

Magical thinking is a dummy variable indicating whether respondents believe that what happens in life is decided by a ‘higher force’, such as God, fate or destiny (1 = Strong Magical Thinking; 0 = Weak or No Magical Thinking).

Economic Insecurity is a dummy variable indicating whether respondents feel pessimistic about their own financial situation when thinking about the next 12 months (1 = Feels Pessimistic; 0 = Does Not Feel Pessimistic).

Representation within Political System is measured as the extent that respondents feel represented in parliament. Representation within Political System is a dummy variable indicating the extent to which respondents feel that Parliament represents their interests (1 = Feels Badly Represented; 0 = Does Not Feel Badly Represented).

Trust in Public Figures is an additive scale that runs from 6 to 24, 6 indicating absolutely no trust in public figures and 24 indicating a very high level of trust in public figures. The scale is constructed out of 6 ordinal items that measure trust in senior officials of the European Union (EU), senior officials of the US government, government ministers in my country, religious leaders, military leaders in my country and people who run large companies. Mokken scale analysis, which is a stochastic cumulative scaling model for ordinal variables, was used to determine the scalability of the survey items (Mokken, 1971). The overall homogeneity of the scale is 0.48, which indicates medium scalability (see Table A3 in the supplementary materials for the results of the Mokken scale analysis).

Political System Rejection is a dummy variable indicating whether respondents reject the political system (1 = Rejects political system; 0 = Does not reject political system). Respondents received the value 1 when they stated that thinking about the political system in their country the statement ‘The system is broken and it would take a total change of system to put things right’ best reflected their view.

Respondents’ news sources are measured on the basis of a set of dummy variables indicating respondents’ use of this source at least two to three times a month to access news, respectively.

Newspaper (including online); Television, Radio, Social media, Non-associated News Website/Blog (not associated with a newspaper); Magazines, Podcasts, Email newsletters or RSS feeds (1 =Yes; 0 = No).

The country-level variables included in our models are operationalized as follows:

Quality of Democracy is measured using the EUI 2018 Democracy Index of which the scores are weighted averages to an excess of 60 questions of which most are expert assessments. The scores run between 0 and 10 and are rounded to two decimals.⁷ Score 0 represents a full autocracy and score 10 represents a full democracy.

Economic Inequality is measured using the most recent GINI index as reported by the World Bank,⁸ which is a measure of statistical dispersion representing the income or wealth distribution of a nation’s residents. The scores run between 0 and 100, where 0 represents perfect equality and 100 represents maximum inequality.

Proportional Representation is measured as a dummy variable (1 = Present; 0 = Absent). The variable is based on data provided by IDEA Electoral Systems Design Database.⁹

Gross Domestic Product is measured nominally expressed in billions of dollars on the basis of the October 2018 International Monetary Fund World Economic Outlook.¹⁰

Results

In order to construct the dependent variable Conspiracy Thinking, we conducted common factor analysis on the responses to conspiracy theory items (see ‘Methods’ section for more details). As in all factor analyses, the variance of the conspiracy beliefs can be understood as emanating from three elements, namely, (1) generic conspiracy thinking (factors accounting for common variance); (2) item-specific beliefs that are not generic, but idiosyncratic; and (3) random noise. See Table 2 for the variance that generic conspiracy thinking explains of the total variance of these conspiracy theory items per country. The total variance explained by the first factor that reflects generic conspiracy thinking varies between 38.58% (Portugal) and 53.38% (Sweden). The unexplained variance reflects random noise and item-specific elements. The latter reflects, among other factors, the endorsement of specific conspiracy beliefs by the dominant socio-political discourse in societies’ contexts, which is at least in part the result of history, but also the extent to which politicians, political parties, media and other sources of reference in recent times endorse specific conspiracy beliefs. Table 2 also displays the highest and lowest communality scores. So, for instance, the lowest communality in France belongs to item B and it reflects the extent to which individual variation in responses to item B can be understood as emanating from generic conspiracy thinking.

Table 2.

Explained Variance and Lowest/Highest Communality Scores Conspiracy Theory Items.

