Will researchers play a ‘secondary’ role in future?
04 March 2013

In the next issue of IJMR (55/2) I report on two research related conferences I attended representing IJMR in December and January. As you will read, both of these focused primarily on the impact the digital world is having on research. 

But much of what was discussed was not, in my mind, market research as we know it. It seems to me that market, and social, research is moving away from structured methodology based on creating new data through questionnaires, focus groups etc – what has been traditionally called primary data - to a world where researchers are becoming increasingly involved in analysing and interpretation data that has already been created, or is created in a relatively undirected way – what might be termed secondary data. 

Researchers may be creating new insights, but the data was originally created for other purposes. This was the focus of both conferences, with little about ‘traditional’ research methods. 

Yvonne McGivern in ‘The Practice of Market Research’ (Pearson Education), the recommended textbook for those taking the MRS Advanced Certificate qualification, defines primary research as data that ‘do not exist prior to data collection’. Secondary data, in comparison, is defined by McGivern as ‘data that was collected for a purpose other than the current research objectives – in revisiting them you are putting the data to a secondary use’. 

As with most standard market research textbooks, the remainder of the book focuses on methods for collecting, analysing and communicating primary data based research projects. These are standard definitions, widely found in research textbooks – I’ve recently used them in my own teaching. 

Does the accepted definition of secondary data apply to research in digital channels, especially social media, or is this a new source of primary data?

Even the renaissance in ethnographic methods can be viewed as the deployment of another form of secondary research. The observed behaviour is often not directed by the research – it is being undertaken for other reasons, the role of the researcher is to interpret the observed behaviour and create the insights.

This change has important implications, for quantitative and qualitative research. The semi-structured outputs from focus groups presents the same challenge as found in social media data. Qualitative analysis techniques, such as correspondence analysis and text mining are being increasing applied by market researchers to data from these channels. 

In particular, it changes the basis of how the sector has evolved to date. 

The competencies covered within market research textbooks start to look out dated. The Codes of Conduct that underpin research remain based on versions of the interview process. If this trend to secondary data is accelerating, do we have a grasp on the new competencies necessary, and are these becoming part of the accepted training and education programmes for researchers? What does this mean for ethics? 

In addition, does this also remove the foundation of surety that comes from traditional primary research based methods, leaving researchers exposed as they cannot cite the principles of tried and tested research methodologies in defence of their work? Clients need to trust the research undertaken on their behalf, and researchers need to have confidence in the data they use. 

I’m not saying that the whole future of research is bound up with social media, or that this massive new source of consumer (or citizen) thinking and behaviour has no role to play in market research, but the changes to the sector could be profound. 

It’s an exciting new world, offering researchers opportunities that cannot be ignored, but does it change the basis of what we do from being primary researchers to data miners? 

Not that many years ago, market researchers saw data mining as the province of direct marketing, but now we can’t seem to get enough of it! 

Savage & Burrows raised major concerns about the future of survey research in the social sector, ‘Wither the survey?’ (IJMR Viewpoint Vol.50 Issue 3), does the same apply to market research? Does it matter?