Lies, conspiracy theories, propaganda and other false and misleading content are not new. However, the rise of the Internet and online platforms have changed the scale and speed by which such content can be spread, creating risks to the safety and well-being of people and society. This Spotlight first sheds light on media consumption and trust in media sources. It then presents cross-country data on privacy concerns and individuals’ perceived control over their personal data on websites, applications and social media. Insights into how exposure to additional context influences the ability of individuals to identify the veracity of information on line follow. The Spotlight concludes with perspectives on how media literacy initiatives may need to evolve in response.
OECD Digital Economy Outlook 2024 (Volume 2)
Abstract
Lies, conspiracy theories, propaganda and other false and misleading content are not new, but the Internet has changed the scale and speed by which such content is developed and spread. False and misleading content is not necessarily illegal, but its rapid and global spread is harmful for society and democracy. In today’s information society, information plays a significant role economically, socially and culturally, and information and communication technologies are widespread and intensively used. As people increasingly source information from websites and social media, false and misleading content on line raises risks for the well-being of people and society.
To address the pernicious effects of false and misleading content on line, it is important to understand how people consume information and what influences media consumption. By examining individuals’ reported trust in various media sources, it is possible to gain a deeper understanding of the broader information ecosystem in which people are exposed to both factual, and false and misleading content on line. This can in turn support the design of more effective public policies.1
Concerns about privacy on line and feeling not in control over personal data reduce trust in the information ecosystem. Understanding the extent to which privacy concerns and feelings of control over personal data vary across countries and demographic groups can shed light on which factors influence trust in digital environments. This, in turn, provides insights into how to build a more trusted information ecosystem.
Moreover, knowing how to help people become better at identifying false and misleading content on line is likewise an important step in mitigating its harmful effects. While technology is certainly one part of the solution, people also play an important role. In particular, equipping people with the skills to detect false and misleading content through media literacy2 is often advocated. However, more evidence is needed to understand how to best design these initiatives and which demographic groups are most at risk.
This Spotlight presents new evidence from the OECD Truth Quest Survey (Box 3.S.1) on media consumption and privacy. It sheds light on where people source information and how trustworthy they perceive various media sources. It then examines attitudes towards privacy and control over personal data. The Spotlight further analyses the behaviour of people as they interact with true and false and misleading content. It concludes with perspectives on strengthening the information ecosystem by rethinking the design of media literacy initiatives.
The OECD Truth Quest Survey measures the ability of people to identify false and misleading content on line in a real life setting across 21 countries (OECD, 2024[1]). The survey provides cross-country comparable evidence on media literacy skills by theme, type and origin (i.e. whether information is generated by humans or artificial intelligence [AI]). It assesses the effect of AI labels on people’s performance and offers insights into where people obtain information, as well as their perceptions about their media literacy skills, among other issues. In this way, the survey contributes to the statistical literature on measuring false and misleading content.
In the survey, participants interact with both true and false content on an interface that resembles a “real life” social media site. Design features such as avatars and scores aim to encourage engagement. The survey was designed by the OECD and administered by an external polling company to ensure a representative sample in each country. It was translated and localised into the primary languages of each country covered, and it was administered in January and February 2024.
The OECD Truth Quest survey was administered to approximately 2 000 people in each of the 21 countries covered by the survey. The respondents are representative of the population of each country based on demographic variables including age, gender, sub-national region, educational attainment, and income level using country-specific quotas. Quotas were calculated based on data from national statistical offices and related institutes. Post-stratification weights were calculated to ensure nationally representative samples. In total, 40 765 people completed the survey across five continents. More detail about the OECD Truth Quest Survey methodology can be found in OECD (2024[1]).
The overall Truth Quest score measures people’s ability to correctly identify false and misleading content on line (Figure 3.S.1)*. Key data from the OECD Truth Quest Survey can be found on the OECD Going Digital Toolkit (OECD, 2024[2]).