Country	Explained variance expressed in %	Lowest Communality	Item	Highest Communality	Item
France	42.178	0.302	B	0.418	D
Sweden	53.384	0.434	C	0.630	D
Germany	44.091	0.193	B	0.460	E
Great Britain	50.126	0.339	A	0.521	F
Italy	44.739	0.266	A/B	0.578	D
Portugal	38.584	0.247	B	0.448	F
Poland	42.725	0.189	A	0.525	D
Hungary	40.515	0.211	A	0.561	D
United States	45.892	0.310	C	0.636	I

Method: Common Factor Analysis extracting one factor using tetrachoric correlations. Extraction method: Unweighted Least Squares. Second column presents the variance of the total variance of the separate items explained by the first factor. Third and fourth columns display the lowest communality score and the item it belongs to. The fifth and sixth columns display the highest communality score and the item it belongs to. Communality scores should be interpreted as the extent to which individual variation in responses to that item can be understood as emanating from generic conspiracy thinking.

In this study, we explain people’s tendency to conspiracy thinking across nine countries on the basis of individual and context characteristics. First of all, contrary to our expectations the variance that can be accounted for by proper names of countries is low, namely 2% (see Table A6 in the supplementary materials). This justifies our choice to focus in our presentation of results on individual-level predictors of conspiracy thinking. Table 3 displays the effect of individual characteristics on conspiracy thinking. Model 1 in Table 3 starts out with the set of sociodemographic variables and shows that male, unemployed and older people have an increased tendency to conspiracy thinking. We find no effect for secondary education. We do find an effect for tertiary education, but this effect disappears when controlling for sample composition effects, see Model 3. Therefore, and contrary to much of the literature, we reject our Education Hypothesis. This might lead us to reflect upon what type of skills higher education offers in different national contexts, and how strongly analytical skills, agency and empowerment are linked to higher education, which seem to be more speculative hypotheses than demonstrable theses. It might also make us reflect further on the link education entertains with race, wealth, socialization and selection effects (Uscinski and Parent, 2014). The sociodemographic variables only explain 1% of the variance of generic conspiracy thinking.

Table 3.

Effects of Individual Characteristics on General Conspiracy Thinking.

Individual Characteristics	Model 1	Model 2	Model 3	Model 4
Male	0.085** (0.021)	0.088** (0.021)	0.092** (0.021)	0.046
Unemployed	0.048* (0.024)	0.033 (0.023)	0.051* (0.024)	0.024
Unmarried	−0.0420 (0.022)	−0.026 (0.021)	−0.033 (0.021)	−0.017
Secondary	−0.020 (0.049)	0.000 (0.04)	0.014 (0.046)	0.006
Tertiary	−0.139** (0.047)	−0.093* (0.045)	−0.075 (0.046)	−0.037
Between 25 and 35	0.052 (0.047)	−0.004 (0.045)	0.011 (0.045)	0.004
Between 35 and 45	0.125** (0.047)	0.062 (0.045)	0.074 (0.044)	0.028
Between 45 and 55	0.134** (0.047)	0.081° (0.045)	0.079° (0.044)	0.031
Over 55	0.143** (0.043)	0.110** (0.041)	0.102** (0.041)	0.050
Ideological Extremity		0.048° (0.025)	0.047° (0.025)	0.020
Centre		0.078** (0.026)	0.084** (0.027)	0.036
Right-wing		0.302** (0.024)	0.310** (0.025)	0.144
Magical Thinking		0.261** (0.024)	0.255** (0.025)	0.112
Trust in Public Figures		−0.034** (0.003)	−0.036** (0.003)	−0.131
Political System Rejection		0.226** (0.023)	0.227** (0.023)	0.110
Feeling Represented in Parliament		0.069** (0.023)	0.069** (0.024)	0.036
Economic Insecurity		0.080** (0.024)	0.086** (0.024)	0.037
Newspapers		−0.099** (0.022)	−0.078** (0.021)	−0.038
Non-associated Website/Blog		0.080** (0.022)	0.091** (0.020)	0.043
Social Media		0.109** (0.021)	0.117** (0.021)	0.058
Sweden			−0.231** (0.047)	−0.067
France			−0.079(0.048)	−0.023
Germany			−0.231** (0.038)	−0.096
Portugal			−0.035(0.043)	−0.011
Poland			−0.420** (0.043)	−0.131
Great Britain			−0.305** (0.072)	−0.048
Italy			−0.176** (0.044)	−0.056
Hungary			−0.360** (0.044)	−0.115
Constant	0.787** (0.063)	0.850** (0.076)	1.023** (0.086)
Adjusted R²	0.010	0.105	0.126
AIC	18,013.331	17,321.171	17,163.948

AIC: Akaike information criterion.

