I Value It, but I Don’t Use It: Attitudes Toward Fact-Checking Among Portuguese University Students

Abstract

Fact-checking appeared as one of the responses to disinformation, establishing itself inside the journalism profession as another of its ethos. However, its social relevance relies on public perception and engagement. Despite growing evidence of approval, low familiarity and limited active use of fact-checking news remain common patterns, particularly among younger audiences. This study examines familiarity, contact, and attitudes toward fact-checking in a convenience sample of Portuguese university students, exploring associations with news consumption habits, political interest, skepticism, and ideological orientation through an online survey, with 356 university students, across different scientific areas. Results indicate that students show favorable attitudes toward fact-checking (M = 3.70) and recognize its social value but report moderate-to-low familiarity with its practices (M = 2.85), infrequent access (M = 2.37), and minimal sharing behavior (M = 1.82). Interest in reading fact-checking content emerged as the strongest predictor of positive attitudes (β = 0.506), outperforming familiarity and access frequency. Lower skepticism was associated with more favorable attitudes and showed no significant relationship with political orientation. Those more conservative displayed slightly less favorable attitudes. Our findings suggest that the primary challenge for fact-checking is not normative acceptance, but motivational engagement since favorable dispositions do not automatically translate into active consumption or sharing within everyday routines.

1. Introduction

The rise of fact-checking as a mechanism for public correction must be understood within the context of a broader crisis of authority, intensified by circulation on digital platforms and algorithmic curation. In this environment, misinformation cannot be treated as an isolated deviation and should instead be understood as a symptom of a communicational ecosystem in which factuality tends to be subordinated to beliefs, political identities, and rhetorical strategies, as well as to affective regimes of adherence to discourse (Lewandowsky et al., 2017). This shift from the centrality of fact to its appearance of plausibility contributes to the erosion of the reliability of sources and cumulatively compromises the collective ability to distinguish between verification and fabrication in the public space (Harsin, 2015; McIntyre, 2018).

It is in this terrain, traversed by symbolic disputes and tensions over authority, that fact-checking asserts itself as a practice. Fact-checking crystallizes a dual claim: on the one hand, it reinscribes itself in the classic journalistic ethos by relying on systematic and publicly justifiable verification procedures; on the other hand, it assumes itself as a normative response to the erosion of informational integrity and discursive responsibility in the public sphere (Graves, 2016; Amazeen, 2020; Cheruiyot & Ferrer-Conill, 2018). In fragmented ecosystems, where traditional editorial control loses its centrality and circulation is reorganized around platform dynamics, fact-checking tends to operate as a form of ex post epistemic control, whose legitimacy depends both on methodological robustness and on the public perception of transparency and impartiality.

While fact-checking has established itself as a normative response to disinformation and as a relatively institutionalized professional practice, its social relevance depends on a second level of analysis: that of public reception. In various contexts, there is simultaneously low familiarity with and reach of fact-checking and, at the same time, high levels of declared acceptance when the practice is presented or explained to respondents (Lyons et al., 2020; Robertson et al., 2020). This gap is theoretically relevant because it prevents practical integration from being inferred from normative attitudes and requires familiarity, contact, and evaluation to be treated as empirically distinct dimensions that are only partially convertible between each other (Kyriakidou et al., 2023; Lyons et al., 2020).

A second implication results from the socially stratified nature of this circulation. Familiarity, awareness of the existence of fact-checking, and demand for it tend to increase with factors such as age and education, and, particularly consistently, with political interest and knowledge (Saldaña & Santos, 2023; Bode & Vraga, 2021; Nyhan & Reifler, 2016).

In analytical terms, this shifts the focus to the question of conditions of access, predispositions, and information routines. The practice may be perceived as socially useful, but it remains peripheral if it does not cross barriers of visibility and integration into everyday news, even among audiences with higher formal education (Kyriakidou et al., 2023; Godel et al., 2022).

Finally, this issue of reception is not neutral from a political-identity point of view. Attitudes and acceptance can be modulated by motivated bias and partisan identities, even though evidence of backfire is considered rare (Nyhan & Reifler, 2010; Wood & Porter, 2019; Nyhan et al., 2019; Shin & Thorson, 2017). At the same time, some studies point to ideological asymmetry in resistance to and acceptance of fact-checking, with greater resistance associated with more conservative positions in certain contexts (Garrett & Bond, 2021; Jia & Lee, 2024). Taken together, these elements justify that the analysis of attitudes towards fact-checking includes not only indicators of trust and perceived effectiveness, but also dimensions of skepticism related to perceptions of bias, as well as political-institutional variables.

But what are the benefits of consuming fact-checking for the public? Several studies acknowledge that fact-checking has cognitive value, helping to reduce false beliefs (Porter & Wood, 2021), but it is also important for strengthening journalistic accountability and disciplining elites. Nyhan and Reifler (2015) noted that the practice of fact-checking improves the quality of political discourse, in the way it pressures and monitors politicians’ statements. On the other hand, Amazeen (2020) argues that fact-checking represents a democracy-building tool, particularly in countries with weaker democratic institutions. Furthermore, reading fact-checking is important because it can improve civic and political engagement (York et al., 2019; Bowles et al., 2025). York et al. (2019) showed that consuming fact-checking improves the perception of political reality. The sharing of fact-checking can also reveal that it is not just about disseminating a piece of information, but about helping to overcome the bottleneck of corrective distribution on social networks (Kim et al., 2022), even contributing to a public sphere based on truth, subject to verification and scrutiny (Westlund et al., 2024). At the local level, the practice of fact-checking can even help improve community journalism (Jerónimo & Esparza, 2023).

