Data not seen: The uses and shortcomings of social media metrics

Abstract

This paper looks at the use of online data, especially social media metrics, to assess media audiences, with particular focus on musicians. It shows how audiences are defined by different sets of people, and grounds the use of social media to understand audiences in the history of mass media audience measurement. The second half of the paper focuses on visible social media metrics-likes, followers, and other such counts-outlining their appeal as measures and highlighting their fallibility and ambiguity. Throughout, the paper argues that different people construct information systems to collect, store, analyze, and interpret data, and that these are shaped by value systems. Metric and big data analysis generally serves economic values, while other approaches to data may be more appropriate for assessing social and personal values.

De platformsamenleving. Strijd om publieke waarden in een online wereld

“…De data van deze gebruikers wordt wel een cruciale bron van inkomsten. Bezien vanuit een techno-commercieel perspectief draait commodificatie om het te gelde maken van gebruikersdata (Baym 2013). Wie verdient er aan data?…”

Balancing the potential and problems of digital methods through action research: methodological reflections

“…boyd and Crawford's article, cited above, is one example. and various contributions to a special issue of First Monday on big data (such as Baym, 2013 andBoellstorff, 2013) provide further critique, highlighting the making and shaping of digital data through methods (see also Barnes, 2013 andGitelman andJackson, 2013).…”

“…Small-scale, qualitative studies open up a space for reflecting on the affordances of digital methods, we argue. This point is made by Baym (2013), when she states that 'Now, more than ever, we need qualitative sensibilities and methods to help us see what numbers cannot'. We conclude that such sensibilities need to extend beyond academic, social scientific and humanities domains.…”

Motivations for participating in ‘viral politics’

“…However, these results may be attributable in part to the fact that the study centers on a distinctly non-partisan hashtag (#elec-tion2012), whereas other threads of interaction on Twitter may have potentially been more partisan (and perhaps rhetorical) in nature. Moreover, while such 'big data' metric analysis can gauge the nature of the content that circulates on Twitter, smaller scale qualitative methods are better equipped to assess the personal meanings that users attach to their digital communication practices (Baym, 2013). The present study thus builds upon the small, but growing, body of qualitative empirical work on viral politics (e.g.…”