N = 6926 Method is Fixed Effects OLS Regression Models. Table displays for Models 1 to 3 unstandardized regression coefficients. The standard errors are in parentheses. For age under 24 is the reference category. For education primary is the reference category. For ideological position leftwing is the reference category. For the country dummies the United States is the reference category. Model 4 displays the x-standardized regression coefficients of Model 3, which means the x-variables expressed in standard deviations and the y-variable in its original units. X-standardized regression coefficients show the relative importance of the independent variables.

°Significant at 0.10; *significant at 0.05; **significant at 0.01.

Model 2 contains in addition to the sociodemographic variables the hypothesized individual-level predictors of generic conspiracy thinking. Adding these individual–level predictors significantly improves the fit of the model, which explains 10.5% of the variance of generic conspiracy thinking. The results show that people’s ideological position is relevant for people’s tendency to conspiracy thinking. The more people position themselves ideologically towards the right of the political spectrum, the stronger their inclination to conspiracy thinking. The variable Ideological Extremity is only significant at 0.10 level. We find that people’s propensity to conspiracy thinking is indeed conditional on the extremity of their ideological position, but more so for people who position themselves on the right side of the political spectrum. We find some evidence for our Ideological Extremity Hypothesis. Our finding is in line with recent work that points to ideological asymmetry in conspiracy thinking, that is, right-wing extremists are more likely to engage in conspiracy thinking than left-wing extremists, because right-wing extremists have personality predispositions, such as low tolerance for uncertainty and a desire for simple solutions, that encourage conspiracy thinking (van der Linden et al., 2020). In addition to finding people’s propensity to conspiracy thinking to be linked to their ideological position, we find conspiracy thinking to be higher for people who score high on magical thinking. The results support our Magical Thinking Hypothesis.

People tend to have a higher propensity to conspiracy thinking when they have less trust in public officials, feel pessimistic about their own financial situation, feel not represented by parliament and reject the political system. We thus find support for our Political Distrust, Economic Insecurity, Representation within the Political System and Political System Rejection Hypotheses.

Finally, Model 2 also examines the link between people’s media use and their propensity to conspiracy thinking. We find reading newspapers to be negatively associated with conspiracy thinking and using social media, websites and blogs that are not associated with newspapers to be positively associated with conspiracy thinking. We did not find any effect for the use of television, radio, magazines, podcasts and email letters or RSS feeds as news source on people’s propensity for conspiracy thinking. We did not incorporate these variables in our final model displayed here. We have checked that by omitting these variables the changes in regression coefficients of individual-level variables are not due to the changes in number of observations on which the analyses are run. See for these robustness tests Table A7 in the supplementary materials.

We know that people respond differently in specific contexts. When we enter the country dummies to control for the differences between countries (Model 3), we see that the coefficients of the country dummies for Sweden, Germany, Poland, Italy, Great Britain and Hungary are negative and significant. This indicates that people residing in these six countries have on average a slightly lower level of conspiracy thinking than the United States, which is the baseline for the model. No significant differences were found for Portugal and France; these countries have on average a similar level of conspiracy thinking to the United States. Including the country dummies improves the fit of our model and helps us explain 12.6% of the variance, but leaves us with 87.4% unexplained. This suggests that the independent variables evidently carry little explanatory power and thus that the role of demographic and socio-political attitudes identified in our model as well as literature as drivers of generic conspiracy thinking is limited.

To know the relative importance of the independent variables for explaining conspiracy thinking, we standardized the independent variables and not the dependent variable. The x-standardized regression coefficients are calculated on the basis of Model 3, which are displayed in Model 4. For example, the results show that 1 standard deviation increase in feeling rejected by the system on average results in an 0.110 increase in conspiracy thinking. The most important variables for explaining generalized conspiracy thinking in our model are Rightwing, Trust in Public Figures, Magical Thinking and System Rejection. We are confident that the findings presented are robust and not driven by a particular country (see for the results of our jackknife sample sensitivity analysis Figures A8 in the supplementary materials).