Overall OECD Truth Quest score, 2024
Note: AVG = Average. * To calculate the overall Truth Quest score, the total number of correct responses is divided by the total number of claims seen. A country score is thus an average of all respondents’ results. The average is calculated as a simple average of the 21 country scores from the OECD Truth Quest Survey: Australia, Belgium, Brazil, Canada, Colombia, Finland, France, Germany, Ireland, Italy, Japan, Luxembourg, Mexico, Netherlands, Norway, Poland, Portugal, Spain, Switzerland, United Kingdom and United States.
Source: OECD (2024[3]), “Ability of adults to identify the veracity of content on line”, OECD Going Digital Toolkit, based on the OECD Truth Quest Survey, https://goingdigital.oecd.org/indicator/80.
As digital technologies have increasingly become part of daily life, the channels by which people consume information have multiplied. Smartphones and applications (apps) now play an important role in how people consume information. In response, news outlets have shifted to offering information through a wider variety of channels (Barthel et al., 2020[4]; Al-Zoubi, 2024[5]). Understanding where people source information and how much they trust it plays an important role in improving the integrity of the information ecosystem (OECD, 2024[6]).
Recent data show that some demographic factors influence media consumption. Information from websites or apps is the most common source, on average, across age groups. However, age is nonetheless one of the most important factors that influence the consumption of other media sources (Figure 3.S.2). The largest difference exists for social media3 , where people aged 18-24 are 25 percentage points (pp) more likely to source information from social media than those aged 65 and older. Those aged 65 and older more frequently consume information from all of the other media sources surveyed than people aged 18-24: public TV (20 pp), print publications (19 pp), public radio (16 pp), private TV (11 pp), private radio (4 pp) and news websites or apps (1 pp).
Income and education also play a role in influencing media consumption. Individuals living in high-income households (top 20%) are more likely than those in low-income households (bottom 20%) to get information from any source except for social media (3 pp). Similarly, those with tertiary education are more likely than those with low education to obtain information from any source except for social media (2 pp). The largest gaps between high-income and high-educated people and low-income and low-educated people are related to news websites and apps (11-12 pp), as well as public radio (12 pp). In the United States, the gap for print publications between those with highest and lowest levels of education reaches 30 pp and it is likewise large for news websites and apps (24 pp). In the United States, those with tertiary education are also more likely to get information from social media (10 pp), while the opposite is true for people in Finland (10 pp), Norway (11 pp) and Switzerland (13 pp).
Gender differences in media consumption are less striking. Men tend to obtain information from all sources more frequently except for social media, the source where women are 8 pp on average more likely to get information. Some interesting country-specific results emerge for social media, which is more frequently a source of information for women in all countries except for the United States, where the gap is below 1 pp. In Finland and Norway, women are 16 pp more likely to source information from social media. For other media sources, the largest gap is in Spain where men are 21 pp more likely to get news from private radio. Conversely, women in Spain are 21 pp more likely to get news from private TV.
Media source by frequency and age, 2024
Notes: Respondents were asked to indicate whether they “often”, “sometimes”, “rarely” or “never” get news from public television, private television, public radio, private radio, print publications, news websites or apps, and social media sites or apps (e.g. Facebook, Twitter/X, Instagram, WhatsApp, Telegram). Social media sites and apps are surveyed as an aggregate.3 This figure shows the share of individuals who “often” and “sometimes” get information from the various media sources. All individuals are defined as people aged 18 and older.
Source: Authors’ calculations based on the OECD Truth Quest Survey (OECD, 2024[1]).
Age differences are also evident in people’s trust in media sources. Young people aged 18-24 generally trust all media sources more than people aged 65 and older. The one exception is public sources (TV and radio), where there is a trust gap of 5 pp with the older cohort trusting information more (Figure 3.S.3). The largest trust gap exists for social media, where people aged 18-24 are 20 pp more likely than those aged 65 and older to trust information they obtain on social media. Social media is also the least trusted source of information overall.
Trust level, by age, 2024
Notes: Respondents were asked to indicate how much, if at all, they trust the information from public television, private television, public radio, private radio, print publications, news websites or apps, and social media sites or apps (e.g. Facebook, Twitter/X, Instagram, WhatsApp, Telegram). Possible responses include “a lot”, “some”, “not too much” and “never”. Social media sites and apps are surveyed as an aggregate.3 This figure shows the share of individuals who trust information “a lot” and “some” from the various media sources. All individuals are defined as people aged 18 and older.
