In the wake of the COVID-19 pandemic, conspiracy theories about the origin, impact, and spread of the SARS-CoV-2 virus have proliferated on social media platforms and the internet. Some of the most notable include the virus being a bioweapon designed in China, intentionally spread around the world for economic gains, of greatly exaggerated impact, not different from the common flu, or even a hoax entirely (Frankovic 2020). While the situation generated by the pandemic has provided a breeding ground for the development of new variants of conspiracy theories, conspiratorial belief has been around for quite some time and has made the object of extensive research (e.g., Sunstein & Vermeule 2009; the European Journal of Social Psychology 2018 special issue, introduced by van Prooijen & Douglas 2018). As previous studies have indicated, conspiracy belief increases especially in contexts of heightened collective stress, anxiety, uncertainty, confusion, and existential threat, such as situations of societal crisis, caused by natural disasters (floods, earthquakes, fires), epidemics, civil unrest, violence, terrorism, and war (van Prooijen & Douglas 2017).
Far from merely false and trivial, conspiracy theories can prove to be dangerous and harmful. They are consequential to people’s attitudes and behavior in key aspects of their life, such as health, safety, and relationships, and can have negative societal consequences. For instance, experimental studies report that participants who were exposed to conspiracy theories about global warming and climate change were less confident in the scientific consensus on climate change, less willing to participate in and sign a petition against global warming, donate to charity, volunteer (van der Linden 2015), or reduce their carbon footprint (Jolley & Douglas 2014a). Similarly, exposure to materials in support of anti-vaccine conspiracy theories has been associated with decreased intentions to vaccinate (Jolley & Douglas 2014b) and fears concerning the safety of vaccines inflamed by conspiracy theories were shown to contribute to declining vaccination rates (Salmon et al. 2005; Offit 2011). In the political realm, belief in conspiracy theories has been associated with the endorsement of political violence (Vegetti & Littvay 2021) and political extremism (Bartlett & Miller 2010; van Prooijen, Krouwel & Pollet 2015).
Given the implications of conspiracy theories for people’s behavior and attitudes, especially as they relate to health and safety, this article investigates COVID-19 conspiracy belief in a diverse sample of undergraduate and graduate students from Romanian universities. The article examines factors pertaining to respondents’ social position, education and analytical skills, information habits, and psychological state as correlates of belief in several conspiracy theories related to the origin, spread, and seriousness of the SARS-CoV-2 virus and accompanying COVID-19 disease. Conspiracy belief is measured using four dimensions adapted after Brotherton, French & Pickering (2013), namely government malfeasance (a belief that the government is acting against its citizens); malevolent global conspiracies (the spread of COVID-19 is controlled by a small global elite); personal well-being (COVID-19 is spread through 5G technologies on an unassuming public); and control of information (governments have withheld timely and accurate information on COVID-19 from the public).
Results indicate that belief in these conspiracy theories is stronger among students with lower social status, who primarily receive their information from TV and discussions with their peers, follow the news less frequently than before the pandemic, and score lower on analytical skills. These factors correlate with most of the conspiracies we investigate.
Conspiratorial belief and its predictors
Regardless of their subject matter, a common point for most conspiracy theories is the belief that (a group of) hostile, threatening, and powerful actors with evil intentions secretly collude in order to achieve malevolent or unlawful goals (Zonis & Joseph 1994; Bale 2007; Clarke 2002). These actors can include anyone from prominent individuals (political leaders, businessmen, such as Bill Gates or George Soros), to particular (out-) groups (e.g. freemasons, ethnic minority groups such as Jewish or Muslim), branches of industry (e.g. pharmaceutical), companies, governments or other organizations intent on dominating the world or causing harm. Other common characteristics are that conspiracy theories are not falsifiable or supported by factual evidence, and often explain their subject matter as a cover-up by powerful entities (Freeman & Bentall 2017), thus providing an enemy to be blamed for the occurrence of a negative event or deemed responsible for unanswered questions regarding different events.
