Understanding the production and circulation of social media data: toward methodological principles and praxis

Social media data have provoked a mixed response from researchers. While there is great enthusiasm for this new source of social data – Twitter data in particular – concerns are also expressed about their biases and unknown provenance and, consequently, their credibility for social research. This article seeks a middle path, arguing that we must develop better understanding of the construction and circulation of social media data to evaluate their appropriate uses and the claims that might be made from them. Building on sociotechnical approaches, we propose a high-level abstraction of the ‘pipeline’ through which social media data are constructed and circulated. In turn, we explore how this shapes the populations and samples that are present in social media data and the methods that generate data about them. We conclude with some broad principles for supporting methodologically informed social media research in the future.