Civic Literacy and Disinformation in Democracies

Abstract

The aim of this study is to explore civic literacy as an approach to counter disinformation in democracies. From group interviews, we elicit, categorize, and analyze diverse perspectives on disinformation in Sweden, previously upheld as a country with high civic literacy levels. We focus on people’s understandings of disinformation, their assessment of their own abilities to discern disinformation, and their ideas about how increased resilience to disinformation could be achieved. Our findings, based on input from 73 interviewees across Sweden, suggest that shared basic knowledge on disinformation is lacking. Moreover, there is a related weak understanding of what constitutes authentic information. Those with low awareness operate on a logic of beliefs, implying that measures to improve factuality and objectivity could not even be aspired for. Still, there are also constituents showing advanced understandings. The majority of respondents call for new measures to strengthen citizen knowledge and skills and generate many proposals to that end. Our results indicate that citizen competence needs to increase considerably to keep up with the rapidly evolving disinformation environment. A concerted drive to boost citizen knowledge and skills, tailored to different constituencies, is needed for the democratic system to work as intended.

1. Introduction

The role of civic literacy in addressing disinformation in democracies is important but not well understood. The concept of civic literacy in modern democracies was defined by Henry Milner (2002, p. 3) as “the proportion of their inhabitants demonstrating the knowledge and skills to act as competent citizens… as a condition for attaining certain social outcomes… as means to an end”. Civic literacy enables citizens to understand and assess public information and communication, to engage in civic action, such as voting and holding elected leaders to account. When civic literacy was conceptualized in this way, the world had not yet experienced the current flows of disinformation, propelled by the rapid rise of social media and online news sources posing as alternatives to traditional media.

Over the past decade, populations in many open democracies have become targets of disinformation. The topics of disinformation are marked by both continuity and change, purposefully exploiting societal tensions and fault lines around issues such as migration (Giandomenico and Linderstål 2023), climate change (Oscarsson et al. 2021), the COVID-19 pandemic (Colde et al. 2020), or general elections (LaForge 2020; Havlicek et al. 2018). Research on democratic societies demonstrates that some people are more receptive to disinformation than others. Vulnerability to disinformation is empirically linked to factors such as education, age, ideological leaning, and rurality (Johansson et al. 2022; Rodríguez-Pérez and Canel 2023; Humprecht et al. 2021; Baptista et al. 2021; Wang et al. 2022).

The connection between civic-literacy-enhancing policies and citizen empowerment and awareness was previously found to be particularly strong in Sweden and its neighboring countries in Scandinavia. The Scandinavian countries were singled out as a result of extensive comparative case study research (Milner 2002, pp. 170–77). Twenty years after these findings, we return to Sweden where we probe the role of civic literacy to function as a bulwark against disinformation.

The aim of this study is to explore civic literacy as an approach to counter disinformation in democracies. We focus on three areas linked to citizen competence: people’s understanding of disinformation (knowledge), their assessment of their own abilities to discern disinformation (skills), and their ideas about how increased resilience to disinformation could be achieved. We elicit the perspectives and perceptions of adult Swedish citizens from communities marked by compound structural vulnerabilities to disinformation. We do so to enhance the likelihood of capturing a broad range of views, including from individuals who are receptive to disinformation.

Our findings, first, provide deeper insights into how people understand and approach disinformation, thus complementing previous large-N research. Second, by focusing on adults outside of formal education, we target a largely overlooked constituency. Disinformation research tends to be more focused on children and youth in education settings. Third, we address civic literacy in broad terms, encompassing all ‘literacies’ needed for people to act as competent citizens in contrast to most contemporary studies, which tend to take more narrow approaches.

2. Civic Literacy and Disinformation in Democracies

Disinformation and its negative systemic effects ‘are not new’ to history (Allcott and Gentzkow 2017). However, after the U.S. presidential election in 2016, the entire world became familiar with the expression of fake news (Meneses Silva et al. 2021). Since then, disinformation has been increasingly regarded as a systemic threat to democratic governance (Lewandowsky et al. 2017; Tenove 2020; Neo 2021, 2022).

2.1. Disinformation

Disinformation, defined as false information with the intention to create harm (Tandoc et al. 2020), often attempts to erode trust in the state and in political processes and institutions and to fragment and polarize society. Disinformation comes in different forms. Even if information is false, it can still fall below the threshold of obvious lies (Allcott and Gentzkow 2017; Lazer et al. 2018). Disinformation can be a mix of authentic and slightly distorted information. Disinformation thus removes the accuracy of public debates and affects the quality of communication between policy makers and citizens, eroding trust and deliberative capability. Research on social media unfriending shows that disinformation campaigns may target connectors and erode trust between different groups in society (Asmolov 2018). Disinformation messaging often tends to be emotionally charged, as frustration and anger increase people’s propensity to buy into the messaging and to unknowingly diffuse false content through social media (Horner et al. 2021; Chuai and Zhao 2022). While misinformation also consists of false information, it lacks the intent to do harm.

