News credibility is one of the most researched topics in journalism. Generally, people evaluate the quality of news based on credibility, which can lead to subsequent use of the news (Kiousis, 2001). Credibility was found to be related to public deliberation and civic engagement, which are the necessary ingredients for a more communicated, informed, and engaged citizenry (Nah & Chung, 2020).
Despite a plethora of literature on news credibility accumulated for a considerable period of time, the topic has gained even more intense research attention over the past two decades or so. One reason for the ever-increasing attention could be the decreasing news credibility around the world. The annual Edelman Trust Barometer Global Report (Edelman Trust Institute, 2023) and the Digital News Report (Reuters Institute, 2023) have shown a decrease in news credibility in many democratic countries for a number of years. For Korea, the Edelman Trust Barometer Global Report 2023 (Edelman Trust Institute, 2023) reported the lowest trust in media with 27 points out of 100 and the biggest drop in the trust level from the previous year with a 6-point decrease among the 27 countries studied. Another reason may be the fast-changing digital news media environment. The environment brought an abundance of news and news media, and people are now flooded by news and information in everyday life. It has become the news users’ task to choose news out of the inundation of information. On the other hand, news has entered the market with tighter-than-ever competition vying for audience attention just to survive.
The motivation of this study was the decline in news credibility in Korea. It might be intuitively imagined that people evaluate overall news based on their everyday news use experience. As such, it could be asked what Koreans’ everyday news experience is like that overall news credibility is so low. Alternatively, it could be asked if Koreans do not trust news in general as much as they trust the news they use every day. In reaction to this curiosity, this research aimed to explore what people’s everyday news experience is like and how such individual experiences might translate to news credibility in general on the major media of television and portals. More specifically, the purpose of this study is to investigate the following: first, how people might choose and/or encounter news in everyday life, because how individuals choose news can affect their credibility evaluation of the news (Metzger et al., 2020); second, how they evaluate the news they use in comparison to news in general on television and portals; and third, how their evaluation of the news they use is related to their evaluation of news in general on television and portals.
Examining the credibility of news people use, or “credibility of news I use,” is meaningful because, in the current news environment where news users are given higher control of what news they expose themselves to, the choice of news is more pronounced and, in effect, each individual’s everyday news experience and the perception of it might be different from others’ experiences and perceptions. While there are views that mass news media are generally homogeneous in content (McNair, 2009), and audiences, too, are notably homogeneous in their tastes (Neuman, 1991), studies in partisan news exposure have provided evidence for individuals’ different news preferences and heterogeneous attitudes toward the same news (e.g., Metzger et al., 2020; Stroud, 2010; Stroud & Lee, 2013). Further, how individuals’ evaluation of their everyday news experience compares to and relates to credibility of overall news on media may offer fresh explanations about news credibility. This study pays attention to “news I use” and examines how its credibility compares to and relates to credibility of news on television and portal websites. In doing so, it explores the possibility that people with stronger political ideology judge news in general and the news they use differently.
News credibility is most often understood as an attribute that shapes one’s perception of a message (Duncan, 2022). It is explained as the result of an evaluation as perceived by users and a product of a cognitive process through which a message is screened and chosen (Liu, 2004). Credibility is often evaluated on the basis of a combination of expertise, trustworthiness, completeness, accuracy, and bias (Flanagin & Metzger, 2000). Whether it be source, medium, or message that is the object of the analysis, many studies have employed one or more of these attributes to measure credibility. The measures have been used for various traditional media: newspapers, network television, and cable television (e.g. Gantz, 1981; Ibelema & Powell, 2001), as well as online news media (e.g. C. J. Chung et al., 2012; D. S. Chung et al., 2022; Schweigner, 2000). Some online news credibility research integrated online environment- and technology-specific characteristics into analyses, some of which included interactivity, multimediality, and hypertextuality (C. J. Chung et al., 2012).
Studies have also measured credibility with one single attribute directly capturing the extent to which one perceives news or news media believable, trustworthy, or credible (e.g. Ibelema & Powell, 2001; Major & Atwood, 1997; Nelson & Kim, 2020; Schweigner, 2000). Because credibility is a perception by nature, and the commonly employed items such as expertise and trustworthiness are essentially perceptions as well, the single direct measure of perceived credibility has served research purposes well and contributed to an increasing understanding of credibility.