Source: Authors’ calculations based on the OECD Truth Quest Survey (OECD, 2024[1]).
Differences in people’s trust in various media sources in relation to education and income reflect the same patterns as media consumption. People with tertiary education and those living in high-income households (top 20%) place higher trust in all media sources except for social media. In terms of gender, there are few differences between men and women, although there is some cross-country variation. For example, women in Spain trust all types of media sources more than men in Spain do: print publications (11 pp), public TV (10 pp), public radio (8 pp), social media (8 pp), news websites or apps (7 pp), private TV (6 pp) and private radio (1 pp).
While social media is the least trusted media source overall (43%), there are notable differences across countries (Figure 3.S.4).4 Individuals in the Latin American countries surveyed tend to trust social media more than people do in other regions.5 In contrast, individuals from the United Kingdom have the lowest trust in information from social media, with about a quarter of people trusting information on social media some or a lot. Across all countries covered, young people aged 18-24 trust information more than people aged 65 and older (Figure 3.S.4). The most noteworthy trust gap between young and older generations is in the United States, where 49 pp separate the two age groups.
Trust in information on social media, by age, 2024
Notes: Respondents were asked to indicate how much, if at all, they trust the information from public television, private television, public radio, private radio, print publications, news websites or apps, and social media sites or apps (e.g. Facebook, Twitter/X, Instagram, WhatsApp, Telegram). Possible responses include “a lot”, “some”, “not too much” and “never”. This figure shows the share of individuals who trust information on social media “a lot” and “some”. Social media sites and apps are surveyed as an aggregate.3 All individuals are defined as people aged 18 and older.
Source: OECD (2024[20]), “Share of adults who trust information from social media sites or apps”, OECD Going Digital Toolkit, based on the OECD Truth Quest Survey, https://goingdigital.oecd.org/indicator/83.
Other demographic differences are also noteworthy. People living in high-income households and those with tertiary education trust information from social media less than those living in low-income households and those with low or no education. This is true for all countries with relatively few exceptions. The largest gaps between people with tertiary education and those with low education are observed in Finland (17 pp), Japan (17 pp), the Netherlands (16 pp), Norway (19 pp) and Switzerland (22 pp). In terms of income, the largest gaps exist in Italy (16 pp) and the Netherlands (16 pp). While on average the differences between men and women are small (3 pp), women are more likely to trust information from social media at least to some extent in Norway (17 pp) and in Finland (10 pp). Conversely, men trust information from social media more in the United States (7 pp).
Importantly, people who trust information from social media have a lower ability to detect its veracity (OECD, 2024[1]). On average across countries, those who trust information on social media a lot demonstrated lower ability to identify its veracity (54%) compared to those who trust information on social media somewhat (59%) and not much or not at all (62%).6 When looking at more nuanced levels of trust, the gap between those who trust social media a lot and those who do not trust at all reaches 8 pp. In the United Kingdom, the difference between those who trust information on social media a lot and not at all is 14 pp. However, in countries where trust in information on social media is high, such as Brazil or Colombia, the difference is relatively small (3 pp).
Protecting individual privacy promotes safety, dignity, and freedom of thought and expression. Concerns about privacy, and control over and the protection of personal data, reduce trust in the information ecosystem. Social media platforms, for example, offer zero price services in exchange for personal data, which are then used as part of the platform’s business model (e.g. for targeted advertising). The extensive amount of personal data gathered, processed and exchanged over online platforms generates risks to individuals’ privacy, especially as personal data are sometimes used in ways unanticipated at the time of collection. Technological advances have also made it easier to re-identify people via personal data that were intended to be anonymous.
How much does privacy matter to people in digital environments? Data show that privacy on websites, apps or social media is a concern for most people (Figure 3.S.5).7 On average, over half of people (56%) avoid using certain websites, apps or social media due to privacy concerns. The share of people that avoid certain online spaces because of privacy concerns ranges from 65% in Portugal to 36% in Japan. Only 14% of people on average are not concerned about privacy when using websites, apps or social media.