Disinformation may purposefully attack the very system of democratic governance through the influence of regular citizens unknowing about the harmful intent (Kuklinski et al. 2000; Edelman 1964). Disinformation can be seen as a major threat to democracy.

The empirical literature identifies certain structural population features associated with greater vulnerability to disinformation. Lower education is significantly correlated with vulnerability across country settings. People with lower education tend to be more susceptible to disinformation and to spread false information more than highly educated people (Seo et al. 2021; Baptista et al. 2021; Humprecht et al. 2021). High age is another factor consistently associated with receptivity to disinformation. Research across European and North American contexts shows that older people are more prone to believe in and spread disinformation compared to youth (Baptista et al. 2021; Brashier and Schacter 2020; Guess et al. 2019; Rodríguez-Pérez and Canel 2023). In Sweden, age has a limited but significant effect on the susceptibility to disinformation (Johansson et al. 2022). Other studies point to ideology as a factor, suggesting that people identifying with extreme political views are more likely to share false social media content (Hopp et al. 2020). Respondents with far-right ideology are shown to be more likely to accept fake news (Baptista et al. 2021; Rodríguez-Pérez and Canel 2023; Calvillo et al. 2021) and believe in conspiracy theories (Strömbäck 2022). Gender is showing mixed results where some studies find men to be more resilient to disinformation than women (Arin et al. 2023; Rodríguez-Pérez and Canel 2023); others demonstrate opposite results (Humprecht et al. 2021), and a third set of studies find men and women to be equally vulnerable (Almenar et al. 2021). In Sweden, gender seems to be overshadowed by other factors (Johansson et al. 2022). Rural residency and geographic location more broadly can also affect susceptibility to disinformation (Havlicek et al. 2018; Austin 2023; Ali et al. 2023). Studies show that rural residents in Sweden have lower trust in central authorities and in nationally produced media, and are more likely to hold conspiracy beliefs compared to city residents (Strömbäck 2022; Rothstein and Holmberg 2022; SOM-Institutet 2023).

Resilience to disinformation is here defined in terms of both ‘preservation’ and ‘adaptation’ (Kalniete and Pildegovičs 2021). ‘Preservation’ has to do with preserving a core identity, feature, or function. ‘Adaptation’ has to do with the ability to develop, even thrive, when confronted with external pressures. What is to be ultimately preserved and developed is citizens’ critical thinking skills, trust, and pro-social behaviors when faced with continuously evolving disinformation pressures. The term ‘resilience’ can thus be said to signify adaptive resistance.

2.2. Civic Literacy

In democracies, civic literacy can be seen as fundamental in dealing with disinformation. Whereas autocracies can exercise top-down control of information flows and information content, information in democracies is freer and the opportunities for restrictions are more limited. In democratic systems, a stronger onus is thus with citizens and their own ability to handle information, be it false or authentic. Citizens in democracies are trusted to be competent—possessing the requisite knowledge and skills to handle politically relevant information. Key for civic literacy is that there is systemic intent to inform citizens and that information is made readily available to them to facilitate informed decisions (Milner 2002, pp. 2–5). While there are various problems inherent in democracies (Lijphart 1975; Putnam 2001; Peters 2005; Hendriks 2010), the focus here is on the factors that enable citizens to function as a “bulwark” against negative impacts of disinformation.

Earlier civic literacy research focused on knowledge and skills needed for offline civic action, such as election participation, community involvement, volunteering, social mobilization, or taking part in demonstrations (Dahlgren 2003). Now, online information flows have increased. Civic action now extends to online contexts, for example, through sharing or commenting on political content on social media or signing online petitions (Smith 2013). In recent years, civic literacy research has, consequently, come to focus increasingly on civic media literacy (Middaugh 2019; Robertson and Ortgies-Young 2023), civic social media literacy (Middaugh 2019), civic data literacy (Carmi et al. 2020), and civic digital literacy (Nygren and Brounéus 2018).

Civic literacy approaches to disinformation hold that informed citizenship largely results from the ability of schools and public education institutions to teach individuals the skills necessary to identify, understand, and reject false information (Pennycook and Rand 2019; Horn and Veermans 2019; Buschman 2019; Brisola and Doyle 2019; Kaufman 2021). This approach presumes that disinformation can be objectified, delineated, characterized, and isolated for study and that knowledge and skills can be taught to citizens at a young age for them to be resilient to disinformation in the future. The power and scope of the civic literacy approaches directed at children and youth lies in mandatory schooling (Lutzke et al. 2019; Mirra et al. 2021). The growing efforts to boost media civic literacy or data civic literacy are aimed at youth in the education system (Dumitru et al. 2022).

