News literacy, social media behaviors, and skepticism toward information on social media

ABSTRACT

Amid growing concerns about misinformation on social media, scholars, educators, and commentators see news literacy as a means to improve critical media consumption. We use a nationally-representative sample to investigate the relationship between news literacy (NL), seeing and posting news and political content on social media, and skepticism toward information shared on social media. This study finds NL and related orientations contribute to who is seeing and sharing information on social media, with those who are more knowledgeable about media structures seeing and sharing less content. Moreover, those who are more news literate and value NL are more skeptical of information quality on social media. Seeing and posting news and political content on social media are not associated with skepticism. This study suggests that NL plays an important role in shaping perceptions of information shared online.

KEYWORDS:

• Social media
• news literacy
• political communication
• misinformation

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributor

Emily K. Vraga (Ph.D., University of Wisconsin-Madison) is an associate professor in the Hubbard School of Journalism and Mass Communication at the University of Minnesota, where she holds the Don R. and Carole J. Larson Professorship in Journalism. Her research focuses on how individuals process news and information about contentious health, scientific, and political issues, particularly in response to disagreeable messages they encounter in digital media environments. She is especially interested in testing methods to correct misinformation, to limit biased processing, and to encourage attention to more diverse content online. [email: [email protected] or Twitter @ekvraga].

Melissa Tully (Ph.D., University of Wisconsin-Madison) is an associate professor in the School of Journalism and Mass Communication at the University of Iowa. Her research interests include news literacy, civic and political participation, and global media studies. She is particularly interested in the relationship between individuals' news literacy and their news and civic engagement as well as the potential for news literacy efforts to improve information processing. [email: [email protected] or Twitter @tullyme].

Notes

1 The Theory of Planned Behavior examines Perceived Behavioral Control, rather than self-efficacy, but Ajzen reflects on the similarities of these two constructs and acknowledges Bandura’s definition of self-efficacy as inspiration for his own model (Ajzen, 1985).

2 As the survey occurred as part of a larger data collection, participants were required to take the survey on a desktop computer. This requirement led to some skew in the stratification by group. The larger project did not influence the analyses reported in any way.

3 These numbers approximate the number of social media users reported by Pew for Facebook (79% of online adults) but over-represents Twitter users: Pew reports 24% of online adults use Twitter, compared to 37% of our sample (Greenwood, Perrin, & Duggan, 2016).

4 As these values are artificially inflated by the number of non-users who receive a value of ‘1,’ we also ran these correlations among only those who use a social media site. Correlations are still high, ranging from r = .68 to r = .94.

5 We ran all models separately for ‘political content’ and ‘news content.’ Results are not meaningfully different for these two types of content, which are included in the appendix.

6 We also test these models with the full six items validated by Vraga and Tully (2015) and find they do not substantially change the results presented here. We thank the anonymous reviewer for their suggestion to focus on a more limited set of SPML measures that better reflect the efficacy component we are attempting to measure.

7 Posting on Facebook and Twitter is correlated at r = .68, p < .001, between Facebook and YouTube is r = .69, p < .001, and between Twitter and YouTube is r = .67, p < .001.

8 We also examine these relationships looking only at individuals who use a given platform to ensure including those who do not use a platform among the ‘never’ group does not create skew in our results. The results are similar to those reported using the full models, although some of the relationships are reduced to non-significance with the reduced power. Please see the supplemental appendix.

9 SPML and NL are correlated at r = .27, p = .00, SPML and VML are correlated at r = .53, p = .00, and VML and NL are correlated at r = .48, p = .00.

10 Bivariate correlations reveal a small positive relationship between SPML and skepticism (r = .07, p = .05), which may be explained by the positive association between SPML and knowledge. The regression results suggest that believing one is media literate is not sufficient and may be detrimental for democratic processing.

11 We thank a reviewer for this suggestion.

Additional information

Funding

This work was supported by University of Iowa [Grant Number 000].