Christopher Wlezien summarises the findings of his recent article on the relationship between news and public opinion. His analysis of three different cases of US public opinion over time finds that public attitudes may ‘cause’ news coverage more than they are caused by it contrary to the dominant assumptions.
Increasingly, we choose our media. There has been choice for some time, however – well before the advent of the internet and social media. This competition provides incentives for news organisations to gauge and respond to public sentiment. But there are other reasons to expect news to be responsive. First, reporters and editors are samples of the broader population and so may be representative. Second, even if they are not like the rest of us, they may see their role as representing what ‘the public’ cares about and thinks.
We thus might expect the news to follow the public for various reasons. This may help explain why news organisations have been involved in public opinion polling, particularly when the costs of doing so were much greater than they are today. They went out of their way at real expense to collect the information, presumably to help guide their reporting (and editorialising), both the topics they cover and how they cover them.
Yet, most research (and commentary) on the media and the public assumes that the latter follows the former. The assumption may seem understandable given the importance of media in our lives, but is the flow only in one direction from the media to the public? Or does the public also drive media coverage? Answers to these questions are important, as they matter for effective political accountability and representation; indeed, they shed light on whether the public really is – can be! – influential in politics and policymaking.
Disentangling the causal relationship(s) between media coverage and the public is difficult. In such circumstances, scholars often rely on ‘cross-lag models’, depicted in Figure 1, which is the approach I adopt in my article (you can find out more about the methodology here). It is a conservative approach, as results will tend to understate the influence of news on opinion and the influence of the public on the news, but it still allows some assessment of directional influence, where positive evidence is to be taken seriously.
An empirical analysis of three cases
For this analysis, I build on my previous research to analyse how news relates to public opinion in three cases in the United States: economic perceptions, candidate support, and policy preferences. These are three different but important cases and the regular focus of coverage and pollsters’ survey questions and also the subject of much scholarly attention.
Economic news tone and public perceptions
To measure the public’s economic perceptions, I use the University of Michigan’s Survey of Consumers in the US. I rely on both economic retrospections – assessments of the past year – and prospections – expectations for the next year. For the analysis, I use monthly averages beginning in January 1980, which is when the news data are available. For this, I rely on previous research measuring tone of ‘economic’ coverage from the New York Times and Washington Post. That research identifies ‘positive’ and ‘negative’ words in domestic economy articles to create a basic indicator of tone.
I found that monthly correlations are 0.42 for retrospections and 0.38 for prospections. Does this mean that the public follows the news or leads it? Statistical analysis following the cross-lag approach indicate that both are at work, but the public leads coverage more consistently than it follows. This can be seen in Figure 2, which simulates the effects of ‘shocks’ in the tone of coverage on perceptions and the reverse. The left-hand column shows the small effects of news content on retrospections and prospections, and the right-hand column shows larger effects of both economic retrospections and prospections on news. The differences are clear, particularly for retrospections, and note that results are strikingly similar in the UK.
Figure 2: Impulse response functions relating economic perceptions and news tone
Candidate news tone and electoral support
Analysis of news coverage and voter preferences during the 2016 US presidential election campaign reveals similar patterns. The time series of preferences is based on hundreds of national polls that register how respondents would vote ‘if the election were held today.’ To measure media coverage, I rely on a measure that captures the tone of news relating to ‘Clinton’ and ‘Trump’ in nine leading newspapers. Figure 3 displays simulated effects of the news and preferences on each other based on a cross-lag analysis. Much as we saw for economic perceptions, the public leads the news more than it follows (though there is a good amount of variation across the campaign).
Figure 3: Impulse response functions, poll share and news tone, final 200 days, 2016
Spending news coverage and public preferences
Finally, I assess the relationship between news about government spending and the public’s budgetary preferences in defense, welfare, and health (see also my article with Stuart Soroka). The measure of preferences is based on data from the General Social Survey (GSS), which regularly asks the US public about their support for ‘more’ or ‘less’ spending. The measure of media coverage differs from that for the first two cases, in that it is not based on tone but on the substance of coverage. I use content from 17 newspapers.
Figure 4: Impulse response functions, from news to preferences
Figure 5: Impulse response functions, from preferences to news
Figures 4 and 5 highlight the difference in news and opinion effects in the three domains. While Figure 4 shows no real influence of coverage on preferences, Figure 5 reveals positive effects of the public on news content in all three domains, ones that are significantly greater than 0 for health and welfare. The public once again appears to lead the content of the news of spending.
What should we take from this? Public opinion appears to ‘cause’ news coverage in each of the three cases considered here, and the reverse holds less frequently and always to a lesser degree. It may be that media influence on the public occurs more quickly and so is not as evident as that of the public on the media using cross-lag models. This highlights the limits of such analyses; they only take us so far. That the results are fairly consistent across the three different cases is noteworthy, and underscores the role that the public can play in news coverage, one that should always be entertained, not settled by assumption.
By Christopher Wlezien, Hogg Professor of Government, University of Texas.