Without situating civic literacy in the education field as such, studies suggest that resilience to disinformation is influenced by critical thinking abilities (Dame Adjin-Tettey 2022; Petersson 2022). Higher cognitive ability decreases the risk of falling for disinformation (Tandoc et al. 2020; Wolverton and Stevens 2019). Simone Chambers (2021, p. 160) argues that citizens could both be seen as vulnerable to disinformation and trusted to deal with it. Although there are ideas about how to increase media literacy among adults, it is not clear how, when, and where the education of adults would take place (Lutzke et al. 2019; Mirra et al. 2021). The adult population, while constituting the majority of the citizenry, is a largely neglected target group in civic literacy efforts to counter disinformation (Boler et al. 2024).

This study fills three gaps: first, we focus on the knowledge and skills of adult citizens outside of formal education; second, we elicit a wide range of views, including from those who are more prone to believe disinformation; third, we use civic literacy as an umbrella term for all kinds of literacies needed to exercise competent citizenship. Most contemporary research tends to narrow in on specific civic ‘literacies’—such as media, social media, data, digital, information, or electoral literacy.

3. Materials and Methods

Our interest is to capture and categorize how disinformation is understood, how people consider their own skills to assess disinformation, and how they think that disinformation should be handled going forward. To elicit a broad range of citizen perspectives, both those who are receptive and resilient to disinformation and those in-between, we target communities marked by compound structural vulnerability to disinformation. We do so, as we, ex ante, cannot know exactly which individuals will turn up for our interviews, expecting those who are more vulnerable to be less prone to volunteer. To avoid potential local context bias, we select three communities across Sweden with similar structural characteristics.

3.1. Community Selection

We select communities marked by similar compound structural vulnerability to disinformation. Our first selection criterion is rural area (‘landsbygd’ in Swedish). Several studies on Sweden show that rural Swedish residents have lower trust in central authorities and in nationally produced media and higher likelihood to hold conspiratorial beliefs (Strömbäck 2022; Rothstein and Holmberg 2022; SOM-Institutet 2023; Oscarsson et al. 2021). Our second criterion is low prevalence of tertiary education. The level of tertiary education in the locations is markedly lower than among city populations (Statistics Sweden 2023a). Income is a third criterion where annual average income ranges between USD 26,000 and 28,500. Ideological profile is a fourth criterion where communities with a high share of Swedish Democrat (SD) voters are selected. The SD ideology is largely based on right-wing conservative values (Sverigedemokraterna 2019; Nilsson et al. 2020).

The more operational selection of communities is as follows: (1) rural municipalities (i.e., rural categories 1 and 2 according to the Swedish Agency for Economic and Regional Growth 2021); (2) those that feature a prevalence of tertiary education between 10 and 12 percent, according to Statistics Sweden (2022); (3) those that feature low average annual income levels between SEK 292,000 and 302,000, according to Statistics Sweden (2022); (4) those that demonstrate a ratio of SD voters ranging between nearly 40 and 44% in the general election of 2022 (Statistics Sweden 2022). To ensure geographic spread, three communities are selected: one in the northern region of Sweden (Norrland), one in mid-Sweden (Svealand), and one in the south of Sweden (Götaland).

3.2. Participant Selection

To render a diverse group of respondents, village associations (‘byalag’ in Swedish) are used as entry points into each community. Focal points of the village associations are contacted about a research project on news and news perceptions looking to interview people from rural localities in Sweden. Through the focal points, volunteer participants of 18 years and above who are residents of said communities are recruited. The aim is a total of 75 volunteer participants, 25 from each of the three communities.

The result is a group of 73 volunteer participants. See

Table 1. Participant profile.

Category					Total
Gender	Women		Men
	38		35		73
Age	<67 years		67+ years
	36		37		73
Education	Elementary	High School	Vocational	University
	9	17	22	25	73

Note: Shaded rows to enhance readability.

for the full participant profile based on self-reported age, gender, and education level. The group is fairly gender balanced. Although participants are of different ages, ranging from 29 to 89 years, they are older than the general population at an average age of 62 years. About one-third has up to high school education, one-third has vocational training, and one-third tertiary education. We thus have a (self-)selection of more university-educated participants compared to their actual share of their community population (33% versus 10–12%). However, 33% university educated is representative of the Swedish national population (Statistics Sweden 2023b). Since the study is not designed to collect any sensitive personal data and there is no intent to process any such data, no ethical approval by the Swedish Ethical Review Authority is required1.

3.3. Semi-Structured Group Interviews

The method of semi-structured group interviews is used to pose specific questions and steer the discussion toward certain topics. The benefit of group interviews is to be able to control topics and questions, while still allowing participants to reflect freely around given topics. Compared to the focus group discussion set-up, semi-structured group interviews are not as reliant on group interaction and inter-participant dynamics (Cohen et al. 2011, p. 436).

The interview sessions are divided into three parts, where the first is devoted to news, the second part to social media, and the third to social interaction around news-related information. No questions concerning sensitive personal data, on, for example, political or religious views, are asked. The questions are both pointed to structure the content and also open to generate a diversity of answers. For the full list of questions asked, please see Appendix A.