Across many studies, a number of different factors have been found to predict credibility: demographics (e.g. Flanagin & Metzger, 2003; Johnson & Kaye, 1998, 2002, 2014), information-processing (e.g. Metzger et al., 2003, 2020), motivation (e.g. Lee & Ahn, 2017; Mulder, 1980), news media use (e.g. Greenberg, 1966; Westley & Severin, 1964), media dependency (e.g. Johnson & Kaye, 2000, 2002, 2009, 2014), soft or hard news (e.g. Ha & Lee, 2011; Miller & Kurpius, 2010), and news engagement such as commenting and liking (e.g. D. S. Chung et al., 2022). Drawing from the previous studies, this research attempts to extend the literature by exploring how the credibility people hold for the news they use is related to credibility of overall news on television and news portals respectively. It also examines how people’s political ideology strength might interact with the credibility of news they use and affect their credibility evaluations of television and portal news. This research chose television and news portals among major news outlets since they are the two most used media for news in South Korea (Korea Press Foundation, 2022).
Credibility of “News I Use”
“News I use” is a concept that has been researched for some time. Notably, the annual Digital News Report by Reuters Institute contains the item of “trust in news I use,” and it is often studied in comparison to “trust in news overall” to understand bias and selectivity. Examining how individuals evaluate the news they use in comparison to news in general bears significance because it addresses individuals’ differential credibility judgement based on their news selection and consumption practice. News people use is likely to be a combination of news on traditional media, such as television and newspapers, and online media such as news portals, online video platforms, and social networks, as most people obtain news via more than one news outlet (Korea Press Foundation, 2022). The set of news that is exposed to the user is likely to be chosen by the user based on various needs, motivations, and preferences. Thus, each individual is likely to have a different set of news from a combination of more than one news outlet in their everyday news diet. “News I use” contains not only the assumption that the news is used by the user but also that the user has one or more routes for the selection of and attention to the news that leads to the consumption of it. One of the routes that may influence people’s news selection could be selective exposure.
The selective exposure thesis claims that people tend to expose themselves to information that corresponds to their own views and perspectives and avoid information that does not match their beliefs. When encountering attitude-challenging information, people experience cognitive dissonance, a psychological discomfort and mental burden (Festinger, 1957), which puts people under the pressure to reduce it. In order to reduce cognitive dissonance, individuals seek out attitude-consistent information, while avoiding the counter-attitudinal. Indeed, there is a plethora of research results confirming partisanship selective exposure in people’s news selection and consumption (e.g., Garrett, 2009; Knobloch-Westerwick & Meng, 2009; Stroud, 2008, 2010). Knobloch-Westerwick and Meng (2009) found that news users were more likely to read the news if the headline was in line with their existing views and values. Stroud (2008) showed that individuals depended on at least one attitude-consistent news source in their news consumption. Literature on partisan selective exposure has demonstrated that people generally expose themselves to attitude-confirming news.
In terms of credibility, there is the strong evidence that people tend to evaluate attitude-consistent information and news sources higher in credibility than attitude-challenging ones (Kahan et al., 2009; Meyer et al., 2010; Oyedeji, 2010). This differential credibility perception can be explained by the cultural cognition view of human cognition and opinion formation in psychology. It suggests that people process information through the filters of their personal and cultural identities with due considerations of any relations and effect to the values and beliefs corresponding to their identities, and finally form opinions about the information (Kahan et al., 2009). On this basis, Kahan et al. (2009) concluded that people perceived like-minded sources to be fairer and more honest than differently-minded sources regardless of the message itself. In sum, the discussion of selective exposure and credibility so far suggests that people’s news choices based on selective exposure is likely to lead to higher credibility of “news I use.”