A further exploration of the main effects and the interactions between context and individual variables (i.e. moderation effects) did not deliver substantial effects, that is, they did not meet the criterion of at least explaining 1.5% of the 87.4% unexplained variance: the effects were so small in size that they are negligible. The following context variables (Quality of Democracy, Economic Inequality, Proportional Representation, Gross Domestic Product) and the following individual factors (Economic Insecurity, Feeling Represented in Parliament, Political System Rejection) were examined. For reasons of brevity, these analyses are not reported here, but they can be found in Table A9 in the supplementary materials.

Conclusion

This study examined what drives people’s tendency towards conspiracy thinking in Europe and the United States. We did so by analysing 2018 survey data of respondents living in Sweden, Germany, Poland, Italy and Hungary, Portugal, France, Great Britain and the United States using representative samples of the population. To our knowledge this is the largest integrated comparative study to date. This study shows that it is possible to study conspiracy thinking comparatively using a set of specific conspiracy theories. We found that the dominant structure of conspiracy beliefs consists of a single common factor, which we labelled conspiracy thinking. This finding has at least two implications. First, that it would be incorrect to consider individual and specific conspiracy theories as only relating to their specific content (e.g. as reflecting simply belief in global warming or belief in the existence of aliens). Second, these findings also imply that individuals’ subscription to individual and specific conspiracy beliefs cannot be reduced to only this generic tendency. This therefore leaves room for at least two complementary strands of further research: (1) determinants of and consequences of generic conspiracy thinking and (2) determinants of and consequences of specific individual conspiracy beliefs, after having controlled for their generic drivers.

At the individual level we found that people who are male, old, unemployed, positioned at the ideologically extremes, who do not feel represented by parliament, feel economically insecure and use for their news source blogs and nonmainstream social media instead of newspapers have on average a stronger tendency towards conspiracy thinking. We did not find an effect for education, marital status and other media sources. However, the most important predictors for conspiracy thinking are being rightwing, feeling distrust towards public officials, engaging in magical thinking and rejecting the political system. This comparative study thereby demonstrates the broader applicability of these factors already identified in some earlier (mostly single country) studies. When one relies on single country studies to replicate earlier work, one cannot rule out that (minor) methodological differences affect the generalizability of the claims made. Such problems do not afflict a comparative study as presented here. Our model – including all these individual-level predictors and controlling for country heterogeneity – only explains 12.6%, indicating that many other important predictors of conspiracy thinking were not incorporated. This suggests that the focus on socio-structural variables as the main explanatory factors is perhaps misplaced and that psychological and personality constructs need to be considered further.

The results in this study show surprisingly that only 2% of the variance lies at the country level, which tells us that to explain generic conspiracy thinking we must look at people’s characteristics more so than the characteristics of the country in which they reside. In addition, we find surprisingly that the country-level and cross-level predictors (i.e. moderation effects) suggested in the literature had no substantial effect. This unexpected finding can be explained in several ways. First, the finding can be interpreted to reflect that conspiracy thinking is a purely psychological phenomenon, not or little affected by context (i.e. by social, political and economic characteristics of a country). Second, it is also possible that country characteristics do affect people’s propensity to engage in conspiracy thinking but that the measurement instrument used to gauge conspiracy thinking is too abstract to reveal this. This is a likely possibility as all countries included in our sample are very similar in terms of economic wealth, value system and a shared interpretation of contemporary history. A future comparative study should therefore try to expand its scope beyond Western, wealthy and democratic countries. Third, country characteristics may moderate people’s tendency to conspiracy thinking, but other characteristics than the ones included in our analyses. Future research should try to model other elements of the political and societal context, such as for instance political culture and the quality of the information environment. A further study should develop new and more specific theories on how country differences matter. Therefore, although this study surprisingly finds limited variance at the country level and no substantial effect of the modelled characteristics, this cannot be regarded as definitive support for the notion that context is irrelevant as a moderating factor.

We did find, however, across the board country effects, namely that people residing in Sweden, Germany, Great Britain, Poland, Italy and Hungary score overall slightly lower on conspiracy thinking that people in the United States. The overall level of conspiracy thinking in France and Portugal was equal to that of the United States. These differences can potentially be explained in two ways. First of all, this can be the result of differences in language. When the conspiracy theory items were translated, the items may have received in some languages a somewhat stronger (or, conversely, weaker) meaning. Second, these across the board effects might derive from country-specific cultural or historical differences that are not captured by the statistical models estimated. However, to test whether these findings are indeed the result of cultural and historical differences, it would be helpful to have more than one moment of measurement per country.