Share of adults who avoid certain websites, apps or social media due to privacy concerns, by age and gender, 2024
Notes: Respondents were asked to indicate whether they agreed with the following statement: “I avoid using certain websites, apps or social media due to privacy concerns”. Possible responses include “strongly agree”, “agree”, “neither agree nor disagree”, “disagree” and “strongly disagree”. Social media sites and apps are surveyed as an aggregate.3 All individuals are defined as people aged 18 and older.
Source: OECD (2024[7]), “Share of adults who avoid using certain websites, apps or social media due to privacy concerns”, OECD Going Digital Toolkit, based on the OECD Truth Quest Survey, https://goingdigital.oecd.org/indicator/84.
The percentage of people avoiding using certain websites, apps or social media due to privacy concerns increases with age, income and education level. The largest gap exists between the youngest and the oldest generations (16 pp). On average, 65% of people aged 65 and older indicate that they avoid some online spaces due to privacy concerns compared to only 49% of those aged 18-24. In Colombia, the age gap reaches 28 pp, while in Norway the difference between the youngest and the oldest groups is only 3 pp.
People with the highest level of income and education more frequently (5 pp) avoid using certain websites, apps or social media due to privacy concerns than those with the lowest level of education and income. In Finland, people with tertiary education and those living in high-income households (top 20%) are 19 pp more likely to avoid certain websites and apps than Finnish people with the lowest levels of income and education. Conversely, people with tertiary education in Luxembourg (4 pp) and Spain (5 pp) are less likely to avoid online spaces because of privacy concerns. With a few exceptions, such as Japan (5 pp), men are also more likely (4 pp) to avoid using certain websites, apps or social media due to privacy concerns. This trend can be pronounced (i.e. Norwegian men are 13 pp more likely than women to avoid certain online spaces).
Control of personal data is another important aspect of trust. About one-third of people feel they do not have control over their personal information on line and one-third do (Figure 3.S.6).8 The remaining third neither feel in control nor not in control of their personal information. Cross-country variations emerge, with nearly half of Spaniards (49%) not feeling in control over their personal information compared to nearly one in four Japanese (24%).
Perception of control over personal information on line, 2024
Notes: Respondents were asked to indicate whether they agreed with the following statement: “I feel I have control over my personal information when using websites, apps or social media”. Possible responses include “strongly agree”, “agree”, “neither agree nor disagree”, “disagree” and “strongly disagree”. Social media sites and apps are surveyed as an aggregate.3 Respondents include all individuals aged 18 and older.
Source: Authors’ calculations based on the OECD Truth Quest Survey (OECD, 2024[1]).
On average, 31% of men and 35% of women do not feel in control of their personal information on line. In most of the countries, women are more likely than men not to feel they have control over their personal information when using websites, apps or social media. Only in Germany (0.5 pp) and the United States (2 pp) are men more likely than women to not feel in control of their personal information on line.
The share of individuals who feel they have control over their personal information when using websites, apps or social media tends to decrease with age. On average, 35% of those aged 18-24 feel they are in control over their personal information on line compared to 29% of those aged 65 and older. The largest gap between these two age groups is in the United States (32 pp), while the smallest gap is Italy, Norway and Switzerland (0 pp).
Media sources can increase trust by providing a high level of data protection and giving users control over their data. Further research into developments in the advertising and data protection landscape could be informative in this respect. For example, questions about the interplay between privacy concerns and practices that require consumers to accept the collection of data to purchase online subscription services by different demographic groups could be useful to explore, including in the context of data-driven advertising.
As concerns about the negative effects of false and misleading content intensify, policy makers increasingly focus on media literacy (Canadian Heritage, 2021[8]; US Department of State, 2022[9]). A range of entities engage in media literacy activities, including governments, schools, universities, online platforms and non-profit organisations. These aim to help individuals better assess and verify the accuracy of information on line. Media literacy initiatives tend to focus on developing cognitive, critical and technical skills that help discern fact from fiction, and enable meaningful participation in public interactions, discussions and debates.
As two of their core elements, media literacy initiatives advocate that individuals check facts and critically appraise the information they consume. A priori, one would expect that people who search for more information about a headline would be better able to identify its veracity. However, how often do people actually read more context about a headline, and does it influence their ability to identify true, and false and misleading content?