As Bale (2007, 51) argues, one of the main appeals of conspiracy theories lies in the oversimplified explanations or narratives that they provide for generally complex cause-and-effect relationships, thereby helping people make sense of distressing events (Hofstadter 1966). This partly explains the surprisingly widespread popularity of conspiracy theories (Pipes 1997; Sunstein & Vermeule 2009). For instance, over half of Americans believe in at least one conspiracy (Oliver & Wood 2014), albeit the intensity of conspiratorial belief varies in the population. Moreover, people that hold one conspiracy belief are more likely to hold others on unrelated topics, even some that are mutually inconsistent or contradictory, which suggests that similar underlying psychological processes are at work (Goertzel 1994; Swami, Chamorro-Premuzic & Furnham 2010; Wood, Douglas & Sutton 2012). There is also evidence of a positive association between belief in conspiracy theories and intuitive thinking; people with stronger analytic thinking skills appear to be less prone to conspiratorial belief (Swami et al. 2014; Stanley et al. 2021). These findings form the basis of the perspective that views conspiracy belief as a form of motivated reasoning, general worldview and belief system, much like a political ideology (Goertzel 1994).
In a study of individual differences in conspiracist ideation, Brotherton, French & Pickering (2013) have identified five facets of conspiracist belief that strongly predict the endorsement of specific conspiracy theories. These are: a) government malfeasance, the belief that governments routinely commit secret acts against the security and interest of their own citizens; b) malevolent global conspiracies, which refer to the existence of a small group of elites that control important global events; c) extraterrestrial cover-up, or the idea that governments refuse to disclose evidence of alien existence; d) personal well-being, namely conspiracies referring to the spread of diseases and use of mind-controlling technologies on an unaware public, and e) control of information, in which organizations (such as governments, corporations or the media) suppress information from the public.
Who is more likely to endorse conspiracy theories? Based on previous research, a series of psychological, socio-demographic, and situational factors seem to predict conspiracy belief (Douglas et al. 2019). From a psychological perspective, belief in conspiracy theories appears to be rooted in negative emotions, such as anxiety (Grzesiak‐Feldman 2013), uncertainty (van Prooijen & Jostmann 2013), stress (Swami et al. 2016), feelings of powerlessness (Jolley & Douglas 2014a; van Prooijen 2017), and lack of control (Whitson & Galinsky 2008; Douglas, Sutton & Cichocka 2017), especially for people with an external locus of control (Hamsher, Geller & Rotter 1968). Through the simplified explanations they provide for a distressing event or threat, conspiracy theories offer people a mechanism to make sense of these events, to overcome their negative emotions, and regain a lost sense of order and control under conditions of uncertainty and fear (Hofstadter 1966; Robins & Post 1997; Kruglanski et al. 2006). Consequently, interventions that increase people’s sense of control over their environment reduce conspiracy beliefs (van Prooijen & Acker 2015).
In terms of socio-demographic factors, conspiracy belief appears to be more widespread among less educated (van Prooijen 2017; Oliver & Wood 2014; Douglas et al. 2016) and lower income individuals (Uscinski & Parent 2014; Freeman & Bentall 2017). A study by van Prooijen (2017) revealed that cognitive complexity, feelings of powerlessness, and subjective social class mediate the relationship between education level and conspiracy belief, with people with a high education level having stronger analytical thinking skills, feeling less powerless, and having a higher perceived social class. Members of racial and ethnic minority groups are also more likely to endorse conspiracy theories, especially with regards to the dominant group conspiring against them (Crocker et al. 1999; van Prooijen & van Dijk 2014; Davis, Wetherell & Henry 2018; van Prooijen, Staman & Krouwel 2018). Such beliefs were attributed to higher levels of perceived external threat, feelings of being undervalued and underprivileged, and lower social status.