By laying the groundwork towards a general theory of conspiracy thinking, this study had offered a first, tentative, step towards these endeavours.

Supplemental Material

sj-pdf-1-psx-10.1177_0032321720972616 – Supplemental material for Conspiracy Thinking in Europe and America: A Comparative Study

Supplemental material, sj-pdf-1-psx-10.1177_0032321720972616 for Conspiracy Thinking in Europe and America: A Comparative Study by Annemarie S Walter and Hugo Drochon in Political Studies

Footnotes

Acknowledgements

We would like to thank the Leverhulme Trust, through the ‘Conspiracy and Democracy’ research project at CRASSH, Cambridge, for funding the survey. Thanks also to Hugo Leal and to Joel Rogers de Waal of YouGov for helping design it and carry it out. We thank Professor Cees van der Eijk for his methodological advice, and the participants of the Comparative Research in Political Communication panel at the Fifth Conference of the International Journal of Press/Politics for their comments. Special thanks to Charles Pattie and to two anonymous reviewers for their insightful comments and queries that helped clarify and strengthen the piece.

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: This study was supported by Leverhulme Trust.

ORCID iD

Annemarie S Walter

Supplementary Information

Additional supplementary information may be found with the online version of this article.

Table A1: Percentage of People that Endorse a Specific Conspiracy Theory per Country.

Table A2: Number of Specific Conspiracy Theory Statements that People Believe in Expressed in Percentages per Country.

Table A3: Trust in Public Figures Scale.

Figure 1: Scree Plots Common Factor Analyses (Table 2 Manuscript).

Table A4: Overview Communalities Specific Conspiracy Theory Items People Believe In.

Figure 2: Frequency Distribution Conspiracy Thinking Per Country.

Table A5: Distribution Dependent Variable Conspiracy Thinking Per Country.

Table A6: Estimation Intraclass Correlation Coefficient.

Table A7: Robustness Test Using a Larger Sample.

Table A8: Jackknife Sample Sensitivity Analysis for Including/Excluding Specific Countries.

Figure 3: Visual Representation Table A6.

Table A9: Explaining Remaining Variance with Country-Level and Cross-Level Predictors.

Figure 4: Scree Plots Robustness Test Item Selection (Common Factor Analyses Jackknifing Items).

Notes

Author Biographies

Annemarie S Walter is Assistant Professor in Politics and International Relations at the University of Nottingham. She received her PhD from the University of Amsterdam in 2012. Her broad areas of research interests include election campaigns, party strategy and political scandals. Her work has been published in journals such as Political Psychology, Political Communication and Party Politics.

Hugo Drochon is Assistant Professor in Political Theory at the University of Nottingham. He received his PhD from the University of Cambridge in 2012 and was a Postdoctoral Research Fellow on the ‘Conspiracy and Democracy’ research project at CRASSH, Cambridge. He is the author of Nietzsche’s Great Politics (Princeton 2016).

References

Allison

(2009) Fixed Effects Regression Models. Los Angeles, CA: SAGE Publications.

Bogart

Bird

(2003) Exploring the Relationship of Conspiracy Beliefs About HIV/AIDS to Sexual Behaviors and Attitudes Among African-American Adults. Journal of National Medical Association 95 (1): 1057–1065.

Bryan

Jenkins

(2016) Multilevel Modelling of Country Effects: A Cautionary Tale. European Sociological Review 32 (1): 3–22.

Darwin

Nieves

Holmes

(2011) Belief in Conspiracy Theories. The Role of Paranormal Belief, Ideation and Schizotypy. Personality and Individual Differences 50: 1289–1293.

Dieguez

Wagner-Egger

Gauvrit

(2015) Nothing Happens by Accident, or Does It? A Low Prior for Randomness Does Not Explain Belief in Conspiracy Theories. Psychological Science 26 (11): 1762–1770.

Douglas

Sutton

Callan

, et al. (2016) Someone Is Pulling the Strings: Hypersensitive Agency Detection and Belief in Conspiracy Theories. Thinking and Reasoning 22: 57–77.

Douglas

Uscinski

Sutton

, et al. (2019) Understanding Conspiracy Theories. Political Psychology 40: 3–35.

Drochon

(2018) Who Believes in Conspiracy Theories in Great Britain and Europe? In: Uscinski

(ed.) Conspiracy Theories and the People Who Believe Them. New York: Oxford University Press, pp.337–346.