Despite the international expansion of fact-checking and the growth of the literature on its uses, perceptions, and effects, the materials previously studied emphasize that the available evidence remains heavily dependent on specific sociopolitical contexts and media ecologies, which limits direct generalizations and reinforces the need for situated knowledge, especially outside WEIRD environments (Vinhas & Bastos, 2025). In the case of southern Europe, and in particular Portugal, we find a relatively late development of fact-checking and a focus on independent initiatives, often associated with electoral cycles and tensions typical of media systems with political parallelism (Hallin & Mancini, 2004; Graves & Cherubini, 2016; Moreno-Gil et al., 2022; Baptista et al., 2022b). Although the national literature is already beginning to map the phenomenon, including studies on public perceptions and attitudes towards fact-checking (Baptista et al., 2023) and exploratory analyses of professional practices and challenges (Jerónimo & Esparza, 2023), it remains pertinent to further explore, with quantitative instruments and a focus on specific audiences, the relationship between familiarity, contact, attitudes, and political-institutional variables in the Portuguese context.

In Portugal, although the practice of fact-checking is relatively recent (Baptista et al., 2023), in recent years fact-checking initiatives have multiplied through the creation of networks and partnerships between platforms and media outlets (Cardoso et al., 2024). The first fact-checking project began as an initiative of the digital newspaper Observador in 2015, joining the IFCN (The International Fact-Checking Network) in 2017. A year later, in 2018, Polígrafo was born, the first project dedicated exclusively to fact-checking. Launched during the international Websubmite conference, Polígrafo quickly became popular, notably through its partnership with the television channel SIC, creating the Polígrafo SIC segment. Currently, Polígrafo works in partnership with the television channel TVI, with the Polígrafo TVI program, broadcast every Monday. In addition to these projects, the Público newspaper also created a section dedicated to fact-checking, called Prova dos Fatos (Proof of Facts), developing a more contextual and explanatory approach. Alongside these initiatives, the Lusa news agency launched Lusa Verifica (Lusa, 2025), a new fact-checking service (Oliveira, 2020; Cardoso et al., 2024; Baptista et al., 2023).

This is precisely where the present study comes in. By focusing on Portuguese university students, the study allows us to observe a segment that, in principle, combines higher formal education and intense digital exposure, but whose familiarity and practical integration of fact-checking should not be assumed to be linear (Lyons et al., 2020; Robertson et al., 2020).

Thus, this article seeks to contribute to characterizing levels of familiarity, contact, and attitudes toward fact-checking, exploring associations between these dimensions and patterns of news consumption and political interest, and integrating dimensions of skepticism and political-institutional framing in the reading of these relationships.

Given the still heterogeneous nature of the evidence on familiarity, circulation, and acceptance of fact-checking, this study takes an exploratory approach that favors descriptive mapping and the identification of associations between central dimensions of public reception. The analysis is guided by a set of research questions:

RQ1. 

What are the levels of familiarity, contact, and attitudes toward fact-checking among Portuguese university students?

RQ2. 

To what extent are news consumption habits and political interest associated with greater familiarity and contact with fact-checking?

RQ3. 

To what extent are familiarity and contact related to more favorable attitudes, including trust, perceptions of impartiality, and perceived effectiveness?

RQ4. 

What role do skepticism and political-institutional variables play in shaping these associations?

2. Young People, Informational Literacy, and News Consumption

Young adults, and in particular university students, constitute a privileged observatory to capture the contemporary reconfiguration of news consumption and, by extension, the social conditions in which contact with corrective practices such as fact-checking occurs. The accumulated evidence suggests that these audiences organize access to information mainly from digital platforms and networks, reducing the centrality of traditional journalistic brands in their daily repertoire (Swart, 2021; Reuters Institute for the Study of Journalism, 2025). It is important, however, to resist the automatic deduction that associates youth and schooling with higher levels of literacy or more demanding assessment standards. The literature describes, with greater nuance, literacy as a situated and contextual practice, operated in ordinary routines and under specific constraints of time, attention and informational environment. In this context, Marchi (2012) shows that many young people not only distance themselves from classic journalistic norms, such as objectivity, but also reorient their credibility criteria towards markers of authenticity, proximity and social identification.

The centrality of these practices makes it particularly relevant to distinguish between media literacy, understood as the ability to evaluate sources, identify biases, and recognize journalistic criteria for credibility, and algorithmic literacy, that is, understanding the curation and ordering mechanisms that structure information exposure in platform environments. The news experience is often mediated by personalized feeds and network circulation that shifts attention from the medium to the flow, making the origin, framing, and authority of messages less transparent (Thorson & Wells, 2016; van Dijck et al., 2018). Studies on user beliefs about algorithmic curation show, for example, that many young people have incomplete or intuitive interpretations of how algorithms select content, which can affect how they assign relevance, trust, and the need for verification to information they find in feeds (Rader & Gray, 2015). In this context, recognizing when a claim requires verification and identifying fact-checking as a relevant resource depends not only on traditional critical skills, but also on the ability to interpret the very architecture of visibility and authority constructed by platforms (Livingstone, 2018; van Dijck et al., 2018).

This gap in algorithmic literacy has consequences that go beyond the technical understanding of the feed and affect the self-perception of critical competence. One point emphasized in the literature is the illusion of informational competence, that is, the tendency to overestimate one’s ability to evaluate information and resist misinformation, even when operating with incomplete interpretations of the mechanisms that organize content visibility (Rader & Gray, 2015; Van Marcke et al., 2025). In this case, subjective confidence in one’s ability to distinguish can function as a cognitive shortcut to reduce the deliberate search for external verification and make it more likely that evaluation will depend on criteria of immediate plausibility rather than consistent validation routines (Cho & Woo, 2025; Walter et al., 2021).

