Integrating trace data into interviews: Better interviews, better data.

In this paper, we present a transnational case study on binge-watching and media habits. The contributions of this paper are two-fold. The first contribution is methodological, proposing 'data-prompted interviewing', a variant of trace interviewing. Participants' own trace data is used to help them recall their own behavior in more detail, which in turn enables them to identify gaps or flaws in trace data records. This case study, which includes sixty interviews in platform logs from YouTube, Netflix and a large self-tracking service, provides evidence that 'data-prompted interviewing' yields robust results. Using data to prompt interviewees' memories substantially improved their recall, and interviewees' insights into platform data substantially improved its interpretation. The second contribution of this case study is theoretical, engaging with calls for reflexivity in media studies by empirically embracing Jenssen's idea of dual 'data movements' and operationalizing reflexivity into concrete methods. Looking at the platform logs with the interviewee revealed patterns and habits in watching behavior which an interview alone would have entirely missed. But interviews also revealed quirks in the platform data which would have completely misled a researcher working with platform logs alone. The combination produces more insightful and robust research results for both 'big data' and 'small data' approaches. Recent data protection laws, which give millions of people the right to access their own trace data from platforms, make this approach practical for a wide range of questions in media studies. We propose that media studies scholars embrace reflexivity as not just theory but method, and through data-prompted interviews, fully embrace how the field affects research, and how research affects the field. [ABSTRACT FROM AUTHOR]