Einstein

Glick

(2015) Do I Think BLS Data are BS? The Consequences of Conspiracy Theories. Political Behavior 37: 679–701.

10.

Enders

Smallpage

(2018) Polls, Plots, and Party Politics: Conspiracy Theories in Contemporary America. In: Uscinski

(ed.) Conspiracy Theories and the People Who Believe Them. New York: Oxford University Press, pp.298–318.

11.

Enders

Smallpage

(2019) Who Are Conspiracy Theorists? A Comprehensive Approach to Explaining Conspiracy Beliefs. Social Science Quarterly 100 (6): 2017–2032.

12.

Fenster

(2008) Conspiracy Theories: Secrecy and Power in American Culture. Minneapolis, MN: University of Minnesota Press.

13.

Freeman

Bentall

(2017) The Concomitants of Conspiracy Concerns. Social Psychiatry and Psychiatric Epidemiology 52 (5): 595–604.

14.

Galliford

Furnham

(2017) Individual Difference Factors and Beliefs in Medical and Political Conspiracy Theories. Scandinavian Journal of Psychology 58: 422–428.

15.

Goertzel

(1994) Belief in Conspiracy Theories. Political Psychology 15 (4): 731–742.

16.

Granberg

Holmberg

(1988) European Monographs in Social Psychology. The Political System Matters: Social Psychology and Voting Behavior in Sweden and the United States Cambridge: Cambridge University Press.

17.

Hogg

Nkala

Dietrich

, et al. (2017) Conspiracy Beliefs and Knowledge about HIV Origins among Adolescents in Soweto, South Africa. PLoS ONE 12 (2): e0165087.

18.

Hofstadter

(1996) The Paranoid Style in American Politics and Other Essays. New York: Knopf.

19.

Hollander

(2018) Partisanship, Individual Differences, and News Media Exposure as Predictors of Conspiracy Beliefs. Journalism & Mass Communication Quarterly 95 (3): 691–713.

20.

Imhoff

Bruder

(2014) Speaking (Un-)Truth to Power: Conspiracy Mentality as a Generalised Political Attitude. European Journal of Personality 28 (1): 25–43.

21.

Jasinskaja-Lahti

Jetten

(2019) Unpacking the Relationship Between Religiosity and Conspiracy Beliefs in Australia. British Journal of Social Psychology 58: 938–954.

22.

Jost

Glaser

Kruglanski

, et al. (2003) Political Conservatism as Motivated Social Cognition. Psychological Bulletin 129 (3): 339–375.

23.

Kim

Cao

(2016) The Impact of Exposure to Media Messages Promoting Government Conspiracy Theories on Distrust in Government. Evidence from a Two-Stage Randomized Experiment. International Journal of Communication 10: 3808–3827.

24.

Knight

(2001) Conspiracy Culture: From Kennedy to the X Files. Abingdon: Routledge.

25.

Knight

(2002) Conspiracy Nation: The Politics of Paranoia in Postwar America. New York: New York University Press.

26.

Krouwel

Kutiyski

van Prooijen

J-W

, et al. (2017) Does Extreme Political Ideology Predict Conspiracy Beliefs, Economic Evaluations and Political Trust? Evidence From Sweden. Journal of Social and Political Psychology 5 (2): 435–462.

27.

Maas

CJM

Hox

(2004) Robustness Issues in Multilevel Regression Analysis. Statistica Neerlandica 58 (2): 127–137.

28.

Mancosu

Vassallo

Vezonni

(2017) Believing in Conspiracy Theories: Evidence from an Exploratory Analysis of Italian Survey Data. South European Society and Politics 22 (3): 327–344.

29.

Melley

(2000) Empire of Conspiracy: The Cultural Paranoia in Postwar America. Ithaca, NY: Cornell University Press.

30.

Mokken

(1971) A Theory of Procedures of Scale Analysis. The Hague: Mouton.

31.

Mulligan

Habel

(2013) The Implications of Fictional Media for Political Beliefs. American Politics Research 41 (1): 122–146.

32.

Nyhan

Zeitzoff

(2018) Conspiracy and Misperception Belief in the Middle East and North Africa. The Journal of Politics 80: 1400–1404.

33.

Oliver

Wood

(2014) Conspiracy Theories and the Paranoid Style(s) of Mass Opinion. American Journal of Political Science 58 (4): 952–966.