The group interviews are held in local community facilities. Each interview session is conducted by two of the co-authors, with one serving as lead interviewer. Each group consists of about five participants. Participants are informed that no attribution will be made, either to them as individuals or to the interview location. Information will be treated with confidentiality by the research team, and any identifying information will be removed in resulting publications. Given the risk that fellow participants in the group sessions could reveal information, participants are asked not to reference “who said what” afterward. Questions are subsequently posed in several question rounds, allowing every participant to respond to each question. Different individuals are asked to start each round to ensure that not one particular individual would dominate the responses. The semi-structured and facilitated group interview set-up creates a sense of flow.

The interviews, conducted in Swedish, are recorded and transcribed. MAXQDA data analysis software is used to code and analyze the narratives emerging from the interviews. Participants are given randomized individual reference numbers. Interview content that is selected for inclusion in the paper is subsequently translated from Swedish into English by the authors.

We use a basic narrative approach to capture participant perceptions and reflections on disinformation, focusing on content and cognitive categories (Lieblich et al. 1998). The empirical material is thus treated as a whole, although produced through group interviews in different locations. In the narratives emergent from the interviews, the interest is to categorize and systematize how disinformation is understood, whether or not people feel confident about their own abilities to assess disinformation, and how resilience to disinformation could be achieved. In other words, we use a narrative working approach to the transcripts, searching for how participants reveal understandings of disinformation and how to build resilience against it (Lie et al. 2023). We present our results in a thematic format, based on our understanding and definition of disinformation and factors relevant to resilience to disinformation (Hollway and Jefferson 2000; Riessman 2008).

4. Results

4.1. How Do People Understand Disinformation?

When it comes to knowledge on disinformation, most participants demonstrate at least some awareness of what disinformation is and how disinformation operates, but a sizeable share also show a low degree of understanding. No or little awareness is more prevalent among the older participants, especially those in their 70s and 80s. The highest level of awareness is shown by the younger participants, with some notable exceptions. Low awareness is also more associated with education below university level. There are no significant gender differences. Overall, there is a marked variation in the understanding and analysis among participants. Below, we capture the more commonly shared understandings, along with the minority of participants who demonstrate a weaker understanding and another minority that shows more advanced and reflective reasoning.

4.1.1. Common Understandings

As regards the understanding of what disinformation is, a common notion is that disinformation consists of evidently false information and is not that widespread. If several media outlets say the same thing, that is, if a person can cross-check between sources, the information is deemed to be safe. The understanding of fake news as evidently fake or ‘outlier information’ is captured in the following statement: “I have not noticed any influencing campaigns from the outside”2. When participants freely reflect around how disinformation is manifested, a variety of phenomena come up. These include cyber-attacks, such as the hacking of the webpages of newspapers or public institutions. Fraud against the elderly in the form of scam calls is also brought up. The wider threat posed by artificial intelligence is a shared source of concern.

Distance emerges as a factor widely associated with disinformation. The idea is that news information produced closer to home, local news especially, is more reliable. “I disconnect from news from abroad because you do not know what is true. Swedish news is better in that respect”3. With local news, direct personal verification is possible. People can go directly to the source, following the principle of I-only-trust-it-when-I-see-it. Disinformation is about “outsider” actors wanting to harm Sweden4.

A broadly shared sentiment is that the primary target of disinformation is someone else. As one respondent puts it, “Disinformation doesn’t target me, only the broad masses”5. Children are widely considered as a target group6. Under-age TikTok users are seen as especially vulnerable7. There is an unease about the constant, unknown influencing of children and youth who spend a lot of time online. The elderly are also identified as vulnerable as they have not grown up with technology and not learned media literacy or source criticism at school8. “The younger you are, the more on top you are, the older the more unaware,” in the words of a participant9.

Social media is widely recognized as a space for disinformation production and diffusion. According to one participant: “Influencing is everywhere but especially online and on social media”10. When participants are asked to reflect on how disinformation operates, some suggest that baits are put out on social media. One person suggests particularly that “Flashback is an arena to spread disinformation, just put out a click bait”11. The speed of online information diffusion is seen as linked to increased risk of false or inaccurate information. Social media helps disseminate (dis)information by offering text and images in powerful combinations.

4.1.2. Low-Awareness Minority

Among participants with low awareness of disinformation, some noteworthy perspectives emerge. One is that there is no objective information or factual reporting at all. All news information is considered to be subjective interpretations of reality. News reporting is always seen to take sides. Since everything is seen as politically controlled and influenced, including education programs for Swedish journalists, virtually no news sources can be trusted. “There is a lot of fake news out there. But we have found likeminded friends and acquaintances, so we stick to that circle of people and live on”12. There is nothing that can be done to make news reporting or information more objective. It is impossible to know if a news item is true. The notion is that you cannot trust anything, only yourself, and simply must form your own opinion on topics. How such personal opinions are being formed is unclear, however, and when probed for awareness on journalistic methods or criteria for news selection or production, little reasoning is provided. This perspective is shaped by a high degree of mistrust.