Conspiracy theories also seem to be more appealing to people who rely on social media platforms (Enders et al. 2021; Stecula & Pickup 2021; Foley & Wagner 2020), a limited number of information sources, or do not have access to information available in the wider society (Sunstein & Vermeule 2009). People who consume news predominantly via social media have a higher likelihood of coming across misinformation and conspiracy theories, due to the low level of regulation of online channels compared to conventional media platforms (Vosoughi, Roy & Aral 2018). Moreover, people with weaker social ties, especially in the context of marginalization and poverty, are more likely to believe in conspiracy theories due to a higher sense of perceived threat (Freeman & Bentall 2017).
COVID-19 conspiracy theories and belief
The COVID-19 pandemic has been an event of massive scale, which has impacted peoples’ lives globally. With uncertainty and anxiety running high, especially in the beginning, the pandemic fits the mould of the complex and disrupting event that could potentially generate considerable conspiracy beliefs (van Prooijen & Douglas 2017). Similar surges in conspiracy belief were also reported during previous sanitary crises, such as the 1918-1919 Spanish flu pandemic or the 2009 H1N1 outbreak (Spinney 2017; Bangerter et al. 2012). Moreover, the fact that the pandemic originated and spread from China added to its potential to be seen as an instrument of foreign control and malevolence. Given the efforts of governments across the world to curb the spread of the SARS-CoV-2 virus through different types of protective measures, restrictions and recommendations, several studies investigated the behavioral implications of COVID-19 conspiracy beliefs, which have previously been linked to distrust toward health authorities and their recommendations.
A study conducted in England reported that conspiracy belief about the COVID-19 pandemic is associated with non-compliant and risky behaviors that go against government-issued guidelines, such as frequent hand washing, social distancing, or limiting the number of interactions with people outside the household (Freeman et al. 2020). Similar results regarding face mask wearing and social-distancing behaviors were reported in studies using samples from the United States (Hornik et al. 2021; Romer & Hall Jamieson 2020) or Croatia (Erceg, Ružojčić & Galić 2020). In Romania, people who endorse conspiracy theories perceived the social distancing measures imposed by the government as less adequate and reported lower levels of compliance with these measures (Maftei & Holman 2020), although Corbu et al. (2021) report higher compliance. Higher levels of conspiracy beliefs were also associated with less concern about the spread and seriousness of COVID-19 among Romanians (Achimescu, Sultănescu & Sultănescu 2021). Given that the virus is highly contagious, such results point to the practical implications of the public’s conspiracy beliefs for controlling the spread of the virus. What is more, COVID-19 conspiracy beliefs have been linked to reduced intentions to be vaccinated against COVID-19 in the future and negative attitudes towards vaccination science (Bertin, Nerra & Delouvée 2020; Romer & Hall Jamieson 2020). These results are particularly salient for efforts to control the virus, since vaccination is one of the surest ways of achieving this goal (Greenwood 2014).
Conspiracy theory believers’ increased propensity toward problematic behavior during the pandemic may be explained by the association between conspiracy belief and lower levels of trust in or even rejection of science (Lewandowsky, Cignac & Oberauer 2013; Lewandowsky, Oberauer & Cignac 2013) and lower trust in governmental and health institutions (Lutkenhaus, Jansz & Bouman 2019). This aspect was particularly salient during the pandemic, since government response to the rapidly evolving COVID-19 crisis was based on the expertise and recommendations of medical professionals, epidemiologists, and physicians. Moreover, many conspiracy theories surrounding the SARS-CoV-2 virus downplayed its seriousness, effects, or even denied its existence, thereby leading to a (false) sense of little to no perceived risk or threat to one’s health, which is associated with the degree to which people engage in protective behaviors, such as vaccination (Brewer et al. 2004). Indeed, people who believe that the pandemic is a hoax reported less compliant behaviors with official recommendations for the containment of the virus (Imhoff & Lamberty 2020). Like in the case of other conspiracy theories, people who engage in analytic thinking were less likely to believe COVID-19 specific theories, such as the pandemic being a hoax (Stanley et al. 2021; Erceg, Ružojčić & Galić 2020). These results support the importance of teaching analytic thinking skills in schools in order to equip children and youth with the mental tools necessary to grasp complex societal problems and critically assess the validity of information coming from different sources.