From here, it becomes more intelligible why familiarity and even favorable attitudes towards fact-checking do not convert, by themselves, into active involvement, either in sharing verifications or in reporting false content. Among young people, news consumption is often structured by curated feeds and incidental contacts with information, which shifts attention to a logic of intermittent and fragmented exposure. In this scenario, news avoidance and information overload function as structural barriers to sustained attention and participation, reducing the likelihood of any gesture that requires time, energy, and cognitive availability (Boczkowski et al., 2018; Fletcher & Nielsen, 2019; Skovsgaard & Andersen, 2020; Song et al., 2017). It is precisely at this point that corrective practices reveal their demand: fact-checking presupposes the recognition of the problem, a minimum cognitive investment to suspend immediate plausibility, and, in many cases, the willingness to make a judgment public, for example by reporting or sharing. Therefore, it tends to focus on more motivated segments, or on moments of heightened attention, rather than stabilizing as a generalized, everyday routine (Fletcher & Nielsen, 2019; Skovsgaard & Andersen, 2020).

3. Fact-Checking as a Media Practice

While verification and corrections have long been part of journalistic practice, contemporary fact-checking represents a more recent specialization and public-facing format, often organized through dedicated teams, sections, or independent outlets. The literature describes this practice not only as an informational product, but as a set of routines and transparency mechanisms that seek to make verification a visible, auditable, and socially legitimate procedure, reinforcing the accountability of public actors and the public correction of errors (Graves, 2016; Amazeen, 2020; Cheruiyot & Ferrer-Conill, 2018). In this sense, fact-checking operates as a professional and normative response to an informational environment marked by saturation of misleading content and erosion of trust, seeking to re-inscribe factuality in the public circulation circuit based on explicit criteria (Graves, 2016; Amazeen, 2020).

Its institutionalization, however, is not uniform and tends to take on different organizational models, with implications for visibility and perceived authority. A recurring distinction separates initiatives based in newsrooms, closer to traditional journalistic routines, from independent or NGO-format initiatives, often structured as specialized organizations, with their own verification repertoires and specific strategies for public legitimation (Graves & Cherubini, 2016). In southern European contexts, the relevance of independent projects has been highlighted, sometimes driven by electoral circumstances and symbolic competition dynamics in the media ecosystem, which reinforces that the reception of fact-checking depends not only on the method, but also on the institutional place from which the verification is enunciated (Moreno-Gil et al., 2022).

This position as a corrective practice, however, makes fact-checking structurally contested. By claiming authority and publicizing verification criteria, fact-checking exposes itself to accusations of bias and selectivity, especially in polarized environments (Tandoc et al., 2018; Waisbord, 2018). Thus, the credibility of fact-checking depends simultaneously on methodological robustness and the ability to maintain an image of impartiality before heterogeneous audiences, in a field where factual validation rules can be appropriated as a resource for symbolic struggle (Waisbord, 2018; Graves, 2016).

Attitudes Toward Fact-Checking: Trust, Credibility, and Skepticism

Attitudes toward fact-checking must be understood as a decisive factor in its ability to mobilize the public, emphasizing that credibility acts as a mediating mechanism, while individual differences in trust and assessments of the type of source can modulate its reception (Liu et al., 2025; Kyriakidou et al., 2023; Brandtzaeg & Følstad, 2017). From this perspective, measuring attitudes would involve capturing dimensions such as trust and impartiality and assuming that these dimensions are linked to broader predispositions of media trust and assessments of the institutional legitimacy of those who verify (Liu et al., 2025; Kyriakidou et al., 2023).

Skepticism, in turn, is often associated with perceptions of bias and suspicions about the real motivations of fact-checkers (Taber & Lodge, 2006; Walter et al., 2021). The concept of skepticism mobilized here differs from the generalized skepticism towards news, described in the context of hybrid consumption as a diffuse distrust of mediated information in general (Fletcher & Nielsen, 2019). Here, skepticism refers specifically to distrust of the motivations, impartiality, and credibility of the fact-checkers themselves. A significant portion of the literature discusses how partisan identities and motivated reasoning processes can condition the acceptance of corrections, generating selective resistance and a predisposition to interpret fact-checking as an instrument of ideological favoritism (Nyhan & Reifler, 2010; Kraft et al., 2015; Lyons et al., 2020). At the same time, some studies point to patterns of ideological asymmetry in certain contexts, in which more conservative audiences show greater resistance or lower acceptance of fact-checking, although the degree and consistency of this effect vary between samples and countries (Garrett & Bond, 2021; Lyons et al., 2020; Jia & Lee, 2024; Aruguete et al., 2025). Thus, measuring skepticism should not be treated as a mere inverse of positive attitude, but as a specific dimension that captures how the fact-check is read through the lens of bias and contested authority (Lyons et al., 2020; Aruguete et al., 2025).

It is important, however, to avoid simplistic readings that take resistance as inevitable or assume the robust existence of widespread counterproductive effects. The literature has been critically reevaluating the so-called backfire effect, suggesting that it is less prevalent than some initial formulations indicated and that, in many cases, corrections can produce effects even in politically charged environments (Wood & Porter, 2019; Nyhan et al., 2019). At the same time, studies on attitudes and reach show recurring patterns that are particularly relevant to reception studies, such as low familiarity and low reach, but high stated favorability when fact-checking is presented, as well as greater use among younger and more educated segments (Lyons et al., 2020; Baptista et al., 2023; Kyriakidou et al., 2023; Robertson et al., 2020). Together, these results support the need to analyze attitudes, skepticism, and practices as empirically distinct dimensions, whose relationship depends on political predispositions, trust, and conditions of visibility and integration into information routines (Baptista et al., 2023; Robertson et al., 2020).

4. Methods

The main objective of our study was to understand the attitudes and perceptions of Portuguese university students regarding the practice of fact-checking. We sought to analyze their degree of familiarity with, contact with, and attitudes toward fact-checking, associating it with levels of credibility and trust in the media, behavioral patterns of consumption, sharing, and verification of information, and the students’ political party orientation.