Another possible route to news is passive selection. Since the classic strong-media-effects-and-passive-audience models in the early times of media studies, when television had just made its dominance as a mass medium, there have been various other views of media effects. The limited media effects models of cognitive dissonance and selective exposure focused on viewer selection of attitude-consistent contents and avoidance of attitude-challenging information to media use (e.g., Sears & Freedman, 1967). The uses and gratification approach attended to viewers’ active selections to match their affective state (e.g., Zillmann et al., 1980). Contrastingly, the market research perspective focused on the structural aspects of television as a commercial medium and claimed that program choice was determined by scheduling rather than viewers’ active choice of contents (e.g., Owen et al., 1974). In more recent literature, while active audience choices were largely found, passive viewing styles were also evidenced (e.g., Cooper & Tang, 2009; Hawkins et al., 1991).
On the Internet, despite the increased audience control, there is the view that news users can generally bear the user-empowering interactivity only to the extent that requires minimal effort and easy use (Harrison, 2006). Indeed, obtaining news is still generally a passive behavior (Harrison, 2006). As such, while some media exposure occurs via a deliberate selection, others occur through choices dictated by certain conditions such as time, place, demographics, program scheduling, and algorithms.
Television and portals are two of the most accessible and easiest choices for people who view news habitually, just want to pass some time, or have nothing in particular to do. In particular, smartphone users carry the device most of the time they are awake. Sometimes they may actively use their smartphone to seek out information to achieve specific purposes. Other times, they may use it passively, for instance, scan through random news habitually on the metro train or in a long line waiting their turn, filling the empty time between the times they spend with specific purposes. Research shows, among various motivations for news use, the information-seeking motivation is closely related to higher credibility (Lee & Ahn, 2017; Mulder, 1980). Based on this claim, people’s credibility perception of news might not be particularly high if the news was passively chosen, rather than deliberately sought after.
Another way through which news users obtain news is inadvertent exposure. Television, as a traditional news medium, does not offer the pick-and-choose, one-news-item-by-one-news item feature as online news does. A number of different news items and stories are delivered in the package of a news program and presented to the audience via meticulous programming. This way, people can inadvertently encounter news they did not specifically choose to consume. In the online environment, also, news exposure is not always selective. Brundidge (2010) and Jun (2012) have found that inadvertent exposure does occur in online space. Both traditional and online media studies have demonstrated that news exposure happens incidentally more often than deliberately (Boczkowski et al., 2018; Fletcher & Nielsen, 2018). With a lower information-seeking motivation, accidental and inadvertent exposure to news might not be related to credibility. It is also quite possible that people are exposed to sensational news simply because the headlines are more visible. Sensationalism in news coverage has generally been found to be evaluated as unreliable (Haw, 2020).
In sum, based on the discussion above, “news I use” is likely to contain news obtained from a combination of different news media via varying routes, which include selective exposure, passive choice, and inadvertent encounters. News encountered through selective exposure tends to carry higher credibility, while news exposed via passive choice and inadvertent encounters might not be regarded as highly credible. According to Reuters Digital News Report 2023 (Reuters Institute, 2023), 38% of Koreans reported that they trusted “news I use,” while 28% indicated that they trusted in “news overall.” For many other countries covered in the report also, the percentage for “trust in news I use” was higher than the percentage for “trust in news overall.” While this result is helpful in understanding how people evaluate news they use, questions still remain about how people rate their everyday news in comparison to the news overall, specifically on television and portals, since those two media are the two most frequently used media for news in Korea (Korea Press Foundation, 2022). Therefore, the following research question is posed:
RQ1: How does credibility of “news I use” compare to credibility of overall news on television and portals respectively?
This research was also interested in understanding if trust people hold for news they actually use translates to news credibility of media in Korea in general. A number of studies have shown that frequent use of news and news media is positively related to credibility of the media (Johnson & Kaye, 2000, 2009; Kiousis, 2001). Further, frequent use of media positively affects not only credibility of the media used but also all news media in general (Ha & Lee, 2011). These findings suggest that people’s higher credibility assessment of the news and media frequently used can translate to higher credibility of overall news and media. It is indicative of the possibility of a positive relationship in credibility between “news I use” and news on television and news portals overall.
H1: Credibility of “news I use” is positively related to credibility of news on television overall.
H2: Credibility of “news I use” is positively related to credibility of news on portals overall.
