We investigate the often-stated, but disputed claim in the political science and political communication literature that increasing media choice widens inequalities in political knowledge, & find little support

Increased Media Choice and Political Knowledge Gaps: A Comparative Longitudinal Study of 18 Established Democracies 1995-2015. Atle Haugsgjerd, Stine Hesstvedt, & Rune Karlsen. Political Communication, Jan 25 2021. https://doi.org/10.1080/10584609.2020.1868633

ABSTRACT: We investigate the often-stated, but disputed claim in the political science and political communication literature that increasing media choice widens inequalities in political knowledge. The assumption is that in a high-choice media environment, the politically interested will consume more news while the uninterested will avoid such content, leading, in turn, to widening differences in political knowledge. Although previous studies show that high media choice increases political knowledge gaps in the United States, comparative longitudinal evidence is currently lacking. To fill this gap, we draw on data from four rounds of the Comparative Study of Electoral Systems. Overall, we do not find general support for the high-choice knowledge gap thesis. In most countries, there is no indication that inequality in political knowledge has increased over time. Building on recent insights from political communication research, we question key assumptions of the high choice knowledge gap thesis.

KEYWORDS: Political knowledgehigh choiceknowledge gapsincreasing political knowledge inequalitynews avoidance

Discussion and Conclusion

Overall, we do not find support for the high-choice knowledge gap thesis as a general theory. Contrary to the high-choice thesis, our analyses provided little evidence that inequality in political knowledge increases over time, and increased Internet use and broadband access had no effect on knowledge inequality. Further, longitudinal analysis of Norwegian data showed that knowledge was not increasingly stratified by political interest and education from 1997 to 2017. The comparative multilevel analysis also indicated that education has not become a stronger predictor of political knowledge over time or in contexts with high levels of Internet use or broadband access. Overall, the results thus suggest that although increased media choice facilitates increasing personalization of media consumption, this does not necessarily mean that the information-poor escape the constant flow of political news coverage and that knowledge inequalities in high-choice societies increase accordingly.

In the theoretical section, we noted newer strands of research that help to clarify these results. In particular, we highlighted the “infrastructural” view of media use, suggesting that individual preferences are less pivotal than assumed in arguing for the high-choice knowledge gap (e.g., Webster, 2014, pp. 23–48). Preferences are constrained by channel repertoires (Taneja et al., 2012) and situational factors (Wonneberger et al., 2011), as well as by the architecture of the digital political communication system (Taneja et al., 2018). On this view, traditional and digital media infrastructures limit the extent to which people (willingly or unwillingly) avoid news about politics and current affairs. Indeed, there is empirical evidence that the inadvertent audience has not disappeared (e.g., Fletcher & Nielsen, 2018; Thorson, 2020), and the longitudinal evidence for increasing news avoidance is inconclusive (Karlsen et al., 2020; Strömbäck et al., 2013). In short, more choice does not necessarily lead to more news avoidance and increasing knowledge gaps.

In this article, we have focused mainly on the development of knowledge about party positions. As it reflects voters’ basic understanding of the political system, this form of knowledge is essential to the ability to navigate the political landscape, and it is reassuring to find that inequalities do not increase as media platforms multiply. Our analysis of factual knowledge questions – which capture citizens’ dynamic knowledge about current issues covered by the media – yielded largely similar results. However, we would like to emphasize that we do not consider these results conclusive in terms of how increasing media choice affects inequalities in political knowledge. Future research should explore inequalities in different types of political knowledge, including policy knowledge as well as more general political information (cf. Barabas et al., 2014). The lack of variables on media use is also an important shortcoming in the present study. The opportunity to link media consumption and political knowledge would have offered a more comprehensive picture of the impact of media preferences in societies transitioning to high choice. Unfortunately, the lack of relevant data makes this type of longitudinal study difficult to conduct. Panel-studies that include content preferences and media use should, however, provide valuable insights. Richer and more detailed measures of media system fragmentation would also be highly valuable. Nevertheless, we believe the present study provides a good point of departure for such future work.

Another important task for future research is to dig deeper into how specific groups react to the changing media environment. In their seminal work, Tichenor et al. (1970, pp. 159–60) argued that when mass media information increases, “segments of the population with higher socioeconomic status tend to acquire this information at a faster rate than the lower status segments, so that the gap in knowledge between these segments tends to increase rather than decrease”. In contrast to the high-choice thesis, they contended that when the relative difference between the information-rich and the information-poor increases, the latter do not necessarily become less knowledgeable (Tichenor et al., 1970, p. 160). In line with this proposition, we found that political knowledge in most countries is more or less stable for the lowest quintile of citizens. 22 One possible reason for this pattern is that a preference for one genre rather than another (e.g., entertainment over news) can lead to increased consumption of the preferred genre without reducing consumption of the other (e.g., Webster, 2014). Time spent on media consumption is not necessarily a zero-sum game, and different types of media consumption might increase simultaneously (Newell et al., 2008). A reformulation of the high-choice knowledge gap thesis along these lines may provide a better understanding of contemporary political knowledge dynamics. It could also serve to guide future research on how people with little political interest and resources relate to and learn from the different types of news made available by current digital media systems.

Although our findings do not support the high-choice thesis in general, we did find some evidence of increasing inequality in the United States. This result related mainly to information-poor citizens becoming less knowledgeable. On the one hand, the finding aligns well with Prior’s seminal US studies (2005, 2007), as well as with previous studies of media systems and political knowledge that report striking differences between the United States and all other countries (e.g., Aalberg & Curran, 2012). On the other hand, these results must be treated with caution. Most importantly, our main dependent variable relies on citizens’ placement of parties on the left-right scale, which is a less familiar concept in the United States compared to other countries in our sample. Our analysis of factual knowledge questions did not identify a similar increase in inequalities. One interpretation of these different results relates to affective polarization processes in the US. Increasing media fragmentation in the US is intertwined with political polarization of the media, and since the turn of this century, negative sentiments toward opposing parties and their voters have grown considerably (Hetherington & Rudolph. 2018). Hence, due to partisan media, a strong focus on misinformation and fake news (e.g., Allcott & Gentzkow, 2017; Lazer et al., 2018) and affective polarization in the electorate (Iyengar et al., 2019), voters might perceive parties as increasingly ideologically extreme, and therefore increasingly misplace these parties on the left right scale.

Overall, our results suggest that a high-choice media environment does not necessarily lead to a widening political knowledge divide between the information-poor and the information-rich. Although encouraging from a democratic perspective, this finding should nevertheless prompt us to think harder about how greater media choice influences media use, and in turn, political knowledge inequality. As the political communication systems of established democracies undergo rapid change, it becomes crucial to improve our understanding of these matters, both empirically and theoretically. This article identifies a number of avenues for future research in this regard. We should also keep in mind that for much of the study period, the choice was between television and radio channels, traditional newspapers, online text pages and low-quality online videos. The present media systems dominated by 24/7 on-demand and high-quality video content offer quite different choices. In this environment, news and current affairs must always compete with favorite television shows or movies. Perhaps the era of real choice has just begun?