Results on the association between conspiracy belief and the use of limited or mainly online information sources (especially social media platforms) hold up for the case of COVID-19 conspiracies as well. A cross-national study of exposure to communication channels during the early stages of the pandemic (May-June 2020) conducted in 8 countries/regions across the world reveals that people who followed traditional media (such as television, radio, and newspapers) had lower levels of support for conspiracy theories, while those who used digital media and personal contacts as information sources had higher levels of support (De Coninck et al. 2021). Results differed across countries, with more susceptibility to conspiracy theories in countries with higher levels of societal polarization, partisan media, and weaker public media systems. A study from the UK also reported a positive association between the use of social media as a source of information and COVID-19 conspiracy beliefs, with people who endorsed conspiracy theories less likely to engage in health-protective behaviors against the SARS-CoV-2 virus. Then again, people who mainly used broadcast and print media were less likely to hold conspiracy beliefs, given the fact that these media sources were regulated and sanctioned in the event of spreading misinformation (Allington et al. 2021). In the US, in addition to social media use, conspiracy belief about the origin and prevention of COVID-19 was also explained in a two-wave panel study by the use of conservative media, such as Fox News (Romer & Hall Jamieson 2021). In Romania as well, people who use social media platforms as information sources on COVID-19 are more likely to endorse conspiracy theories on vaccines and vaccination (Buturoiu et al. 2021).
Support for COVID-19 conspiracy theories appears to be higher among the younger population (Freeman et al. 2020; Hornik et al. 2021; Allington et al. 2021), people with lower levels of income, perhaps due to lower levels of information (Romer & Hall Jamieson 2020; Hornik et al. 2021), lower levels of education (Romer & Hall Jamieson 2020; De Coninck et al. 2021; Tonković et al. 2021), and ethnic and racial minorities, likely due to higher levels of experienced threat (Freeman et al. 2020; Romer & Hall Jamieson 2020). Results about gender seem to be mixed, with some studies reporting no significant gender differences (Freeman et al. 2020; Tonković et al. 2021; Allington et al. 2021), while others reporting a higher level of support in the case of men (Cassese, Farhart & Miller 2020) or women (Pizarro et al. 2020).
A study conducted in May 2020 on a Romanian sample found a positive relationship between COVID-19 conspiracy belief and age (with older people more likely to support conspiracy theories), as well as education level (being a high school or a college graduate predicts increased beliefs in conspiracy theories), while reporting no significant differences between men and women (Stoica & Umbreș 2021). The results regarding age are particularly concerning, since older individuals have an increased risk of severe illness from the virus and a higher likelihood of losing their life as a consequence (CDC 2020). The unexpected association to education, which runs counter to results from other countries, was explained by the two researchers through the particularities of the Romanian national context, in which trust in government is low, especially among the highly educated, who thus might have questioned the official line of communication related to the pandemic. Moreover, especially in the beginning of the crisis, several public figures and various pundits who endorsed ideas such as the seriousness of the disease being exaggerated or even a hoax, received quite high coverage in the Romanian media. Therefore, the inconsistent messages that the population received contributed to a climate of uncertainty, which favored the spread of conspiracy beliefs. These results are in line with previous studies showing that fragmented public health messaging from political leaders translates into public responses divided along partisan lines (Singer et al. 2020).
Research design and measures
In this study, we were interested in understanding what predicts university students’ belief in COVID-19 related conspiracies. Although many studies had students as their research population, ours is different in two ways. First, our data come from a very large sample of university students that, although not probabilistic, included a considerable degree of diversity in terms of demographics, place of study, year, and program of studies, thus increasing the representativeness of the data. Second, data were collected early in the pandemic, when levels of uncertainty and concern were high, and the information the public received was often controversial, thus providing a particularly fertile ground for the spread of conspiracy theories.