To this end, an exploratory, self-reported, convenience-based online survey was applied to Portuguese university students from various scientific fields. This study was therefore based on a non-probabilistic convenience sample recruited in classrooms. No quota-based procedures were used to balance gender or disciplinary area. A total of 356 students participated voluntarily and anonymously in our study. The survey took place between 20 October and 8 December 2025.

The final sample is disproportionately female and concentrated in the Social Sciences and Humanities, which should be taken into account when interpreting the results. All statistical analyses were performed using SPSS 29.0.1.1 (244) (IBM SPSS, Chicago, IL, USA).

4.1. Participants

Our sample is composed of students with a mean age of 22.61 years (SD = 7.92). In the sample, 245 (68.8%) are women and 107 (30.1%) are men, with the majority of participants attending undergraduate programs (73.9%) at public universities (73.9%). The respondents belong to different scientific areas: 73.6% from Social Sciences and Humanities; 13.8% from Health; 4.8% from STEM (Science, Technology, Engineering, Mathematics) courses; and 0.6% from Environment and Agriculture.

4.2. Procedures

The questionnaire was organized into seven thematic blocks: (1) sociodemographic questions; (2) information consumption; (3) contact with fact-checking; (4) familiarity with fact-checking; (5) attitudes toward fact-checking; (6) skepticism toward fact-checking; and (7) political-institutional variables (Table 1).

Table 1. Survey structure: thematic blocks, objectives, and key question points.

Section/Thematic Block	Objective	Key Question Points
1. Sociodemographic characterization	Describe the participant profile	Age; gender; scientific area; academic degree
2. Information and news consumption	Measure informational habits and media exposure	Frequency of news reading; time dedicated to news; intention to share; interest in following current events
3. Contact with fact-checking	Assess actual exposure to fact-checking	Frequency of access to fact-checking content; frequency of contact; frequency of sharing
4. Familiarity with fact-checking	Measure perceived knowledge about the phenomenon	Degree of knowledge; concept recognition; prior contact
5. Attitudes toward fact-checking	Assess general perception and normative evaluation	Perceived usefulness; social importance; credibility; impact on disinformation; valuation of the role of fact-checking
6. Skepticism toward fact-checking	Measure distrust and critical perceptions	Perception of journalist favoritism; favoritism toward groups; bias; lack of neutrality
7. Political and institutional context	Frame attitudes within the political profile of students	Interest in politics; institutional trust; ideological positioning

Regarding the measures used in the survey, we employed items and formulations already validated by the literature in other studies that sought to understand citizens’ attitudes toward fact-checking, namely to measure the level of familiarity (1—Never heard of it, 5—Very familiar) (Robertson et al., 2020) and attitudes toward fact-checking such as credibility, skepticism, and trust (Baptista et al., 2023; Liu et al., 2025). In this sense, for a more rigorous analysis, four composite indices were created from variables on Likert-type scales: (1) General Attitude; (2) Level of Skepticism; (3) General Engagement; and (4) Political Interest. It should be noted that the construction of the indices was carried out by calculating the mean of responses to the corresponding items. Nevertheless, whenever necessary, we proceeded to recode the items to ensure interpretive coherence (Table 2).

Table 2. Composite indices: formulas, reliability, and component questions.

Indices	Formulas	Cronbach’s α	Questions
1. General Attitude	Attitude_G = MEAN (Q20, Q21, Q22, Q24, Q27, Q28)	0.885	Q.20. Trust in fact-checking; Q.21. Importance of fact-checking for democracy; Q.22. Impartiality; Q.24. Efficacy; Q.27. Fact-checking should be more widely disseminated; Q.28. Fact-checking helps hold politicians accountable
2. Level of Skepticism	Skepticism = MEAN (6 − Q15, 6 − Q26)	0.558	Q.15. Journalists favor one side; Q.26. Fact-checkers favor ideologies
3. General Engagement	Engagement = MEAN (Q12, Q13, Q36, Q37)	0.701	Q.12. Frequency of access; Q.13. Frequency of sharing; Q.36. Shared fact-checking; Q.37. Reported false content
4. Political Interest	Interest_pol = MEAN (Q31, Q32, Q33)	0.852	Q.31. Interest in political news; Q.32. Follows politics; Q.33. Enjoys talking about politics

Note: Cronbach’s alpha is not very effective for measuring only two items, so we performed a Spearman correlation on Q15 and Q26, which measure the level of skepticism, with (ρ = 0.370, p < 0.001).

Regarding the questions associated with habits, interests, and information consumption (Baptista et al., 2022a), our research also seeks to ascertain the relationship established between habits, interests, and practices of consuming information and the perceptions and different levels of use and attitude toward fact-checking. To this end, we asked participants, using 5-point Likert scales, about reading frequency (1—Never, 5—Very frequently), time dedicated to news consumption, active participation with news, and interest in news.

In addition to the composite indices, several variables were analyzed at the item level, as they represent analytically distinct dimensions of fact-checking reception and news consumption. These included familiarity with fact-checking, interest in reading fact-checking articles, frequency of access to fact-checking content, frequency of sharing fact-checking content, frequency of reading news, time spent consuming news, and sharing news on social media.

The composite indices were used only when conceptually broader dimensions were being measured, namely, General Attitude, Level of Skepticism, General Engagement, and Political Interest.

With regard to the measurement of political–ideological orientation, we invited students to self-position on a political spectrum of two dimensions: left–right (1—Very left, 7—Very right) and liberal–conservative (1—Very liberal, 7—Very conservative). Two scales were used so as not to limit the perception of each student’s political values to a single, closed, and horizontal political dimension on the political spectrum. People are motivated by their social, ideological, and partisan values when they self-place on this type of scale (Baptista & Loureiro, 2018; Freire, 2006; Freire & Belchior, 2011; Inglehart & Klingemann, 1976).