34.

Przeworski

Teune

(1970) The Logic of Comparative Social Inquiry. New York: Wiley Interscience.

35.

Radnitz

Underwood

(2017) Is Belief in Conspiracy Theories Pathological? A Survey Experiment on the Cognitive Roots of Extreme Suspicion. British Journal of Political Science 47 (1): 113–129.

36.

Siddiqui

(2020) Who Do You Believe? Political Parties and Conspiracy Theories in Pakistan. Party Politics 26: 107–119.

37.

Singh

(2011) How Compelling Is Compulsory Voting? A Multilevel Analysis of Turnout. Political Behavior 33: 95–111.

38.

Smallpage

(2018) Conspiracy Thinking, Tolerance, and Democracy. In: Uscinski

(ed.) Conspiracy Theories and the People Who Believe Them. New York: Oxford University Press, pp.187–198.

39.

Smallpage

Drochon

Uscinski

, et al. (2020) Who Are the Conspiracy Theorists? Demographics and Conspiracy Theories. In: Butter

Knight

(eds) Routledge Handbook of Conspiracy Theories. Abingdon: Routledge, pp.263–277.

40.

Stempel

Hargrove

Stempel

(2007) Media Use, Social Structure and Belief in 9/11 Conspiracy Theories. Journalism & Mass Communication Quarterly 84 (2): 353–372.

41.

Stockemer

(2017) What Affects Voter Turnout? A Review Article/Meta-Analysis of Aggregate Research. Government and Opposition 52 (4): 698–722.

42.

Swami

Camorro-Premuzic

Furnham

(2010) Unanswered Questions: A Preliminary Investigation of Personality and Individual Difference Predictors of 9/11 Conspiracist Beliefs. Applied Cognitive Psychology 24: 749–761.

43.

Swami

(2012) Social Psychological Origins of Conspiracy Theories: The Case of the Jewish Conspiracy Theory in Malaysia. Frontiers in Psychology 3: 280.

44.

Swami

Furnham

(2012) Examining Conspiracist Beliefs About the Disappearance of Amelia Earhart. The Journal of General Psychology 139: 244–259.

45.

Swami

Coles

Stieger

, et al. (2011) Conspiracist Ideation in Britain and Austria: Evidence of a Monological Belief System and Associations Between Individual Psychological Differences and Real-World and Fictitious Conspiracy Theories. British Journal of Psychology 102 (3): 443–463.

46.

Swami

Voracek

Stieger

, et al. (2014) Analytic Thinking Reduces Belief in Conspiracy Theories. Cognition 133: 572–585.

47.

Uscinski

Parent

(2014) American Conspiracy Theories. New York: Oxford University Press.

48.

Uscinski

Klofstadt

Atkinson

(2016) What Drives Conspirational Beliefs? The Role of Informational Cues and Predispositions. Political Research Quarterly 69 (1): 57–71.

49.

Van der Linden

Panagopoulos

Azevedo

Jost

(2020) The Paranoid Style in American Politics Revisited: An Ideological Asymmetry in Conspiratorial Thinking. Political Psychology. Epub ahead of print 24 June. DOI: 10.1111/pops.12681.

50.

van der Linden

Panagopoulos

Azevedo

, et al. (2020) The Paranoid Style in American Politics Revisited: An Ideological Asymmetry in Conspiratorial Thinking. Political Psychology. Epub ahead of print June. DOI: 10.1111/pops.12681.

51.

van Prooijen

(2017) Why Education Predicts Decreased Belief in Conspiracy Theories. Applied Cognitive Psychology 31: 50–58.

52.

van Prooijen

Douglas

(2018) Belief in Conspiracy Theories: Basic Principles of an Emerging Research Domain. European Journal of Social Psychology 48: 897–908.

53.

van Prooijen

Krouwel

APM

Pollet

(2015) Political Extremism Predicts Belief in Conspiracy Theories. Social Psychological and Personality Science 6 (5): 570–578.

54.

Warner

Neville-Shepard

(2014) Echoes of a Conspiracy: Birthers, Truthers, and the Cultivation of Extremism. Communication Quarterly 62 (1): 1–17.

55.

Wood

Douglas

Sutton

(2012) Dead and Alive: Beliefs in Contradictory Conspiracy Theories. Social Psychology and Personality Science 3: 767–773.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.54 MB