We collected the data in an online survey conducted among university students between the 26th of April and the 30th of May 2020. The study received the approval of the Ethics Committee of the “Lucian Blaga” University of Sibiu. The period of data collection coincided to a certain extent with the national lockdown introduced in Romania between the 24th of March and the 14th of May 2020 in response to the COVID-19 crisis. On the 11th of March 2020, in-person classes were discontinued in all education institutions and various forms of emergency online teaching were adopted across Romanian schools. At the time of data collection, classes were still held online.
The authors distributed the survey on social media platforms and students’ groups, as well as emailed a large part of Romanian universities asking for support in distributing the survey to their students. The questionnaire was filled in by 2026 under- and post-graduate students enrolled in Romanian universities. This response rate indicates that an online survey is an appropriate instrument of data collection for this population, as it fits students’ digital behavior. In terms of the location of study, 61% of respondents study in Sibiu, 10% in Bucharest, 9% in Brașov, 5% in Cluj-Napoca, 4% in Alba Iulia, 4% in Iași, and 3% in Oradea. Smaller percentages of the participants study at universities located in Timișoara, Craiova, Galați, Bacău, Targu Mureș, and Suceava. From the students who participated to the study, 73% were women, 64% resided in urban areas, and 90% were undergraduate students. In terms of the year of study, 36% were first year students, 27% second year, 24% third year, and the rest were fourth or higher year. We notice an overrepresentation of women students in the sample, given that they represent 59% of the population of university students (Romanian Ministry of Research 2017).
The dependent variables measure respondents’ level of agreement with six statements about the nature and the origin of COVID-19. The statements were constructed following the dimensions proposed in Brotherton, French & Pickering (2013) that define conspiratorial belief, namely government malfeasance, malevolent global conspiracies, personal well‐being, and control of information. They were formulated for this specific case after observing the frequency of claims related to the origin and nature of COVID-19 in the public discourse at the time, in both mass-media and social media channels of large circulation. The statements read as follows: COVID-19 is a conspiracy of political elites to restrict citizens’ rights and control them (government malfeasance); COVID-19 is a conspiracy of some interest groups that seek financial gain (malevolent global conspiracies); COVID-19 is used as a weapon in the fight of the superpowers for global supremacy (malevolent global conspiracies), COVID-19 was created by scientists (personal well‐being); COVID-19 is spread through 5G communication networks (personal well‐being); COVID-19 is not as dangerous as it is presented to the public (control of information).
Respondents indicated their level of agreement with each of the statements on 10-point scales, where 1 indicates their complete disagreement and 10 their complete agreement with the statement. Descriptives for each of these variables are shown in Table 1.
Table 1: Descriptives for dependent variables (PDF)
The distribution of these variables shows variations in the levels of agreement with the six statements, from strongest agreement with the claim that COVID-19 was created by scientists (mean = 6.09) followed by agreement with the claim that it is not as dangerous as presented (mean = 5.49), to weakest agreement with the statement that it is spread through 5G technology (mean = 2.06). As the values of the standard deviations indicate, there are also important differences in respondents’ degree of support for each of these statements. Drawing on extant findings pertaining to the correlates of belief in conspiracy theories, we consider influences that stem from respondents’ social position, their information habits, level of education and analytical skills, and psychological state.
Social position is measured through a variable recording the level of education of respondents’ parents. Students were asked whether any of their parents hold a university degree. The variable is recoded to distinguish between those who have at least one parent with university studies from those whose parents do not have university studies. Since the majority of the students are not yet financially independent from their parents, the assumption is that they share their parents’ social position. Education is frequently used as a measure of social position and university studies appear to be a good predictor of other elements that define social position, such as occupational status and income.
Information effects are measured through a series of variables recording respondents’ news consumption at the time of data collection. One question asked how frequently they read the news and actively sought information at the time of the survey about the current situation compared to previously. Respondents could indicate their answer on a 5-point scale, where 1 signifies they read news less frequently and 5 that they have a higher level of news consumption than before. In addition to this measure, respondents could also indicate their main source of information. Three dichotomous variables were created from this question, separating those who indicated television, newspapers (including online), and discussions with others as their main source of information.
