Scraping the Social?

What makes scraping methodologically interesting for social and cultural research? This paper seeks to contribute to debates about digital social research by exploring how a ‘medium-specific’ technique for online data capture may be rendered analytically productive for social research. As a device that is currently being imported into social research, scraping has the capacity to re-structure social research, and this in at least two ways. Firstly, as a technique that is not native to social research, scraping risks to introduce ‘alien’ methodological assumptions into social research (such as an pre-occupation with freshness). Secondly, to scrape is to risk importing into our inquiry categories that are prevalent in the social practices enabled by the media: scraping makes available already formatted data for social research. Scraped data, and online social data more generally, tend to come with ‘external’ analytics already built-in. This circumstance is often approached as a ‘problem’ with online data capture, but we propose it may be turned into virtue, insofar as data formats that have currency in the areas under scrutiny may serve as a source of social data themselves. Scraping, we propose, makes it possible to render traffic between the object and process of social research analytically productive. It enables a form of ‘real-time’ social research, in which the formats and life cycles of online data may lend structure to the analytic objects and findings of social research. By way of a conclusion, we demonstrate this point in an exercise of online issue profiling, and more particularly, by relying on Twitter to profile the issue of ‘austerity’. Here we distinguish between two forms of real-time research, those dedicated to monitoring live content (which terms are current?) and those concerned with analysing the liveliness of issues (which topics are happening?).