In the OECD Truth Quest Survey, respondents saw a news claim “headline” with the option to “read more” context about the claim.9 Overall, people clicked to read more context 29% of the time. People in France and Brazil clicked to read more context the least on average (21%), while people in Japan and the United Kingdom did so the most (39%) (Figure 3.S.7).10 Moreover, in Japan the share of people clicking to read more context in 20 or more claims reached 31%, while in Brazil it was only 8%.
Average percentage of times an individual clicked to read more context (left-hand scale) and overall Truth Quest score (right-hand scale), by gender, 2024
Notes: The percentages are calculated as the number of times a respondent clicked to read more context divided by the total number of claims seen (left-hand scale). The score is calculated as the total number of correct responses divided by the total number of claims seen (right-hand scale). A country score is thus an average of all respondents’ results and expressed as a percentage.
Source: Authors’ calculations based on the OECD Truth Quest Survey (OECD, 2024[1]).
On average, people with low education clicked to read more context 22% of the time, compared to those with secondary (28%) and tertiary education (33%). In all countries, women clicked to read more context (33%) more often than men (25%) on average. The share of people clicking on more context increases with the level of education in all countries except for Japan.
At the same time, exposure to more context increases the odds of correctly identifying the veracity of information on line in ten countries. Counterintuitively, it decreases the odds of correctly identifying the veracity of information on line in two countries (Finland and Portugal). The effect of exposure to additional context varies among content types (Figure 3.S.8).11 For satire, reading more context increases the odds of a correct answer by as much as 54%. Exposure to more context also increases the odds of correctly identifying true content (+12%). Conversely, exposure to more context for information classified as contextual deception leads to a lower accuracy rate (-20%).
Other studies have also found that not all aspects of traditional media literacy interventions always help people learn how to detect the veracity of information on line (Jones-Jang, Mortensen and Liu, 2021[10]; McGrew, 2024[11]). However, “lateral reading” has been identified as an important aspect of media literacy in the context of false and misleading content on line, although more research is needed in this area. Lateral reading is defined as the “strategy of leaving an unfamiliar website to search for information about a source’s credibility via additional sources” (McGrew, 2024[11]).
Odds ratio of correctly identifying the veracity of content, by type, 2024
Note: How to read the figure: A result of 1.54 means that the odds of a correct answer are 54% more likely after clicking to read more context.
Source: Authors’ calculations based on the OECD Truth Quest Survey (OECD, 2024[1]).
To foster a stronger information ecosystem, it is crucial to understand how people consume information and whether they trust it. This Spotlight shows that media consumption and trust levels in media sources vary considerably across demographic groups within and across countries. For example, while social media is an important source of information for many people – particularly young adults – it is also the least trusted source. Thus, social media companies and policy makers could consider developing mechanisms to make social media more trustworthy.
More trustworthy social media is important from a societal perspective because those who trust information from social media tend to have a lower ability to detect the veracity of information on line. On average across countries, those who trust social media a lot had a relatively lower overall Truth Quest score. As a result, governments and other actors may wish to design media literacy initiatives with this in mind, focusing on information on social media and how to better detect true and false content. Again, an understanding of demographic and country-specific differences can help target those most at risk.
As people increasingly consume information on line, either through news websites or social media platforms, media literacy can help people become more critical consumers of information.12 For example, data from this Spotlight show that on average, about one-third of the time people read more context about a headline in the accompanying text. Understanding people’s behaviour in this respect – i.e. people on average are not active searchers for more information – provides insights into the challenges faced by media literacy initiatives to help people become more critical consumers of information.
At the same time, analysis in this Spotlight suggests that reading more context about a headline does not always increase the odds of correctly identifying its veracity. This may indicate that media literacy initiatives should focus on encouraging people to seek additional information on third-party websites through lateral reading. It may also suggest that people simply do not want to expend additional time to verify their understanding, which poses other challenges for media literacy. More research into people’s online behaviour as well as more monitoring of the effectiveness of media literacy initiatives (OECD, 2024[12]) is needed to improve the information ecosystem.