Education and analytical skills have been highlighted as relevant correlates of conspiracy belief (van Prooijen 2017; Oliver & Wood 2014; Douglas et al. 2016). We test the influence of education by considering respondents’ year of study. The variable separates among first, second, third year, and a fourth category of students that includes all those who are in the fourth or higher year of study (including MA and PhD students). We included two other variables that tap into differences in students’ academic performance and analytical skills. The first is whether they receive a study scholarship (yes=1). In Romania, this type of scholarship is granted to students who have very good academic performance and rank at the top of their cohort. The second variable measures how often respondents encounter problems in the process of online learning due to their lack of technological skills (5-point scale variable, ranging from 1=never to 5=very frequently).
Previous research includes anxiety among the main drivers of belief in conspiracy theories (Grzesiak‐Feldman 2013). The respondents in our sample were asked how calm/relaxed they felt in the last two weeks. They could answer on a 6-point scale, where 1 indicates they never felt calm and relaxed in the last two weeks and 6 that they felt this way all the time.
Gender also appears in previous studies as an occasionally significant correlate of conspiracy theory belief, which is why we included it among the predictors, as well as respondents’ place of residence, and another variable that separates students who study STEM (i.e. Computer Sciences and Mathematics; Engineering; Medical Studies; Sciences: Biology, Chemistry, Physics) from those studying other subjects. The latter variable could be seen as a proxy for trust in science.
The distribution of the independent variables is presented in Table 2.
Table 2. Descriptives for independent variables (PDF)
To test the relationship between the factors mentioned above and belief in each of the conspiracy theories we included in our study, we estimate ordinary least squares regression models. Five out of six dependent variables have a close to normal distribution, whereas in the case of belief in 5G conspiracy the distribution is skewed. For this latter case, the distribution was normalized using the natural logarithm. We test the model for each conspiracy statement separately to understand whether the set of predictors is similar for all of them or whether there are differences in what predicts belief in different types of conspiracies or conspiracies corresponding to different dimensions of what has been identified as conspiratorial belief.
Table 3. Correlates of belief in conspiracy theories (OLS coefficients and S.E.) (PDF)
Results indicate that some factors consistently predict belief in (almost) all types of conspiracies, whereas others appear to correlate only with specific statements. The first category includes parents’ level of education (which we took as indicative of respondents’ social position), respondents’ habits of news consumption, and measures of analytical skills. Gender and type of program specialization appear to be conspiracy-specific influences. Respondents who have at least one parent who graduated university are significantly less likely to agree with all conspiracy statements. The strength of the effect varies on a 10-point scale between 0.4 points for the personal well-being type of conspiracies, stating that Covid-19 is created by scientists and spread by 5G technology, and 0.8 points for the government malfeasance type, stating that COVID-19 is a conspiracy of political elites to restrict citizens’ rights and control them.
Respondents’ habits of news consumption are also a significant correlate of belief in almost all types of conspiracy statements. Compared to those who follow the news much rarer than before the pandemic, those who increased their consumption of news to various degrees are less likely to believe in all types of conspiracies, except for those related to personal well-being, namely that COVID-19 is created by scientists and spread by 5G technology. For all the other conspiratorial statements, respondents with higher consumption of news are, on average, about 1 point less likely to agree with the conspiracies, on a 10-point scale. The differences are larger for higher levels of news consumption compared to those who declared looking for information much less frequently than before.
Controlling for frequency of news consumption, the main source of information appears to also matter for conspiracy belief. In general, respondents who indicated TV as their main source of information are significantly more likely to agree with four conspiracy statements claiming that COVID-19 is related to either political or economic elites, is a weapon in the fight of the superpowers for global supremacy, and is created by scientists. The size of the effect varies between 0.6 and 0.8 on a 10-point scale. For the other two conspiracies, namely that COVID-19 is spread by 5G technology and is not as dangerous as presented, although the effect is in the same direction (i.e. TV as the main source of information is a positive correlate of stronger agreement with conspiracies), the coefficient does not reach conventional levels of statistical significance. Similarly, if discussion with others was indicated as the main source of information, conspiracy belief increases. This applies to four statements, namely those linking the pandemic with the conspiracy of political elites, the superpowers’ fight for global supremacy, or the 5G technology, as well as to the belief that the disease is not as dangerous as presented. In these cases, the size of the effect varies between 0.6 and 1 point on the 10-point scale. On the other hand, indicating newspapers as the main source of information decreases belief in all conspiracies, except the one stating that the virus is a weapon used in the fight of superpowers for global supremacy and the one linking COVID-19 with 5G technology. The decrease varies between 0.6 and 1 on the 10-point scale.