5. Results

5.1. Descriptive Analysis

Overall, respondents in this sample reported quite positive attitudes toward fact-checking (M = 3.70, SD = 0.76) and favorable (M = 3.93, SD = 0.96) attitude toward the practice of fact-checking and the social role it represents (Table 3). However, despite recognizing the importance of fact-checking, university students present a moderate–low level of familiarity with this practice (M = 2.85, SD = 1.15), do not seem to regularly access this type of content (M = 2.37, SD = 1.06) and do not habitually share it on their social media (M = 1.82, SD = 1.01) (Table 3). In fact, our results seem to point to a discrepancy between attitudes and effective behaviors. It is also paradoxical that interest in reading fact-checking articles presents, on average, higher values (M = 3.08, SD = 1.02) and that, on the part of students, there is a relatively high consumption of news in general (M = 3.56, SD = 0.91). We also found that the average time spent consuming news is comparatively low (M = 1.83, SD = 0.88), which may indicate that students may be exposed to news with some regularity, but tend to interact with it only for a short time. Additionally, and similarly to the habit of sharing fact-checking articles on social media, news sharing is also not frequent (M = 2.06, SD = 1.0). On the other hand, our results show that political interest today is at a moderate level (M = 3.30, SD = 0.95), indicating that the low engagement with fact-checking should not be attributed simply to widespread political disinterest.

Table 3. Descriptive statistics for key fact-checking, news consumption, and political-interest variables.

	N	Mean	Std. Deviation	Sk	Ku
Level of familiarity	356	2.85	1.150	0.143	−0.782
Reading interest	356	3.08	1.020	0.054	−0.324
Access frequency	356	2.37	1.058	0.271	−0.751
Favorability	356	3.93	0.968	−0.671	0.083
Frequency of sharing	356	1.82	1.011	1.013	0.086
Reporting falsehoods	356	1.41	0.493	1.235	0.445
News reading frequency	356	3.56	0.909	−0.069	−0.584
Time dedicated to news consumption	356	1.83	0.883	1.03	0.836
News sharing on social media	356	2.06	1.00	0.491	−0.708
Political interest	356	3.30	0.953	−0.324	−0.355

In this way, in a first descriptive approach, our results present university students as a passive informational consumption audience, which, while recognizing the importance of fact-checking, appears poorly familiarized with the journalistic concept and with little disposition to engage with the content on social media. In a comparative analysis of news reading frequency, time spent on news, and weak sharing habits, we note that information consumption is low, with fact-checking content remaining peripheral. Indeed, our analysis also revealed that 58.7% of students have never flagged or reported false content, although 41.3% have done so at least once. In the last three months, only 39% of students report having shared fact-checking content at least once, a result consistent with the sharing habits mentioned previously.

In this sense, our results once again reinforce the idea of high recognition of fact-checking, but low active engagement.

5.2. News Consumption and Political Interest as Correlates of Familiarity and Contact with Fact-Checking

In order to understand whether broader news consumption habits and the level of political interest affect the level of familiarity with fact-checking practices and their more frequent use, we calculated Spearman’s correlation coefficient (Table 4). Our results, and answering RQ2, show that all variables considered have statistically significant and positive associations with the level of familiarity and/or frequency of access.

Table 4. Spearman correlations between news consumption, political interest, familiarity, and access to fact-checking.

Variable	1	2	3	4	5	6	7
1. News reading frequency	—
2. Time dedicated to news consumption	0.383 **	—
3. News sharing on social media	0.190 **	0.130 *	—
4. Interest in reading fact-checking articles	0.215 **	0.256 **	0.005	—
5. Political interest	0.380 **	0.304 **	0.242 **	0.362 **	—
6. Familiarity with fact-checking	0.351 **	0.294 **	0.136 *	0.392 **	0.350 **	—
7. Access frequency to fact-checking content	0.248 **	0.314 **	0.151 **	0.554 **	0.330 **	0.484 **	—

Note: Spearman’s rho; N = 356. * p < 0.05; ** p < 0.01.

Therefore, analyzing Table 4, we note that the more frequently news is read, the greater the degree of familiarity with fact-checking (ρ = 0.351, p < 0.001) and, consequently, the higher the frequency of access (ρ = 0.248, p < 0.001). Therefore, the data suggest that students who follow the news more regularly tend to be not only more familiar with the practice of fact-checking, but also access its content more frequently. The time spent consuming news is also significantly and positively associated with familiarity (ρ = 0.294, p < 0.001) and frequency of access (ρ = 0.314, p < 0.001). The same behavior, although with weaker associations, is observed in relation to social sharing. Political interest also shows positive and statistically significant associations with familiarity (ρ = 0.350, p < 0.001) and frequency of access (ρ = 0.330, p < 0.001), which, in some way, reinforces the idea that fact-checking is more likely to become part of the routine of students who are more attentive to public affairs.

5.3. Familiarity, Reading Interest, Access Frequency, and Attitudes Toward Fact-Checking

Through the use of Spearman’s correlation coefficient, we verified what association is established between the level of familiarity and the behavior and attitudes associated with fact-checking, namely reading interest and access frequency (Table 5).

Table 5. Spearman correlations between familiarity, reading interest, access frequency, and attitude.

	Familiarity	Reading Interest	Access Frequency	Attitude
Familiarity	1.000
Reading interest	0.392 **	1.000
Access frequency	0.484 **	0.554 **	1.000
Attitude	0.305 **	0.540 **	0.301 **	1.000

Note: ** p < 0.01.

Analyzing Table 5, we observe that there is a moderate positive correlation between familiarity with fact-checking and interest in reading fact-checking content (ρ = 0.392, p < 0.001), which indicates that the more familiarized students are with fact-checking, the greater interest they tend to demonstrate. We also verified that those who know fact-checking better tend to effectively access its content with greater regularity (ρ = 0.484, p < 0.001). Furthermore, interest and access frequency are also strongly correlated (ρ = 0.554, p < 0.001), which in some way demonstrates that declared interest is an excellent predictor of effective access behavior. Therefore, if on one hand unfamiliarity can be a barrier to consumption, on the other hand, when interest exists, it translates into action, namely in terms of access to fact-checking content.