Strength of Political Ideology and News Credibility
Based on the discussion in the previous section, it can be imagined that one would consume news via one or more of the three routes. Then a question arises: what combination of the three routes would one employ? For news users with stronger political ideology, it is likely that they employ partisan selective exposure more often than others (Garrett, 2009; Knobloch-Westerwick & Meng, 2009; Stroud, 2008; Westerwick et al., 2013).
For news consumers, overall news contains not only the news they use, but also news they do not use. Also, it includes not only attitude-consistent news, but also attitude-challenging news, as well as balanced news. Based on the finding that news consumers’ credibility evaluations for attitude-congruent news was significantly higher than the same for attitude-challenging news (Metzger et al., 2020), it can be assumed that those with stronger political ideology are likely to evaluate “news I use” more highly than news overall. There is also evidence from the hostile media hypothesis for people’s antagonistic bias to even balanced news coverage and sources, particularly when they are highly involved in the issue (Christen et al., 2002; Giner-Sorolla & Chaiken, 1994). News users with stronger political ideology could be more involved in political news and issues, and may view news more antagonistically than others, affecting the credibility of overall news negatively. In sum, individuals with stronger political ideology may employ partisan selective exposure in their news choice more than others, and while they are more likely to evaluate “news I use” to be credible, they may not necessarily judge news on television and portals overall to be equally credible. Therefore, for those with a stronger ideology, the extent to which the credibility of “news I use” corresponds to the credibility of overall news on television and news portals might be limited. Based on this discussion, I propose the following hypothesis:
H3: The relationship between the credibility of “news I use” and the credibility of news overall on television is moderated by strength of political ideology.
H4: The relationship between the credibility of “news I use” and the credibility of news overall on portals is moderated by strength of political ideology.
Data and Analysis
The data used in this study was from the annual News Media Audience Survey conducted by the Korea Press Foundation for the year 2021 and released in January 2022. It was conducted between May and July 2021 via face-to-face interviews with individuals 19 years of age or older. Based on the number of registered citizens in April 2021, the sample was allocated proportionally to each of the 17 metropolitan areas in the country and then stratified at five different administrative district levels in order to represent the population in each district in each level proportionally. The sample size was 5,010, which included 2,513 women and 2,497 men with the proportion of 49.8% and 50.2% respectively. Of the sample, 16.9% were 19 years of age or in their 20s, 16.1% in their 30s, 18.7% in their 40s, 19.5% in their 50s, and 28.7% were in their 60s or older. Forty-five-point-one percent of the sample received university education or higher, 39.5% had a high school diploma, and 15.4% achieved an education level lower than high school.
A paired sample t-test was used for RQ1, and hierarchical regression analyses were performed for television news and portal news respectively to test the relationships and the moderation effects assumed in Hypotheses 1, 2, 3, and 4.
Television News Credibility and Portal News Credibility
Television news credibility and portal news credibility were the two dependent variables in this research. Following previous research on credibility (Kiousis, 2001; Tsfati & Cohen, 2012), respondents were asked to indicate how credible they found news and current affairs information on television and portals (such as Naver and Daum) on a scale of 1 (not at all credible) to 5 (highly credible) (television news credibility: M = 3.8, SD = 0.7; portal news credibility M = 3.5, SD = 0.7)
Credibility of “News I Use”
Respondents were asked to show how credible they found news and current affairs information they use on a scale of 1 (not at all credible) to 5 (highly credible) (M = 3.5, SD = 0.6). As such, this measure is not media-specific.
Ideology strength was the interaction term for examining the moderation effect with news credibility. Respondents were asked to indicate their political ideology on a five-point scale (1 = strongly progressive, 2 = moderately progressive, 3 = neutral, 4 = moderately conservative, 5 = strongly conservative). The measure was then collapsed and recoded to present ideology strength with 1 indicating neutral, 2 moderate (moderately progressive and moderately conservative) and 3 strong (strongly progressive and strongly conservative) (M = 1.6, SD = 0.6).