Existing research found that the use of traditional media reduces belief in conspiracy theories, whereas reliance on social media and talking to others as main sources of information increases support for these theories (Buturoiu et al. 2021; Enders et al. 2021; Stecula & Pickup 2021; Foley & Wagner 2020). Our results square with the latter finding: students who indicated talking to others as their main source of information at the time of the interview were significantly more likely to support four out of the six conspiracy statements we analyzed. On the other hand, our finding that reliance on TV as main source of information also increases belief in conspiracy theories runs contrary to findings of previous studies. We consider that this is due to the particularities of the Romanian media system which fits the model of a polarized pluralist media system, with high political parallelism, namely a media system characterized by strong partisanship of the media outlets and audiences, as well as media outlets that follow partisan interests (Hallin & Mancini 2004; Castro Herrero et al. 2017). Support for conspiracy beliefs was shown to be stronger in countries with weaker public media systems and higher levels of partisan media (De Coninck et al. 2021). Moreover, the finding that using TV versus newspapers as the main source of information relates to higher levels of conspiracy belief is congruent with communication studies wisdom on the difference between soft and hard news (Reinemann, Stanyer & Scherr 2016). TV programs could expose the viewers to more soft news and this might be even more the case for commercial television programs. Unfortunately, in our study we did not ask what TV channel was primarily used as a source of information, thus fusing public and commercial broadcasts viewership in one indicator.
Education and analytical skills yielded mixed results. The year of study does not make a significant difference in belief in conspiracies. Then again, students who receive a study scholarship are less likely to believe in conspiracies linking the disease with political and economic elites, superpowers, and the 5G technology. The size of the decrease in these beliefs varies between 0.3 and 0.6 on the 10-point scale. Lack of technological skills to the point where it interferes with respondents’ learning in the online environment appears to be a consistent predictor of beliefs in conspiracies relating COVID-19 with political, economic elites, superpowers’ fight for global supremacy, and the 5G technology. Compared to those who never encountered problems due to their lack of technological skills, students who experienced such problems to a higher or much higher degree are significantly more likely to believe in the above-mentioned conspiracy theories. Effects vary in the range of 0.5-1 on the 10-point scale.
Contrary to the expectation that higher levels of anxiety increase agreement with conspiracy statements, our results show no significant relationship between the two variables. Gender yielded mixed results. While not significantly related with belief in four conspiracy statements, women appear to be more likely to believe that COVID-19 was created by scientists and significantly less likely to believe that it is not as dangerous as presented to the public. That is to say, women are significantly more likely than men to endorse a conspiracy pertaining to well-being and significantly less likely than men to believe in a conspiracy pertaining to the control of information. Previous studies found either no significant relationship between gender and conspiracy belief or reported conflicting findings, namely that either men or women are more likely to endorse conspiracy theories (Allington et al. 2021; Cassese, Farhart & Miller 2020; Pizarro et al. 2020). Those that found that men are more susceptible to conspiratorial belief indicated learned helplessness as the root of this relationship, while those that reported women as more inclined to endorse conspiracy theories pointed to personal uncertainty and perceived threat. We conjecture that the health-related nature of the conspiracy theories might appeal more to women’s socially enhanced feelings of care, which, in turn, might make them more alert to potential dangers, thus increasing their anxiety. Indeed, our results indicate that the women in our sample have a higher level of anxiety compared to men (Pearson chi-square=34.52; sig=.000).