At the level of the general attitude toward the practice of fact-checking, we verified that knowing fact-checking can help generate more positive attitudes (ρ = 0.305, p < 0.001), although it does not seem sufficient on its own. It is important to note that interest in reading is more strongly related to positive attitudes (ρ = 0.540, p < 0.001) than the level of familiarity. This leads us, once again, to suggest that motivation or psychological engagement with fact-checking is an important element for its legitimation.

In a third moment of our analysis, we performed a multiple linear regression to verify in what way the level of familiarity with fact-checking, reading interest, and access frequency can serve to explain attitudes toward fact-checking. In the multiple regression, the model revealed that the three variables present statistically significant differences regarding attitudes toward fact-checking (F_(3;352) = 48.021, p < 0.001), explaining 29% of the variance in attitudes (R² = 0.29). The coefficient table demonstrates that interest in reading fact-checking (β = 0.506, p < 0.001) is the only variable with significant values, demonstrating it to be the strongest predictor by far for presenting positive attitudes (Table 6).

Table 6. Multiple linear regression: predictors of attitudes toward fact-checking.

Predictor	B	SE B	β	t	p	95% CI for B
(Constant)	1.573	0.244	—	6.439	<0.001	[1.09, 2.06]
Familiarity	0.074	0.036	0.111	2.034	0.043	[0.003, 0.145]
Interest in reading fact-checks	0.506	0.065	0.506	7.785	<0.001	[0.379, 0.633]
Frequency of access	0.089	0.049	0.098	1.816	0.070	[−0.008, 0.186]

R² = 0.29; adjusted R² = 0.28; F_{(3, 352)} = 48.021, p < 0.001; N = 356.

In order to understand how this composite of variables can relate to the level of skepticism (inverted skepticism index) presented toward the practice of fact-checking, Spearman correlations revealed that the greater the skepticism, the less positive the attitude toward fact-checking (ρ = 0.297, p < 0.001). Positive, albeit weak, associations were also found with interest in reading verification content (ρ = 0.149, p = 0.005) and with access frequency (ρ = 0.117, p = 0.028). No significant relationship was found with the level of familiarity with fact-checking (ρ = 0.093, p = 0.078) (Table 7). Curiously, skepticism did not present significant associations with interest in political issues, nor with the level of institutional trust or with the level of engagement or interaction with fact-checking. This, in some way, suggests that distrust toward fact-checkers is not strongly anchored in political or institutional orientations.

Table 7. Spearman correlations between skepticism and study variables.

Variable	ρ (Spearman)	p
Attitudes	0.297	<0.001
Interest in reading	0.149	0.005
Frequency of access	0.117	0.028
Familiarity	0.093	0.078
Political interest	0.052	0.318

With the analysis of Table 8, one also perceives that knowing fact-checking increases engagement (ρ = 0.411), increases positive attitudes (ρ = 0.305), is linked to interest in politics (ρ = 0.350), but does not reduce distrust.

Table 8. Spearman correlations between familiarity, engagement, political interest, and attitudes.

Variable	1	2	3	4
1. Familiarity	—
2. Engagement	0.411 **	—
3. Political interest	0.350 **	0.406 **	—
4. Attitudes	0.305 **	0.239 **	0.276 **	—

** p < 0.01.

5.4. Effects of Sociodemographic and Political Factors

In order to understand whether men and women have different attitudes toward fact-checking, a t-test for independent samples was performed. The analysis did not reveal statistically significant differences between men and women [t₍₃₅₀₎ = −0.09, p = 0.926], with both genders presenting very similar means (M men = 3.59; M women = 3.70) (Table 9). Furthermore, through a one-way ANOVA, we sought to verify whether the scientific area of students had an effect on attitudes toward fact-checking, and no statistically significant differences were evidenced (F_(3;326) = 0.595, p = 0.619). Post hoc multiple comparison tests confirmed the absence of significant differences between any pairs of areas.

Table 9. Sociodemographic and political effects on attitudes toward fact-checking.

Variable	N	M	SD	Test Statistic	p	Effect Size/ρ
Gender				t(350) = −0.092	0.926	d = −0.011
Male	107	3.69	0.767
Female	245	3.70	0.767
Scientific Area				F(3;326) = 0.595	0.619	η2 = 0.005
Social Sc. & Humanities	262	3.70	0.729
STEM	17	3.70	1.094
Health	49	3.71	0.760
Environ. & Agriculture	2	4.41	0.118
Pol. Orient. (ρ with Attitudes)
Liberal–Conservative	356	—	—	—	0.030	−0.115 *
Left–Right	356	—	—	—	0.607	−0.027
Pol. Orient. (ρ with Skepticism)
Liberal–Conservative	356	—	—	—	0.112	−0.084
Left–Right	356	—	—	—	0.333	−0.051

* p < 0.05.

In order to understand if the gender and scientific area of the students could have any relation with the level of familiarity and engagement with fact-checking, non-parametric tests were carried out. Using the Mann–Whitney test, we revealed a statistically significant difference in familiarity by gender (U = 10,672.50, Z = −2.861, p = 0.004), with men showing higher averages than women (199.26 vs. 166.56). No statistically significant difference between genders was found for overall engagement (U = 12,796.00, Z = −0.357, p = 0.721). Similarly, Kruskal–Wallis tests did not reveal statistically significant differences between the scientific areas, either for familiarity (H(3) = 1.806, p = 0.614) or for engagement (H(3) = 1.349, p = 0.718).

These results suggest that no statistically significant differences were observed between the subgroups regarding attitudes towards fact-checking in this sample. On the other hand, through additional non-parametric analyses, our findings indicate that gender is associated with familiarity with fact-checking, although no significant differences were found regarding engagement or between scientific fields. However, it is important to note that these comparisons between subgroups should be interpreted with caution, considering that they have unequal sizes.