References
[5] Al-Zoubi, A. (2024), “The impact of digital technology on media consumption: A case study of smartphone and app usage”, in Technology: Toward Business Sustainability. ICBT 2023. Lecture Notes in Networks and Systems, Vol. 925, Springer, Cham, http://dx.doi.org/10.1007/978-3-031-54019-6_39.
[4] Barthel, M. et al. (2020), “Measuring news consumption in a digital era”, 8 December, Pew Research Center, Washington, D.C., https://www.journalism.org/wp-content/uploads/sites/8/2020/12/PJ_2020.12.08_News-Consumption_FINAL.pdf (accessed on 8 July 2024).
[19] Basol, M. et al. (2021), “Towards psychological herd immunity: Cross-cultural evidence for two prebunking interventions against COVID-19 misinformation”, Big Data & Society, Vol. 8/1, https://doi.org/10.1177/20539517211013868.
[8] Canadian Heritage (2021), “Ongoing support for research and media literacy projects as Canada continues to fight online disinformation”, News Release, 9 February, Canadian Heritage, Gatineau, Quebec, https://www.canada.ca/en/canadian-heritage/news/2021/02/ongoing-support-for-research-and-media-literacy-projects-as-canada-continues-to-fight-online-disinformation.html (accessed on 1 August 2024).
[10] Jones-Jang, S., T. Mortensen and J. Liu (2021), “Does media literacy help identification of fake news? Information literacy helps, but other literacies don’t”, American Behavioral Scientist, Vol. 65/1, pp. 271-388, https://doi.org/10.1177/0002764219869406.
[13] Lesher, M., H. Pawelec and A. Desai (2022), “Disentangling untruths online: Creators, spreaders and how to stop them”, OECD Going Digital Toolkit Notes, No. 23, OECD Publishing, Paris, https://doi.org/10.1787/84b62df1-en.
[11] McGrew, S. (2024), “Teaching lateral reading: Interventions to help people read like fact checkers”, Current Opinion in Psychology, Vol. 55/101737, https://doi.org/10.1016/j.copsyc.2023.101737.
[3] OECD (2024), “Ability of adults to identify the veracity of content online”, OECD Going Digital Toolkit, https://goingdigital.oecd.org/indicator/80 (accessed on 28 June 2024).
[12] OECD (2024), Facts not Fakes: Tackling Disinformation, Strengthening Information Integrity, OECD Publishing, Paris, https://doi.org/10.1787/d909ff7a-en.
[17] OECD (2024), “OECD Information Integrity Hub”, webpage, https://www.oecd.org/en/blogs/2023/03/oecd-hub-on-information-integrity-joining-forces-to-fight-dis--and-misinformation.html.
[6] OECD (2024), OECD Survey on Drivers of Trust in Public Institutions – 2024 Results: Building Trust in a Complex Policy Environment, OECD Publishing, Paris, https://doi.org/10.1787/9a20554b-en.
[2] OECD (2024), “Online safety”, OECD Going Digital Toolkit, https://goingdigital.oecd.org/theme/10 (accessed on 28 June 2024).
[7] OECD (2024), “Share of adults who avoid using certain websites, apps, or social media due to privacy concerns”, OECD Going Digital Toolkit, https://goingdigital.oecd.org/indicator/84 (accessed on 18 November 2024).
[20] OECD (2024), “Share of adults who trust information from social media sites or apps”, OECD Going Digital Toolkit, https://goingdigital.oecd.org/indicator/83 (accessed on 18 November 2024).
[1] OECD (2024), “The OECD Truth Quest Survey: Methodology and findings”, OECD Digital Economy Papers, No. 369, OECD Publishing, Paris, https://doi.org/10.1787/92a94c0f-en.
[18] OECD (2022), “Misinformation and disinformation: An international effort using behavioural science to tackle the spread of misinformation”, OECD Public Governance Policy Papers, No. 21, OECD Publishing, Paris, https://doi.org/10.1787/b7709d4f-en.
[15] Polizzi, G. (2020), “Information literacy in the digital age: Why critical digital literacy”, in Informed Societies: Why Information Literacy Matters for Citizenship, Participation and Democracy, Facet Publishing, London, http://eprints.lse.ac.uk/102993/.