Controls included the basic demographic variables of age (M = 48.5, SD = 15.9), gender (Male = 1, female = 2), and education (1 = elementary school, 2 = middle school, 3 = high school, 4 = college, and 5 = graduate school), political interest, television news use and portal news use. For political interest, respondents were asked to indicate how interested they were in politics and social issues on a five-point scale of 1 (not at all interested) to 5 (highly interested) (M = 3.3, SD = 0.8). For television and portal news use, respondents were asked to report how many weekdays they had used television news and portal news on a television set and mobile phone, respectively, in the past week on a six-point scale of 0 (none) to 5 (five) (television: M = 3.4, SD = 2.0; portal news: M = 3.2, SD = 2.0).
RQ1 asked if there was any difference in the level of credibility between “news I use” and news on television and portals. Results of a paired sample t-test in Table 1 indicated statistically significant and higher levels of credibility for television news (M = 3.7; SD = 0.7) (t (5009) = 27.09, p < .001) and portal news (M = 3.5; SD = 0.7) (t (5009) = 2.25, p = .020) than “news I use” (M = 3.5; SD = 0.6). The difference was greater and the significance was stronger for television than portals. It should also be noted that the difference was quite small for portal news.
The hierarchical regression models explained 24% of the variance in television news credibility and 29% in portal news credibility. Results in the second block of Table 2 showed a significant relationship to television news credibility (β = .42, p < .001). A significant relationship was also found for portal news credibility (β = .41, p < .001). Therefore, H1 and H2 were supported. Interestingly, among the demographic variables in the first block, age and education were related to news credibility of both of the media in opposite ways. Older age and lower education were significantly related to television news credibility (age: β = .08, p < .001; education: β = -.05, p = .019), while younger age and higher education explained portal news credibility (age: β = -.08, p < .001; education: β = .12, p < .001). Similarly, the direction of the relationships to news credibility were opposite for television news use and portal news use. Television news credibility was positively related with television news use but negatively with portal news use (television news use: β = .14, p < .001; portal news use: β = -.06, p < .001), while portal news credibility was positively related to portal news use but negatively to television news use (television news use: β = -.03, p = .058; portal news use: β = .21, p < .001).
The expected moderation effect was statistically significant for portal news as shown in the third block of Table 2 (β = -.28, p < .001), but not for television (β = -.14, p = .068). Therefore, H4 was supported, while H3 was not. In order to demonstrate the negative moderation effect in H4, Figure 1 and Table 3 were prepared. Figure 1 shows estimated marginal means for portal news credibility with the three separate curves of three different ideology strengths at different levels of credibility of “news I use.” All of the three curves, except for the strong curve, show increased portal news credibility as credibility of “news I use” increases. The strong curve moves to the upper right, except at the highest level of credibility of “news I use.” To examine these increases and decreases more closely, Table 3 was prepared. From Table 3, for the news user group with the lowest level of trust on the news they use not at all credible, the mean of portal news credibility was 2.0 for those with neutral ideology, 2.1 for the moderates, and 1.6 for those with strong ideology. For the group with mid-level trust in the news they use, the mean of portal news credibility was 3.3 for those with neutral ideology, 3.3 for the moderates, and 3.2 for those with strong ideology. For the group with high trust in the news they use, the means for portal news credibility was 4.3 for the neutrals, 4.1 for the moderates, and 3.4 for those with strong ideology. These results show that portal news credibility generally increases as credibility of “news I use” increases, which explains the overall upper-right directions of the curves in Figure 1. However, within each news user group of the five different levels of credibility of “news I use,” portal news credibility decreases at the highest ideology strength, which explains the negative moderation effect of ideology strength.
The credibility of “news I use” was found to be lower than television news credibility as well as portal news credibility. This finding is in contrast to Digital News Report 2023 (Reuters Institute, 2023), where “trust in news I use” is higher than “trust in news overall” for most countries, including Korea. This contrast is likely to be due to the specificity of television and portals as news media in this research. “Trust in news overall” in the Digital News Report did not specify news media. Literature has shown that traditional media such as newspapers and television are generally more trusted than online news media (Metzger et al., 2020; Nelson & Kim, 2020). The substantially higher credibility level of television news seems to have been driven partly from people’s attribution of higher credibility to television as a conventional medium. News users are likely to take a mix of different news selection routes in their news diet. The credibility of “news I use” would be much higher if people employed selective exposure alone. Combined together, all of the three news exposure routes, i.e., selective exposure, passive selection, and inadvertent exposure, could shape an individual’s news experience and are likely to influence the credibility of “news I use.”