In this way, we can observe that fact-checking is more of a political-cognitive phenomenon than a sociodemographic one. Indeed, our results further indicate that there is a statistically significant negative correlation between the general attitude toward fact-checking and political orientation on the liberal–conservative scale (ρ = −0.115, p = 0.030), suggesting that more conservative students tend to present slightly less favorable attitudes. On the other hand, we found no significant evidence with positioning on the left-right political scale (ρ = −0.027, p = 0.607).

Furthermore, we also found no statistically significant associations between the level of skepticism toward fact-checking and the ideological positioning of students, whether on the liberal–conservative scale (ρ = −0.084, p = 0.112) or on the left–right scale (ρ = −0.051, p = 0.333). Therefore, these results indicate that skepticism does not appear to be related to the political self-identification of university students.

6. Discussion

Our results demonstrate that young university students are globally favorable toward the practice of fact-checking, although they are not very familiarized with fact-checking and present low levels of engagement. This paradox leads us to identify a peripheral use of fact-checking by students in their information routines. Therefore, there is recognition, but there is no action that translates into engagement with the content.

Thus, our findings approximate those of other studies that analyzed people’s behavior toward fact-checking and their consumption habits (Robertson et al., 2020; Kyriakidou et al., 2023; Lyons et al., 2020; Baptista et al., 2023; Amazeen et al., 2019). In fact, fact-checking emerges with high approval, including in several European countries (Lyons et al., 2020), but may remain marginal in audiences’ daily lives.

With the low levels of familiarity presented, our study does not corroborate the idea, sustained by other analyses, that young people are more aware of the practice of fact-checking (Robertson et al., 2020; Baptista et al., 2023). Being university students, with high levels of education, higher values of knowledge or familiarity would be expected.

On the other hand, the peripheral use of fact-checking and the low-intensity pattern of news engagement observed in our sample do not seem exclusive to the Portuguese context (Hidalgo-Cobo et al., 2025; Almenar et al., 2022; Vázquez-Herrero et al., 2022; Amazeen et al., 2019). Indeed, phenomena such as news avoidance (Aharoni et al., 2021; Kalogeropoulos & Nielsen, 2017; Lindell & Mikkelsen Båge, 2023) and incidental news consumption (Boczkowski et al., 2018; Park & Kaye, 2021) seem more common among young people, which in some way may help explain our results.

Interestingly, our results also indicate that the consumption of fact-checking, and consequently, knowledge of the practice, is not dissociated from broader informational routines. In fact, students who tend to read news more frequently, dedicate more time to news, and also show greater political interest tend to report a higher level of familiarity with fact-checking (Robertson et al., 2020; Lyons et al., 2020). Furthermore, these students also access verification content more frequently. Therefore, our data seem to suggest that the consumption of fact-checking content can be integrated into the routine of students who are more connected to news and informed about public affairs (albeit marginally). Thus, the problem does not seem to lie in a lack of interest in news, but rather in the weak integration of fact-checking consumption into information consumption habits.

It is also important to note that the weaker associations involving the sharing of news on social media are consistent with what other studies have shown about the sharing of fact-checks, which depends on more specific political and motivational backgrounds than mere general exposure (Amazeen et al., 2019).

Our study found that male students tend to be more familiar with fact-checking. A possible explanation for this trend can be found by considering other studies that have identified gender disparities regarding political interest and political knowledge, with men tending, on average, to report higher levels in both dimensions. Furthermore, these resources, in turn, are positively related to attention given to political news and the use of news, which may increase familiarity with fact-checking as a verification tool (Fraile, 2014; Strömbäck & Shehata, 2010).

A complementary interpretive hypothesis, not directly tested in this study but theoretically grounded, may help contextualize these findings. The literature on informational literacy suggests that subjective confidence in one’s own ability to evaluate information can function as a cognitive shortcut that reduces the perceived need for external verification tools such as fact-checking (Rader & Gray, 2015; Kahneman, 2011). This phenomenon, described as the illusion of informational competence, affirms that individuals who see themselves as critical readers may underestimate their vulnerability to misinformation, making formal verification mechanisms less salient in their routines (Nyhan & Reifler, 2010; Van Marcke et al., 2025). Given that our sample is composed of university students, a group where such self-perceptions of critical sufficiency could plausibly be more pronounced, this mechanism might partially account for the observed gap between favorable attitudes and low active engagement, without necessarily implying indifference or civic disinterest. However, since the self-perception of informational competence was not measured in this study, this remains a speculative avenue that warrants future investigation.

Therefore, if on one hand we find an audience receptive and favorable to fact-checking, on the other hand, we verify that this audience does not seek and does not interact with news (comments or shares) and fact-checking articles. Faced with this scenario, journalism faces an enormous challenge. We believe it becomes crucial to present journalistic content that corresponds to the expectations of young people, namely with new formats and more attractive designs.

Along this line of thought, our results demonstrate that, effectively, the main challenge of fact-checking does not lie in its normative acceptance, but rather in seeking to join efforts to find strategies of motivational mobilization for its active consumption. Our correlations showed that the strongest associations, related to attitude and level of familiarity and access to fact-checking, involve variables more directly linked to motivation, that is, the interest in reading fact-checking articles. In our analysis, interest in reading fact-checking content was the strongest predictor of positive attitudes. Although the levels of familiarity and engagement are also associated with positive attitudes, both present a more reduced effect. Furthermore, positive attitudes also do not predict, on their own, an active action of access or sharing. This pattern therefore suggests the importance of motivational logic in fact-checking consumption.

