collaborative analysis

Collaboration is a common feature of modern research life. Researchers may collaborate at every stage from seeking funding to disseminating findings. They might collaborate with other researchers, or managers, supervisors, participants, colleagues from other disciplines – anyone whose input seems likely to be useful, and who is willing to play along.

 

There has been little information about how to collaborate apart from the occasional blog post. It seems to be something people are supposed to understand how to do by some mystical kind of proximity. I’m happy to say a new book on the subject is now available for pre-order. Reframing and Rethinking Collaboration in Higher Education and Beyond: A Practical Guide for Doctoral Students and Early Career Researchers is by Narelle Lemon and Janet Salmons. This is a delightfully positive book which focuses on a strengths-based approach to making collaboration work.

 

There is also little insight into why people should collaborate. Sometimes people are required to collaborate by research funders, commissioners, managers or publishers. Some researchers seek collaboration for ethical reasons, perhaps because they are working within a participatory model, or they see the value of including different perspectives in the research team. However, research timescales and budgets can be barriers to collaboration, as it generally takes longer and costs more to involve more people in a project.

 

I am particularly interested in the role of collaboration in data analysis. This is the phase of research where people most often work alone. Even when teams of researchers analyse a dataset, they often work individually rather than collaboratively, and standard techniques are used to assess the level of reliability between individual quantitative raters or qualitative coders in an effort to achieve consistency. Is it a coincidence that data analysis is also a phase where researchers seem to be particularly susceptible to ethical breaches such as data manipulation or falsification? The Retraction Watch database contains thousands of examples.

 

Data analysis is a challenging task for several reasons. First, it tends to be badly taught and poorly understood. Journal articles reporting empirical research may contain only a sentence about analysis, perhaps stating which software and/or statistical calculations were used, rather than explaining why and how the analytic work was done. Even research reports with no word limits often give accounts of analytic work that are sketchy at best, which makes it difficult to learn about and from the approaches used by others. Second, even when a researcher is well-versed in analytic techniques, analysing data is an intellectually demanding task. Third, during analytic work, researchers are often aware of competing agendas from participants, funders, managers and so on. Participants’ wishes to be represented accurately and fairly, a funder’s requirement for the work to be completed on time, and a manager’s desire for significant findings may prove incompatible. Fourth, researchers may be unaware of their own biases, and of how others’ agendas and their own biases could be affecting their interpretations.

 

Working collaboratively is particularly helpful in reducing the impact of competing agendas and unconscious bias. Talking through analytic decisions and the reasons behind them can shed light on these and so enable researchers to focus more closely on their data. Indigenous researchers say that it is better to collaborate on data analysis than to work alone (Smith 2012, Walker 2013). The Indigenous research paradigm pre-dates Euro-Western research by tens of thousands of years, so we might do well to listen and learn (Kara 2018:22).

 

Of course budgets and timescales may get in the way of full-scale collaboration on data analysis. However, even small-scale collaboration can be surprisingly helpful. I worked with a colleague on a research project a few years ago which had a reasonable budget, so we built in half a day to sit together with our data and share our thoughts. This was very valuable in two main ways. First, although our worldviews are quite similar, we discovered that we perceive data quite differently. The resulting conversations helped me, as I took the analysis forward, to see more in the data than I would otherwise have done. At times it was almost as though my colleague was still sitting with me, saying ‘Look at that!’ and pointing out something I would otherwise have missed. Second, when my colleague gave her feedback on my initial analysis, it was much better informed than it would have been if we had not spent that time together working with our data.

 

So collaboration, in data analysis, has the potential to prevent malpractice, enrich the analytic process, and improve research findings. Now you don’t even need to be physically together because there are so many technological options that support collaborative working, from the general (Google Docs, Dropbox, Skype) to the specialist (SPSS, NVivo, Quirkos). Yet research ethics committees rarely consider data analysis at all (Kara 2018:47). I would like to see research ethics committees asking for information about how researchers intend to collaborate during data analysis, with whom, and why. Furthermore, I’d like to see them requiring a full rationale from anyone who does not propose to do collaborative analysis, with a clear explanation of the alternative steps they will take to avoid malpractice and enhance the analytic process.

 

We designed Quirkos Cloud to facilite live collaborative analysis of qualitative text data, if you are working from home or from across the world. Download a free trial and see for yourself how Quirkos makes qualitative analysis simple and intuitive!

 

Thanks to Helen Kara for this guest blog post, don't forget to visit her own blog and website at helenkara.com

 

 

References

Smith 2012, Decolonizing Methodologies, Zed Books, London p 130



Walker 2013, research in relationship with humans,, the spirit world, and the natural world, p 303-4, citing Begay and Maryboy 1998:50-55. In Mertens, Cram and Chilisa (eds), Indigenous pathways into social research: voices of a new generation. Left Coast Press, Walnut Creek, CA. (citing Begay and Maryboy 1998:50-55.)

 

Kara 2018, Research ethics in the real world, Policy Press, London, pp22, citing Cram, Chilisa and Mertens, 2013:11; Passingan, 2013:361; Steere, 2013:388-391; Wilson and Wilson, 2013:333.

 

Kara 2018, Research ethics in the real world, Policy Press, London, pp47

 

 

 

Tags : analysis ,   collaboration ,   teamwork ,   research