Abstract
In this presentation I address the challenge of conducting ”real time” research in online social networks. More specifically, I engage the questions of how to study the development of events on social network sites as they are occurring, as well as how to capture the ever-changing nature of the social behavior in these networks (who says what to whom and when)?
Drawing on the idea of liveliness – how is a topic (or topics) changing over time (Marres and Weltevrede 2012) - I investigate different approaches (among others scraping, keyword analysis, log-data analysis, systematic screen dumping etc.) that are useful to study the variation in the type and volume of activity in online social networks. These different approaches encompass various conceptions of what data is and how to capture this data. This opens up for an interesting discussion of the epistemological aspects of these data aggregation tools.
On top of that, there are a multitude of methodological challenges related to studying social media in real-time. Firstly, there is the issue of what data to capture, at which points in time. Secondly, there is the pertinent challenge of access - both understood in terms of ”walled gardens”, but also in terms of, how to access and retrieve data that is lost seconds after appearing online (e.g. the specific graphical and relational constellation of objects in social network sites). The latter relates to the general issue of how to recreate the context of a given social interaction in online as well as offline research. Lastly, there is a need to touch upon issues of representativeness, generalizability and validity of the available data.
Throughout the presentation I will relate the methodological challenges to a more specific discussion of ”media events” (Dayan and Katz 1992; Katz and Liebes 2007) in the era of social and networked media. In this context, I introduce two cases that have attracted substantial attention on online social networks (as well as in the mainstream media): “Hurricane Sandy” (the storm that rampaged the US east coast during a few days in October 2012) and “Red Bull Stratos” (the staged jump from the stratosphere, where an astronaut broke the speed of sound in a free fall). These cases have engaged users on a variety of online platforms (among others Twitter, Facebook, Instagram), and are thereby suitable for an inquiry into the methods used to study online social networks in real-time.
Drawing on the idea of liveliness – how is a topic (or topics) changing over time (Marres and Weltevrede 2012) - I investigate different approaches (among others scraping, keyword analysis, log-data analysis, systematic screen dumping etc.) that are useful to study the variation in the type and volume of activity in online social networks. These different approaches encompass various conceptions of what data is and how to capture this data. This opens up for an interesting discussion of the epistemological aspects of these data aggregation tools.
On top of that, there are a multitude of methodological challenges related to studying social media in real-time. Firstly, there is the issue of what data to capture, at which points in time. Secondly, there is the pertinent challenge of access - both understood in terms of ”walled gardens”, but also in terms of, how to access and retrieve data that is lost seconds after appearing online (e.g. the specific graphical and relational constellation of objects in social network sites). The latter relates to the general issue of how to recreate the context of a given social interaction in online as well as offline research. Lastly, there is a need to touch upon issues of representativeness, generalizability and validity of the available data.
Throughout the presentation I will relate the methodological challenges to a more specific discussion of ”media events” (Dayan and Katz 1992; Katz and Liebes 2007) in the era of social and networked media. In this context, I introduce two cases that have attracted substantial attention on online social networks (as well as in the mainstream media): “Hurricane Sandy” (the storm that rampaged the US east coast during a few days in October 2012) and “Red Bull Stratos” (the staged jump from the stratosphere, where an astronaut broke the speed of sound in a free fall). These cases have engaged users on a variety of online platforms (among others Twitter, Facebook, Instagram), and are thereby suitable for an inquiry into the methods used to study online social networks in real-time.
| Originalsprog | Dansk |
|---|---|
| Publikationsdato | 19 jun. 2013 |
| Status | Udgivet - 19 jun. 2013 |
| Begivenhed | The ICA 2013 conference “Challenging Communication Research” - London, London, Storbritannien Varighed: 14 jun. 2013 → 21 jun. 2013 |
Konference
| Konference | The ICA 2013 conference “Challenging Communication Research” |
|---|---|
| Lokation | London |
| Land/Område | Storbritannien |
| By | London |
| Periode | 14/06/2013 → 21/06/2013 |