Will COVID-19 increase Secondary Data Analysis?

Sarah Hadfield is an ESRC doctoral researcher in Social Policy and Criminology at The Open University. Her research explores childfree young women’s experiences of non-permanent employment and financial autonomy. In this post (published 12 May, 2020), Sarah discusses if social distancing will increase the use of secondary data in research projects. She argues that Secondary Data Analysis can be used in a different way from how it has been traditionally been used in in this time of interruption to researchers’ usual practices.

The response from the academic community to COVID-19 has been swift with thought-pieces and rapid response research. Although most are supportive of this response, others have questioned the ethics of doing research at this time and queried why some are able to do research while others are not. My own position is that I am grateful to those managing to document the impact of COVID-19; the absence of this would be cause for more concern. It is analysis of data that has debunked the myth that the virus is indiscriminate and that ‘we are all in this together’. Figures out last week illustrate that it is Black Britons who are twice as likely to be impacted by COVID-19. Further research is required to understand the reasons why and the long-term consequences. Of course there will be questions around who is able to do research at this time; however, we should continue to respond to researching COVID-19 and if we can continue with our own research considering its impact.  For my colleagues, the material impact differentiates - for some it will simply be impossible to transfer to software or virtual interviews or change their methodologies, while for others the transfer will be easier. The design and delivery of research will need to take place within the rules of social distancing and lockdown, and even if it eases it might be that in-person contact will be reduced for a significant time. It is with this in mind that I consider: Will COVID-19 increase Secondary Data Analysis?

Secondary analysis is the re-analysis of either qualitative or quantitative data already collected in a previous study, by a different researcher normally wishing to address a new research question

Payne & Payne, 2004 :214

It was the processing of questionnaires from two formative 1980s youth studies projects that sparked my passion for research methods, but it taught me that secondary data is divided into a further two camps beyond that of qualitative or quantitative research: analysing data that was collected for secondary analysis and that which was not. Questionnaires such as Understanding Society and the British Household Panel, or qualitative data from the Timescapes and the Mass Observation are designed with the idea that they would be used for secondary analysis. The data I prepared for re-use was not, resulting in methodological possibilities and a greater understanding of how the data was generated as well as unforeseen challenges. I do not think anyone would argue that most academic studies are designed considering someone else will want to re-analyse the data collected. Exploring data that was not intended for re-use means the researcher will need to consider that they might be an outsider and not privy to all discussion and decision-making, and although this could be viewed as a barrier, it also is an opportunity to make methodological advances.

The beauty of Secondary Data Analysis is that in most cases it does not require the researcher to leave their desk; therefore, it could be a partial solution to the stopping of most face-to-face contact. From my own experience of analysing Understanding Society data for my PhD and looking at the Labour Force Survey, I have observed that if I were to design these surveys, I would adapt them to be more suitable for my own research questions, as they presently stand limit the conclusions I can draw. I think, ultimately, because the ideas we develop to conduct research will never neatly sit in with the way someone else collects data, this means that Secondary Data Analysis will not and should not be considered a solution to the restrictions COVID-19 has placed on our research. What I would suggest is that we use secondary data in a different way to potentially reduce the amount of new data we need and to design better research projects in this time of limited social contact.

Instead of looking at Secondary Data Analysis as a singular method, I suggest we continue to take this opportunity to incorporate it in our original research studies to answer some questions and to influence how we design our projects. Critically considering someone else’s tried and tested method and measurement to develop new research can only bring robustness to the project. Where possible, discussing with them the type of data it produced and what it missed creates collaboration and fosters best practice. Furthermore, viewing deposited data can make you consider what you would do differently. Limitations of the data deposited in archives is often explained too. Research is more important now than ever, but collaborative, open, inclusive and ethical research is most important. This is the best way to honour the legacy of those who have deposited data and answer the call for collaboration in this time of interruption to researchers’ usual practices.

Read Sarah's earlier blog, Young Women & COVID-19: We are not all in this together