A resemblant but still different perspective on disinformation is provided by a participant who had grown up in an authoritarian regime and become a Swedish citizen as an adult. “Back there, we did not trust the news. We knew they weren’t true.” All formal or official news outlets were by default perceived as government-provided disinformation. Consequently, if a person wants the truth, they could go to the comments section of online news where ordinary people post their thoughts13.

4.1.3. Advanced Understandings

Some participants, most of whom are well educated, demonstrate a wide range of more deliberative understandings. They address news source production and evaluation and various disinformation strategies used to divide and polarize the population. Advanced reasoning also revolves around a person’s own vulnerabilities and beliefs and responsibility and the recognition that anyone could be a victim of disinformation.

The speed of news distribution is highlighted. Participants acknowledge that cross-checking and verification takes more time. Some therefore prefer print newspapers since this news modality is seen to operate at a slower, more deliberate speed. The importance of information sources, source proximity, and source evaluation are brought up. The authenticity of a news item would depend on the source. The closer a reporter is to the source, the more trustworthy the news.

Psychological factors in terms of emotions, cognition, and critical-thinking skills are also raised. The role of emotional triggers and negative emotions, even “emotional storms”, come up across various interview sessions. Disinformation actors are described to exploit emotional reactions to contentious issues to spread false information. Upsetting news is seen as a trigger on social media.14 “Us against them is effective”15. It is about “creating an us-against-them feeling, stirring up emotions,” a participant stated16. “Feelings are important. Something that I am passionate about, something that confirms my agenda”17. The importance of education is brought up. Without proper education, people are seen as easier to “influence, fool and control”18.

A repertoire of information manipulation measures emerge. One is to convey half-truths by omitting some aspects19 or reporting a twisted truth. As put by a respondent, “not a blatant lie but not true either”20. Another is to exploit societal divisions by taking a divisive topic and using extremist views to polarize society21. A third is to create a sense of insecurity and to provoke fear to which simple silver-bullet solutions are presented22. Scapegoating or having a common enemy are considered to be unifiers23.

Some disinformation strategies are seen as particularly dangerous. Among these are the planting of fake news into established media24 or influencing established politicians. Another is the unintentional spread of false news over and over again, turning lies into truth in the eyes of a large number of people25. One example raised is the LVU campaign aimed at discrediting the Swedish social services among Arabic-speaking audiences26.

Interestingly, the timing of disinformation is brought up by participants who are politically engaged at the local level. “At election time international actors get activated,” one participant comments27. Another participant points to the prospects for disinformation at EU elections28: “In connection with elections, information about rapes and assaults could increase markedly to generate a sense of a lack of safety. However, the number of complaints filed to the police would not increase. People just want scapegoats”29. In addition to elections, one participant notes the use of diversion tactics: “When we are preoccupied with something else, then military information in the borderland of truth, is injected”30.

There is also a recognition that you cannot mistrust everything. There is a need to be able to believe some information for society to work. Some participants admit that they are choosing media that confirm their own pre-existing attitudes and values. “This aligns with my world view”31. If news aligns with one’s own world view, it is seen as easier to believe.

4.2. How Do People Assess Their Own Ability to Identify Disinformation?

In terms of skills to identify disinformation, participants are asked to reflect on their own ability. The majority of participants, about 70 percent, claim that they are unsatisfied with their ability to discern disinformation and want to learn more. The participants who feel satisfied with their own aptitude are younger or demonstrate an overall attitude of distrust.

4.2.1. People Dissatisfied with Their Skills

Overall concerns and worry mark the perspectives of the majority of people who do not feel confident in their own ability to identify false news. “I would like to learn more about source criticism”32, one participant states. “The digital development forces you to become more critical and reflective,” another participant notes33. “With international events it is particularly difficult. Everyone wants to promote their version. The great powers try to assert themselves and exercise influence,” a participant observes34. “The news flow is so large that it is really difficult to know what is true. It would be helpful to learn how to address lies, but it is difficult”35.

Several participants reflect on their roles as parents. “My daughter uses TikTok a lot, where people can easily spread any information,” one participant says36. Another participant comments: “I realize that I know very little. There should be trainings for how parents could better guide their children”37. A participant mentions that the Swedish civil contingencies agency has a list of trustworthy media and online training programs38.

Many also want to learn specifically about artificial intelligence (AI), fearing deepfakes of text, image, video, and audio content. One participant expresses worries about AI-generated fraud via email and phone calls, leading her family to pick code words to verify that a call from a family member is authentic39. “Of course you want to learn how to separate real from fake information. Sometimes I get a mobile alert asking, ‘Can you tell if it is AI or real?’. Then you get to learn what [factors] to look for”40, a participant states.

4.2.2. People Satisfied with Their Skills

Few participants state explicit satisfaction with their own abilities with some exceptions. One participant explains that “To get facts, I just have to search the internet”41. “Yes, I am pretty good at it. I am currently pursuing an education where I am encouraged to use source criticism”42, one participant states. Another participant reports to have sufficient knowledge on how to detect fraud43.