In some way, this consumption logic suggests that fact-checking will be at a disadvantage compared to the consumption of other types of information, which instead follows a logic of incidental consumption or unintentional exposure (Fletcher & Nielsen, 2019). Furthermore, in the current digital ecosystem, where the attention economy prevails, fact-checking may become even more vulnerable. This is because fact-checking requires more sustained attention for being reactive and explanatory.

Similarly to other studies (Robertson et al., 2020; Baptista et al., 2023), our results further confirm the central role of political interest as a predictor of familiarity and positive attitudes toward fact-checking. This may be related to the fact that political knowledge is associated with a greater capacity to detect falsehoods (Vegetti & Mancosu, 2020) or with greater digital literacy and critical evaluation (Amazeen & Bucy, 2019). Although it is also known that political interest can intensify selective exposure and polarization (Barnidge et al., 2020; Spohr, 2017), our results do not allow us to verify this tendency.

Finally, it is important to highlight that our findings are in conformity with other studies in the literature when they suggest that more conservative students tend to present slightly less favorable attitudes (Jia & Lee, 2024; Robertson et al., 2020; Amazeen et al., 2019). However, contrary to other studies (see Liu et al., 2025; Pretus et al., 2023; Shin & Thorson, 2017), we found no evidence that the ideological and political orientation of students may be associated with the level of skepticism.

7. Conclusions

The conclusions of this study allow us to respond directly to the defined research questions. Regarding RQ1 (“What are the levels of familiarity, contact, and attitudes toward fact-checking among Portuguese university students?”), the results suggest that fact-checking is widely recognized as a relevant and socially useful practice, but this normative legitimation does not translate, with the same intensity, into high familiarity or frequent contact. As for RQ2 (“To what extent are news consumption habits and political interest associated with greater familiarity and contact with fact-checking?”) and RQ3 (“To what extent are familiarity and contact related to more favorable attitudes, including trust, perceptions of impartiality, and perceived effectiveness?”), the observed patterns are consistent with the literature that points to the importance of informational routines and motivational predispositions in the differentiation between those who know and seek verifications and those who remain in a more peripheral relationship with this content (Lyons et al., 2020; Robertson et al., 2020). Finally, regarding the exploratory question RQ4 (“What role do skepticism and political-institutional variables play in shaping these associations?”), the study reinforces the pertinence of considering dimensions of skepticism and political-institutional framing as part of the problem of public reception of fact-checking, in an ecosystem where verification is itself an object of authority dispute and suspicions of bias (Tandoc et al., 2018; Waisbord, 2018).

One of the clearest contributions of this work lies in the empirical gap between recognition/valuation and effective practice. Although participants demonstrate globally favorable attitudes toward fact-checking, levels of regular consumption and, above all, of sharing and active engagement remain limited. This pattern approximates what has been described as high support, low use, suggesting that the main challenge is not only normative, but also motivational and routine (Robertson et al., 2020; Baptista et al., 2023). In light of the literature on incidental consumption and news avoidance, it is plausible that platform-curated exposure, informational fatigue, and the low disposition for public interaction with news content constitute structural obstacles to the mobilization of fact-checking as an everyday practice (Thorson & Wells, 2016; van Dijck et al., 2018; Boczkowski et al., 2018; Fletcher & Nielsen, 2019; Skovsgaard & Andersen, 2020). Similarly, it is plausible that mechanisms related to the illusion of informational competence, whereby subjective confidence in one’s critical ability may diminish the perceived need for external verification, could contribute to this gap, although this hypothesis was not empirically tested in the present study and remains a direction for future research (Rader & Gray, 2015; Kahneman, 2011; Nyhan & Reifler, 2010; Swart, 2021).

In this way, a future research agenda must deepen the causes of low active engagement: why do students who recognize the utility of fact-checking rarely share it or use it regularly, and what conditions could increase this mobilization? Subsequent studies can explore, through qualitative and/or experimental means, what factors hinder sharing (for example, reputational costs, conflict avoidance, low salience of corrective content in the feed, or insufficient integration into routines) and what mechanisms could incentivize it (for example, formats more compatible with incidental consumption, greater visibility in platform environments, or strategies of trust and transparency that reduce suspicions of bias) (Graves, 2016; Amazeen, 2020; Tandoc et al., 2018; Wood & Porter, 2019). In synthesis, the results presented here suggest that the social efficacy of fact-checking depends less on its abstract legitimation and more on its capacity to become practicable and mobilizable in real informational routines, especially in young and highly connected audiences, but not for that reason automatically more interventive in the digital public sphere (Fletcher & Nielsen, 2019; Swart, 2021).

8. Limitations and Future Research

We believe that our study should be interpreted in light of some limitations. First, it is important to emphasize that all results and conclusions are based on a convenience and non-probabilistic sample that was collected in a classroom context. In addition, the sample is also unequal in terms of gender and heavily concentrated in the Social Sciences and Humanities, which limits the generalization of the results and may have influenced the overall pattern of attitudes observed towards fact-checking. In the sample of our study, which is characterized as exploratory, there is an overrepresentation of students from disciplinary areas more directly exposed to social media or more sensitive to political content, which, in some way, may have contributed to comparatively more favorable assessments of fact-checking.

Secondly, the questionnaire measured general attitudes, familiarity, and self-reported engagement with fact-checking, rather than responses to concrete fact-checks with visual, social, or political stimuli. For this reason, the results should not be interpreted as invariant with respect to the topic. We are aware that students’ willingness to read, trust, or share fact-checks may vary depending on the relevance of the subject matter, personal involvement, or partisan or political relevance.

Third, the study relies on self-reported measures collected at a single point in time. This limits causal interpretation and leaves open the possibility of discrepancies between stated attitudes and actual behavior.

In the future, we believe it will be important to apply experimental research, in which it is possible to combine multiple tests with exposure to real fact-checks covering diverse topics. In addition, the study could be complemented with a qualitative analysis of how students’ political involvement and perception of personal competence shape actual engagement with fact-checking content.