[14] UNESCO (2024), “Digital literacy”, webpage, https://uis.unesco.org/en/glossary-term/digital-literacy (accessed on 10 July 2024).
[16] UNESCO (2023), “Information Literacy”, webpage, https://www.unesco.org/en/ifap/information-literacy (accessed on 10 July 2024).
[9] US Department of State (2022), Media Literacy Design Manual, US Department of State, Washington, D.C., https://www.state.gov/wp-content/uploads/2022/08/Media-Literacy-Design-Manual-FINAL.pdf.
← 2. There are many types of literacy relevant to false and misleading content on line. Information literacy is an umbrella concept that applies to both analogue and digital media. It is defined as the ability “to seek, evaluate, use and create information effectively to achieve their personal, social, occupational and educational goal” (UNESCO, 2023[16]). The concept of digital literacy, which refers to the skills needed to use digital technologies effectively, as well as awareness of privacy, cybersecurity and critical thinking skills, is closely related. As a result, the term “media literacy” is often used to comprise the range of literacies needed to thrive in the digital age (Polizzi, 2020[15]).
← 3. Social media platforms are diverse and include a range of activities, including social networking, microblogging, messaging, and video-sharing, and content moderation policies vary across platforms. Perceptions of trustworthiness may be influenced by both the activities offered on a particular platform and the platform’s content moderation policies.
← 4. The figure reflects a simple average of the 21 country scores from the OECD Truth Quest Survey: Australia, Belgium, Brazil, Canada, Colombia, Finland, France, Germany, Ireland, Italy, Japan, Luxembourg, Mexico, Netherlands, Norway, Poland, Portugal, Spain, Switzerland, United Kingdom and United States.
← 5. The top four countries that trust information on social media some or a lot are: Colombia (64%), Poland (62%), Mexico (61%) and Brazil (57%).
← 6. The Truth Quest score is a measure of people’s ability (or skill) to detect the veracity of content on line. It is calculated as the total number of correct responses divided by the total number of news claims seen. A country score is an average of all respondents’ results. On average, respondents correctly identified the veracity of content 60% of the time. More information about the survey methodology can be found in OECD (2024[1]).
← 7. The figure reflects a simple average of the 21 country scores from the OECD Truth Quest Survey: Australia, Belgium, Brazil, Canada, Colombia, Finland, France, Germany, Ireland, Italy, Japan, Luxembourg, Mexico, Netherlands, Norway, Poland, Portugal, Spain, Switzerland, United Kingdom and United States.
← 8. The figure reflects a simple average of the 21 country scores from the OECD Truth Quest Survey: Australia, Belgium, Brazil, Canada, Colombia, Finland, France, Germany, Ireland, Italy, Japan, Luxembourg, Mexico, Netherlands, Norway, Poland, Portugal, Spain, Switzerland, United Kingdom and United States.
← 9. All headlines and additional context were derived from real-life news that was professionally fact-checked (except for the AI-generated claims, which were generated by GPT-4). More information can be found in OECD (2024[1]).
← 10. The average score is calculated as a simple unweighted average of the 21 country scores from the OECD Truth Quest Survey: Australia, Belgium, Brazil, Canada, Colombia, Finland, France, Germany, Ireland, Italy, Japan, Luxembourg, Mexico, Netherlands, Norway, Poland, Portugal, Spain, Switzerland, United Kingdom and United States.
← 11. The odds ratios are calculated by dividing the odds of a correct answer after clicking to read more context by the odds of a correct answer without reading the extra context. An odds ratio above 1 indicates that the chance of correctly identifying the veracity of information is increasing after clicking to read more context, and when it is below 1 it is decreasing. All types of content in the OECD taxonomy of false and misleading content were included (Lesher, Pawelec and Desai, 2022[13]). This figure shows the average, which is calculated as a simple unweighted average across the 21 country scores covered by the OECD Truth Quest Survey.
← 12. Media literacy can take many forms, including games (Basol et al., 2021[19]) and other mechanisms for providing media literacy tips (OECD, 2022[18]).