Another line of argument is one of distrust: “I do not buy into anything, because I do not trust people”44. “There is not much to do about it—I do not count on anything to be true”45. “Journalism…well I do not know if they are being influenced by someone. I am really suspicious of everything”46. Another participant points to the impossibility of learning: “I would like to get better, but I believe that it is impossible. There is not much truth left. You have to form your own opinion according to what you believe”47.

Other motivations for not needing to learn about how to identify disinformation differ in nature. Some participants claim that they are “protected” from disinformation as they do not use the internet much for reasons of old age or dyslexia48. Other participants simply state that they are not bothered about disinformation.

4.3. What Opportunities Do People Identify for Strengthening Resilience to Disinformation?

Most participants agree that measures to strengthen resilience to disinformation need to be taken. The participants together generate over 80 proposals, which could be placed into three broader categories: education, civil society engagement, and news media. The more advanced proposals come from people who are well read and, in some instances, also had to tackle issues of disinformation in their daily work as teachers or social workers. A smaller share of participants hold the view that since no information could be trusted anyway, there is little use in doing anything proactive.

4.3.1. Education and Training

The majority of proposals involve education about disinformation in various forms, predominantly through the regular school system but also training for other constituencies. Most suggestions address school education for children and youth. One participant suggests the introduction of social media and information literacy as a new school subject49. This sentiment is echoed by other participants: “Today you need to teach children how to sift through a massive flow of information”50. Teachers and parents need to be better equipped to teach and guide children. “In teachers’ trainings, critical thinking and source evaluation should be integrated”51. More holistic approaches involving children, school teachers, and parents are called for. A teacher-participant says: “Do not avoid the difficult topics during parent-teacher meetings. A more direct dialogue with the parents is needed…It should not be children and adults on their own, everyone should help each other…”52.

The elderly are considered another salient target group for training by many participants. “More targeted efforts for elderly people are needed”53. Some point to existing study material that pensioners’ associations had produced to protect against fraud54 and suggest that those types of associations could also organize disinformation training and lectures55.

Some participants suggest that government-sponsored public education campaigns are a way to reach the general public. “Education campaigns such as ‘Think about this’ or ‘Are you thinking critically about information sources today?’”56 could be launched, according to one participant. Such campaigns could be aired on a public service to target the elderly, who are seen as faithful consumers of traditional media. Younger constituencies would be targeted on social media. Disinformation should be countered in the spaces where false information is already circulating, as one participant suggests: “For example on Instagram, Facebook and TikTok. I believe in short and clear messages. There is a lot of disinformation on social media. Therefore, it is important to spread truth there”57.

For adults, proposals include training to increase critical thinking skills. This could take the form of study circles and lectures. Many emphasize the importance of in-person interactive training and discussions as opposed to online modalities58. “Meeting in-person in a group with a real teacher on more than one occasion would be good”59.

The need for tailored training and information materials for different adult constituencies is also brought up. A differentiated approach is, for example, needed for people with functional variations. “I am dyslectic so it is difficult to read. Smaller groups with a course convener would be helpful”60, one participant states. Another participant works at a service home for persons with intellectual disabilities: “My sense is that our residents are very exposed. They would definitely need support, for example, through easy-to-read manuals with more pictures. An online course would not work”61. Also, new citizens or immigrants with non-Swedish backgrounds come up as a constituency for which tailored disinformation materials and training is needed.

Most participants prefer to learn about disinformation through in-person teaching and interactive discussions, including study circles. Some participants suggest that representatives of government agencies, like the police or the military, could be part of the education efforts62.

4.3.2. Civil Society and Local Community Engagement

Civil society and local community associations are identified as conducive spaces for disinformation education and discussion and for bringing people with different perspectives together.

The view is that people need to meet more and discuss things together. “We speak too little about democracy… If we want to have democracy, we need to discuss things also with those who have different opinions. You need to be able to talk to everyone. It is exciting to debate and sharpen your arguments”63, one participant states. “Strengthening civil society and liberal or popular adult education [“folkbildning” in Swedish] is very important”64, another participated comments. “We should start a new citizens’ movement… it would be a good mission for our public education associations to pursue this”65. One participant suggests an approach of respectful engagement with adults who are sharing false information online: “You should not ridicule people who may believe in false information, but rather explain nicely. For example, start sending links with higher quality information. If you use an accusatory tone, you will not reach them”66.

An opposite perspective is brought up by a participant who reports extensive online engagement but is hesitant over in-person discussions: “If you meet in person it may be more difficult to bring up some political stuff since other people may have a different opinion—I don’t know how good civil society is, it does not guarantee the truth. If you hang out with a group of people, then you get colored by that group”67.

An elderly participant shares that she is a member of an association where people would share experiences of having been exposed to fraud: “In these gatherings you get information on what is going on, so that you are not fooled”68.

