ABSTRACT
The only long term trend data on trust in the American press comes from the General Social Survey (GSS). The erosion of trust in the press as measured by the GSS indicator is indisputable, but its implications for the functioning of American democracy depend on what, precisely, is being measured. In this study we use an experimental design embedded in a representative national probability sample to shed light on what people are thinking of when they say they trust or distrust the American press. Are they thinking about the sources they themselves use for news? The sources that are most popular with the population at large? An average of all possible media sources? We find that individuals express much greater trust in the press when they are asked to consider specific news sources than when they are asked to evaluate a generic news media. Our results suggest that an accessibility bias combined with the proliferation of news sources in recent years may lead individuals to think of distrusted sources when asked to answer generic media trust questions. We therefore argue that different measurement strategies are needed to successfully address trust in the press in the current news environment.
Acknowledgments
The authors thank their classmates in the Political Communication seminar for their contributions to this study. Funding for the data collection was provided by the Institute for the Study of Citizens and Politics: www.iscap.upenn.edu
Notes
1 The systematic bias toward attention to and memory of negative information is well documented. Further, strongly negative events generate more emotional reactions and greater arousal, thus they tend to be more accessible in memory than positive examples.
2 This is consistent with other recent studies (e.g., Dilliplane, Citation2011, Citation2014; Gentzkow & Shapiro, Citation2011; Webster & Ksiazek, Citation2012), which show that partisan news consumption is a modest portion of most people’s media diets; the public perception is, nonetheless, that it is extremely popular.