The results regarding H1 and H2 revealed that the more individuals trusted the news that they used, the more likely that they trusted television and portal news. People’s evaluation of the news they use does translate to their evaluation of media news in general. This finding suggests that most people are safe from the danger of being in an echo chamber, where they hear only news and information corresponding to their own views. The moderation effect was found for portal news only. The moderation effect was not found for television, and the reason seems to be due to the generally high trust attributed to television news as a traditional form of media. Because television news tends to be trusted universally, it is less likely to reflect individual news use and credibility. While H4 was found to be statistically significant, the incremental R2 value for the third block was so small, i.e., 0.2%, that it is suggestive of reservations in the interpretation, particularly for those with stronger ideology.
The results of H4 reaffirmed H2 in showing a strong correspondence between the credibility of “news I use” and portal news credibility. It means that the more people trust the news they use, the more they trust portal news overall. However, the results also revealed how strong ideology could interfere negatively in the extent to which trust in the news individuals use corresponds or translates to portal news credibility overall. At each level of trust in the news people use, the extent of the “correspondence” decreased when their ideology strength was strong. It suggests that strong belief in the news one uses coupled with strong ideological beliefs could potentially lead to bias, i.e., believing news one uses, while not trusting other or overall news, which presents a case of selective exposure, where individuals’ political beliefs guide their news selections and evaluations (Stroud, 2008). For example, for the news users with strong ideology and high trust in the news they use, the mean for portal news credibility was 3.4, a value lower than the sample mean of 3.5.
Judging the news they use as highly credible, while assessing overall portal news as much less credible, politically strong-minded people might avoid news that they did not choose on portals. Further, even if they do expose themselves to unlike-minded news passively and inadvertently, they may interpret the news in the way they want to believe it in the subsequent process of selective perception, which may lead to misinterpretation of news and, in effect, politically polarized world views, political apathy, and intolerance, as well as a less-than-fully-informed citizenry.
The lowest portal news credibility of 1.6 was from the group with strong ideology and the lowest credibility of “news I use.” This group of news users seem to show negativity and cynicism toward news and media despite their strong ideological beliefs. However, as mentioned earlier, these results should be interpreted with reservations due to the small R2 value. Since the negative moderation effect was most visible among those news users with strong political ideology and/or the highest and lowest credibility of “news I use,” the small value might be due to the small proportion of each news user group in the sample (strong political ideology: 4.3%; highest and lowest credibility in “news I use:” 1.1% and 0.3% respectively). Interestingly, for the lowest
and low levels of credibility of “news I use,” portal news credibility was higher for the moderates than the neutrals.
Television news credibility was not subject to the negative moderation effect assumed. It appears that television, as a traditional media, is generally more trusted regardless of the level of trust in the news people use or ideology, thus less of subject to such individual audience characteristics.
This study addressed and revealed how the inner workings of individuals’ news choice, consumption, and credibility judgement could influence the credibility of television and portal news, and how ideology strength could manipulate the inner workings yielding differential credibility evaluations. However, it has limitations that can be improved in future research. First, there is a possibility of difference in use and evaluation of news and media between conservative and liberal people. There is research that shows difference in attitudes toward counter-attitudinal news between conservatives and liberals (Garrett & Stroud, 2014). Second, this study made theoretical assumptions about the three routes to news exposure. In future research, more evidence would be found if the routes can be methodologically operationalized to be observed actually and directly.
The lower level of credibility of “news I use” in comparison to television and portal news found in this study warrants further attention. The result indicates that people’s everyday news use does not meet the expectations they generally hold for news and media. While this result appears to leave a broad task to attend to for the field of journalism in terms of the quality of and access to news, its implications also include the need for renewed attention to media literacy for news users so people can make a more informed news choice based on valid expectations as part better communication in society and a more engaged citizenry.
All of the values of the coefficients and R2 in block 2 were same with or without ideology strength.