Small communities are seen to force people to interact with those different from themselves. The issue of gossip is brought up—and how disinformation could play out and be countered at the village level 69. “We need to develop our group responsibility and be a bit critical toward each other. Not just discussing something that has been said, but discussing what the sources are and how it came about. We need to hold each other accountable when disinformation is being spread”70.

Other participants share their experiences of online activism where they, as individuals, counter disinformation and hold others accountable: “I usually report fake adverts. Sometimes they disappear. I have even chatted with hackers online for fun and reported false profiles. If more people do this, it could make a difference”71.

4.3.3. News Media

Several ideas and reflections concern established media, including public service, and how trust in them could be increased.

Increased fact verification by news media and improved news reporting are raised but also the need for transparency about the news production process and journalistic methods overall. “Newspapers in Sweden need to be better at fact checking”72, one participant notes. Some participants recognize the value of slowed-down news reporting, due to the need for fact checking. “Quality journalism with solid source work is important”73. One participant suggests some kind of ‘label’ or a ‘certificate’ to signal that a news item has been verified and produced according to certain standards74. Having a responsible editor and reporters on the news site are factors to be taken into account75. Some participants even propose restrictions of certain news sources, including those that use incorrect and inappropriate language76.

News selection and the balancing of ‘negative incident reporting’ with events or developments deemed important for society is another theme. Several participants call for a better balance between positive and negative news. The sense is that of being overwhelmed with too much negative news, leading to ‘numbness’ that could lead people to disconnect from news. “There is a lot of focus on creating news that many people read, rather than news that could benefit society. They news organizations need to take more responsibility”77.

A recurring issue is the lack of local news coverage and how this affects trust in news and perceived news reliability. “If you are a national newspaper, you need to take that responsibility, to actually reflect the nation. You need to cover more than the large cities and Mälardalen”78. “There needs to be something for those of us who live on in more rural areas to recognize ourselves in. There are benefits to learning about each other, not just being in a ‘Stockholm bubble’ or a ‘countryside bubble’”79.

Another line of argument is that Swedish news tends to focus too much on news interpretation and news debates instead of factual reporting on current events. “News should be concrete and about facts. They always bring in people to ‘discuss’ the news, and in the end it is all guess work”80. Swedish news is compared in particular to Finnish news, which is deemed to be more factual in conveying current events. “It is also about how you frame the news”81, one participant comments. You can describe a situation in terms of “‘now things are becoming dangerous,’ or you can put them in more objective terms”82.

5. Discussion

Several empirical findings are worth highlighting. First, on knowledge, we find an evident lack of a basic shared knowledge base on disinformation. While most participants demonstrate some awareness of how disinformation operates, even the majority of participants could be seen as scrambling for individual “pieces of the puzzle” of what constitutes disinformation. The different interviews generate different understandings, and there is little convergence. In particular, there is no real common understanding on what disinformation is, how disinformation works, and what the possible impacts are. Several respondents with low awareness are not even operating in the realm of knowledge but in the sphere of beliefs. They regard news information as self-formed beliefs and subjective viewpoints and consequently suggest that objectivity or factuality could not even be aspired for. Their perspectives and those of new citizens with backgrounds from autocratic countries highlight a different relationship to disinformation compared to the majority. Finally, a third category of interviewees, also a minority, express highly advanced reflections on disinformation and various modalities of disinformation.

Second, and relatedly, a common understanding of how authentic news or information is being produced is also lacking. Even participants who express confidence in the public service say that they “choose to trust” the public service. Few show awareness of the method or the concrete steps involved in the news production process, including criteria for news selection, proximity to sources, verification, the role of responsible editors, etc.

Third, on skills, the vast majority of participants, about 70 percent, claim that they are unsatisfied with their ability to discern disinformation and want to learn more. Many respondents express particular concern for children and the elderly in their communities. Participants who had previously received training on disinformation are especially open. Several interviewees signal hope and agency linked to the prospects of addressing (dis)information through increased civic engagement.

A fourth finding is that people’s assessments of their own skills to identify disinformation appear to be linked to their beliefs regarding the possibility of objectivity and factuality. A hyper-critical and distrusting attitude makes people revert to their own belief formation, which is somehow taking place in online and offline echo-chambers.

Overall, our findings on civic knowledge and skills connected to disinformation suggest a variation in vulnerability. Different constituencies seem to be vulnerable to disinformation in different ways. Youth may be more informed of how disinformation works and more skilled at social media usage but are also more exposed to online manipulation and are not necessarily aware of how they are being influenced. The elderly are less informed of how disinformation works, but being more offline, they are also less exposed to online disinformation and more prone to consume public service and established news media. The minority of distrusting people who “form” opinions and beliefs on their own are less inclined to have their ideas challenged beyond the closed circles of the like-minded. Swedish constituents with backgrounds from autocratic countries with no independent media and no public service have different frames of reference when it comes to disinformation and require a tailored approach. So do people with functional variations who may need more accessible and less text-based messaging on disinformation.