Respondents’ place of residence is correlated with only some of the conspiracy theories investigated here. Urban residents are less likely to endorse the statement that COVID-19 is a conspiracy of the political elites. Finally, students enrolled in STEM programs are significantly less likely to believe that COVID-19 is produced by 5G networks.
In this study, we surveyed a sizable and diverse population of university students in the wake of the COVID-19 pandemic. At the time of data collection, Romania was under national lockdown and students were engaged in online learning. The climate at this time was one of heightened uncertainty, fostered also by conflicting messages, a fertile ground for the spread of misinformation and conspiracy theories about the origin, spread, and gravity of the virus. In this context, we investigated predictors of students’ agreement with several conspiratorial statements on the origin and nature of the unfolding pandemic.
The results of our study suggest that belief in COVID-19 conspiracy theories relates to students’ social status, news consumption frequency and source of information, as well as analytical skills. Concurrent with previous research, students who have at least one parent with a university degree, who consumed news more frequently than prior to the pandemic, who indicated newspapers as their main source of information, who receive a study scholarship, and who are not hindered in their learning activities by low levels of technological skills seem to endorse conspiracy theories to a smaller degree. Then again, indicating either TV or discussions with others as the main source of information during the pandemic makes students more likely to endorse conspiratorial explanations about the nature and origin of COVID-19.
This indicates that education, analytical skills (perhaps including “spillover” effects from the parents’ education to their offspring’s analytical and critical thinking abilities), and exposure to higher quality media information provide students with the resources needed to evaluate the information they receive and resist the conspiracy theories that have spread so widely in the wake of the COVID-19 crisis. Students of STEM appear to be more science cognizant and more likely to reject the conspiracy with the highest science-related load, namely that COVID-19 is spread by 5G technology. The psychological factors we investigate here are not significantly associated with conspiracy belief. Anxiety does not drive belief in any of the COVID-19 conspiracy theories we analyzed. As in the case of previous studies, gender effects on strength of COVID-19 related conspiracy beliefs are mixed. Gender is significantly related with two out of six conspiracy statements, although not in a consistent way. While women appear to be more likely to endorse the statement that COVID-19 was created by scientists, they are less likely than men to agree with the claim that the virus is not as dangerous as presented to the public. We consider that this could relate to the fact that women score higher on anxiety feelings compared to men, especially in matters related to health and well-being. We also acknowledge and support the necessity of conducting more in-depth research to understand the roots of these differences, as they are consequential for understanding how conspiracies spread and could be addressed and countered.
Previous studies agree with the finding that higher levels of education reduce belief in COVID-19 conspiracy theories and make people more compliant with government-issued measures, restrictions, and recommendations (although see Stoica & Umbreș 2021). Against this background, our finding that the level of conspiracy belief is quite high among Romanian university students, a population with higher levels of education than the average, is sobering. These results are even more problematic in a country that is lagging far behind the EU average in the share of the vaccinated population and has one of the highest rates of COVID-19 related deaths at the time of writing this article (ECDC 2021).
This study contributes to the literature on conspiracy beliefs by providing new insights on the factors that mitigate the spread and endorsement of conspiracy theories in the Central and East European / post-communist context. By investigating a sizable and diverse sample of university students in Romania at an early stage in the pandemic, we find that factors pertaining to education, analytical skills, and media consumption are key correlates that predict support for conspiracy theories. Given the link between conspiracy belief and health behaviors in the context of the COVID-19 pandemic, these results point to the importance of developing analytical skills and media regulation for curbing misinformation in societal contexts of heightened uncertainty, confusion, and existential threat. However, the extent and strength of conspiracy belief is likely to have changed with the passing of time, as more information became available during the pandemic about the nature, implications, and ways of containing the virus. Therefore, further studies could explore longitudinal change in conspiracy belief. Nevertheless, the findings presented here can provide an explanation for the high levels of vaccine hesitancy and refusal in Romania, which have contributed to the high infection and mortality levels the country faced in autumn 2021.
