KES 4 Summary – quality in social media research

Missed the final Knowledge Exchange Seminar on quality in social media research? Not to fear, read on to feel like you were near!

Emerging methodologies always, and quite rightly, face scrutiny and sometimes scepticism in relation to their rigour. The challenges for social media research methods are similar to those faced by qualitative research two decades ago: demonstrating the complementary value of new methods to those to whom it may appear a threat. Our final Knowledge Exchange Seminar discussing quality took place in this context and the following provides a flavour of the key discussions.

Is it a new method or a new paradigm? Can we apply existing principles or do we need new standards? These were the overarching questions of the day. Epistemologically, the data are different; methodologically, the sample is different. We also know from KES3 that the ethical questions also require new thinking.

Yet while all of this is different, participants at our event were keen to stress that we can approach these issues using what we already know, drawing upon generic principles we apply to all social research. So, ensure we are question-led and don’t fetishize the method – just because social media can be easier and cheaper, only use it where it is appropriate. In fact, if it’s just being used because it’s easy, it’s probably not being done well.

And this brings me on the next important point to emerge. Social media research is a discipline in its own right and a heterogeneous one. It requires specific skills to collect and manipulate the data and real expertise in interpreting the data and, where possible, drawing wider inferences. And it is not all about Twitter or all about scraping. Researchers are collecting data from all sorts of social media sources and analysing volume, meta-data, content, connections and much more.

Out of this, however, there was a strong sense that there needs to be a shared understanding or guidelines about when it’s appropriate and when it not. We can use some general principles to understand this, but we also need empirical testing and experimentation to help this understanding evolve. When we make decisions on mode in a social survey or sample characteristics for a purposive qualitative sample, we draw on a whole host of methodological best practice built up, tested and refined over decades - this network provides a great opportunity for us to do that for social media.

Given nascent nature of this discipline then, we need not to focus on (though continue to acknowledge) potential weaknesses and limitations. People doing social media research are doing pioneering work. At our event, these pioneers weren’t predicting the demise of traditional methods such as the social survey. They were trying to work out what social media research can add to our existing set of methodological tools and what research problems may be better tackled by social media methods.

So the message is really to keep innovating and learning as there remains a lot we still don’t know. We don’t know enough about online behaviour – who does what, where, how and why –how different people are to their offline persona and how these blur and overlap. This is important if we see social media research having the capacity to say something about the offline. And beyond this epistemological question, we are also still experimenting with the best ways to tackle the problems of sampling, data quality and analysis that any method faces. But we can be open about tackling these challenges and welcome the scrutiny as a discipline of collaborative and continuous improvement. Keep using the network members and resources to make this happen!