On suggestions on how to improve resilience to disinformation, the majority find it important to take active measures to counter disinformation. Together, participants generate over 80 proposals on how to create resilience, ranging from school curricula to parental training, teacher training, lecturing by government agencies, and civil society initiatives, along with measures undertaken by news media and public service specifically. Many advocate increased in-person interaction and engagement through study circles, lectures, and dialogues. These creative ideas stand in contrast to a minority of participants who do not see much value in measures to counter disinformation and who purposefully do not want to engage with those who are not like-minded.

6. Conclusions

With the aim to explore civic literacy as an approach to counter disinformation in democracies, we use Sweden, previously identified by Milner (2002) as a country of high civic literacy, as a testing ground. Our interest is in how people understand disinformation, how they assess their own skills in discerning disinformation and in their ideas about how to increase resilience to disinformation. To attain a diverse mix of perspectives and to avoid getting a self-selection of highly informed respondents given our interview volunteer set-up, we deliberately turn to communities marked by compound structural vulnerabilities to disinformation. We elicit, categorize, and analyze the understandings of 73 respondents across Sweden.

How do our findings speak to the role of civic literacy in tackling disinformation? Democracy—with its limits to centralized information control, censorship, and shutdowns—assumes or trusts citizens to be sufficiently competent. Our empirical findings suggest that there is no shared knowledge base on disinformation but also that most people recognize their lack of understanding and skills and want to improve. They tend to be even more concerned about the weakness of others.

In other words, civic literacy as an approach to tackle disinformation has clear limits even in a country like Sweden, historically known for its high civic literacy levels. In part, this may be due to civic literacy being more demanding than ever. When civic literacy was conceptualized some decades back, information sources or channels were more limited. Citizens, to a greater extent, received the same information but would interpret it differently. Today, there is a greater plurality of information channels. Citizens may find themselves in totally different information realms. The burden is now with citizens to assess information sources and determine the authenticity or falsity of information content. A high degree of civic competence is needed across different domains to keep up with the growing volumes of disinformation evolving at a rapid speed due to technological developments. People need cognitive skills, technical know-how, emotional self-awareness, and an ability to navigate massive flows of true, false, and irrelevant information via images, texts, videos, and audio clips. This kind of adaptive resistance requires a broader skill set than previously.

Possible limitations to our findings concern their validity. In terms of internal validity, our—by necessity—small sample may not necessarily reflect the composition of views of the communities visited or those of the Swedish population at large. Furthermore, it is possible that group interview dynamics might have hampered the free expression of viewpoints compared to more confidential one-on-one interview set-ups. On the other hand, for some respondents, being in a group together with some known participants might have provided comfort and ease, and we do find a divergence of perspectives also within group sessions. The external validity of our results has to do with Sweden as a case. Civic literacy as a concept emerged as part of a larger discourse on social capital. Civic literacy has to do with society-wide measures, collective response, and acceptance among citizens for collective measures. This implies that the preconditions for civic literacy vary between countries depending on their histories, political cultures, and institutional set-ups. Sweden has a history of a relatively centralized governance and information provision through public service. This raises issues of comparability with democratic countries marked by more decentralized structures, by a stronger focus on individualism, and a historically weaker focus on public service.

Future research could extend the empirical inquiry to other countries with comparable governance set-ups and histories of public service media. Countries in the Nordic region, such as Norway and Finland, would be a natural start, but other European countries such as the UK and France could also be considered. Finland would be a particularly interesting comparison case, considering Finnish policy and practice on civic literacy and education also in recent decades. Since the Russian annexation of Crimea in 2014, Finland has focused on building up public awareness to disinformation (Bjola and Papadakis 2020).

In terms of policy recommendations, the focus would be to further enable and empower people as citizens and not only as news consumers or social media users (Chambers 2021, p. 161). A larger concerted drive to boost citizen knowledge and skills across different constituencies is needed to ensure a critical mass of competent citizens. Efforts need to be undertaken by citizens and for citizens, as suggested by Chambers, who proposes a combination of “structural, regulatory, and ethical responses” (Chambers 2021, p. 161). Approaches need to be tailored and adaptive. While strengthened disinformation education of children and youth is key, a static or linear approach centered on education of the under-aged is not sufficient. The adult population needs to be catered for as well.

A civic literacy boost could consist of three components. First, a “countering disinformation” effort conveying basic knowledge of what disinformation is and how it works to provide citizens with a common knowledge foundation. Accuracy and care should be taken by decision makers and political actors when using the term “disinformation” publicly. A second component is “promoting quality information and informational hygiene”. An enhanced understanding of the news production process, its principles, and mechanisms must be communicated to clarify how serious news providers work and what is the critical function played by public service. Quality news should be present in the spaces where citizens are, including on social media platforms. A third component has to do with increased civic engagement, which facilitates real-life encounters and exchanges between people of different backgrounds and views, allowing people to break out of their social echo chambers.