Online media have become a prime source of news use in most Western countries (Reuters Digital News Report, 2017). As the audience of legacy offline media is steadily declining, we have entered into an era in which online news no longer complements but replaces offline news use. Keeping in mind differences in the affordances of online and offline news use, this raises many questions with regard to the habitualization of news use. For example, TV news exposure can be best explained by context and habit (Wonneberger et al., 2012), whereas online news users assume a much more active role online (Mitchell et al., 2017). At the same time, social media platforms have become an important source of incidental news exposure (Karnowski et al., 2017). This mode of news use is significant, because extant research shows that news that is encountered while browsing social media for entertainment or social purposes, leads to engaged news use, especially among users with lower political interest (Valeriani & Vaccari, 2016).
We hence need to gain a better understanding of how online news use is engrained into internet use in combination with contextual information that only a combination of survey data with trace data can provide. This study will map the pathways to online news sites to identify triggers and patterns of news use. To do so we will use click stream tracking data, collected through a browser plug-in, in combination with survey data from a sample of 269 Dutch internet users.
We are able to identify patterns of news use by studying the news-related sequences of browsing sessions; thereby identifying different pathways to news such as via social media, via search engines or directly via a news website and relate that a) to the type of user that engages in either pattern, and b) the likelihood that the session is followed up by information seeking or participatory behavior. Subsequently, we will use survey data to explain typical user behavior using multi-level regression analysis and combine both sources of data to build usage typologies. Covariates of interest are a) motivational variables such as political interest and internal efficacy b) complementary news use such as TV news and newspaper use c) demographic background variables, most importantly age.
As we are still in the process of analyzing the data, it is too early to draw conclusions with regard to the implications of our findings. However, we do think that our paper will be an important contribution to the growing body of survey based literature on online news use. Our analysis already indicates that participants severely overestimate their (incidental) exposure to online news. While limited somewhat by the absence of mobile data, our analyses indicate how we should move forward to really understand patterns of usage of online news.