I’m not sure that this is how I’ll process all of the books that I read – in fact I’m almost certain that it isn’t – but I’ll continue this series of posts about Trowler – Doing Insider Research in Universities because I have found it to be a great way to dip my toe into the many issues that I will face in my research.
The next two chapters look at value and robustness in insider research (which, again, I take to be about being able to defend your methodology – and choosing a good one) and then the ethics and politics of insider research in your university, which is pretty much unavoidable.
He opens with a discussion of some of the criticisms that case studies face as well as some of the responses to these. I’ll leave it here in full as it sums it up well.
Case study researchers may find their work subject to the following criticisms (Flyvbjerg, 2006): it only yields concrete, practical knowledge rather than the supposedly more valuable general, theoretical and propositional knowledge; generalization from one case is not possible and so the research does not contribute to the development of knowledge; the research can generate hypotheses, but other methods are necessary to test them and build theory; case study design contains a bias toward verification, i.e., a tendency to confirm the researcher’s preconceived notions. Both Flyvbjerg and Yin (2009) refute these criticisms, but those who research their own universities need to be clear about precisely what their research questions are, what the rationale behind the research design is, and what the truth claims are. This advice holds for any kind of research, but other designs tend to draw less critical fire.
(He also highlights Gomm et al (2000), Simons (2009) and Yin (2009) as great starting points for further investigation of case study methodology)
Trowler dips back into some of the ontological questions that he touched on earlier in the book, comparing the merits of the true vs the useful. (I may be oversimplifying this). This draws on Sayer’s notion of “practical adequacy” in prioritising the usefulness of information. I kind of get it but think I’ll need to dig a little deeper. I can see how some true things mightn’t necessarily be valuable but as for things that are kind of true…?
This is echoed in further discussions of Bassey’s idea of “fuzzy generalizations”. In short, this is about acknowledging that life is complex and theory won’t always accommodate the range of factors at play. So that rather than saying that in situation A, if B happens then C will follow, we might say in situation A, if B happens then C will generally follow between D and E% of the time. It’s not as neat and arguably not as helpful but no doubt more realistic.
In terms of the design of research, Trowler posits twelve questions for researchers to consider to test the rigour and quality of their proposed methods.
1. In designing the research, how do I know that my planned approach will answer the questions I have more fully than any other?
2. How do I design the research to take best advantage of the benefits of insider research while avoiding its pitfalls as far as possible?
3. Conceptually, how do I represent my organization, its culture and its practices? (And how does this representation shape my design?)
4. How and from whom will I secure access to the data I need? (Why them and why not others?)
5. Whom should I inform about the project, and how should I describe it, when I seek ‘informed’ consent? (And how might this affect my data?)
6. How will I ensure that the project is run ethically so that all participants and institutional bodies are protected? (While at the same time being as transparent as possible to readers so they can judge the robustness of my approach and conclusions?)
7. If I am using participant observation, what are the ethical issues in relation to the people I observe in natural settings? (And how might my answer to that question affect my data?)
8. If using interviews, what measures should I take to deal with interview bias? (And will they assure a sufficient degree of robustness?)
9. What should the balance be between collecting naturalistic data and formally ‘collected’ data? (And how can I offer assurances of robustness about conclusions drawn from both?)
10. How should I analyse the different forms of data I have, given that there will almost certainly be a large amount of various sorts? (And how do I ensure that sufficient and appropriate weight is given to each form of data in generating conclusions?)
11. How, and how much, will I communicate my findings to participants to ensure that they are happy with what I intend to make public? (And will this affect the way I present my conclusions to other audiences?)
12. Generally, in what other ways can I satisfy the reader about the robustness of my research and its findings?
(I was a little hesitant to just paste this in holus bolus but it all seems particularly valuable).
In very pragmatic terms, there will also inevitably be a number of ethical and political issues to consider when undertaking insider research in one’s own institution. The question of whether to anonymise the institution and the research participants is a live one – though at this stage, to me, it seems impractical and counterproductive, particularly when it could mean reducing the number of sources of data for fear of not being able to correctly reference them. The lack of transparency could also arguably lessen a reader’s view of the robustness of the research.
I also think that the question of to what extent people change their behaviour under observation is a valid one, however there are no doubt ways to mitigate this.
Politically, senior leaders in the organisation will imaginably want to feel confident that the research won’t damage the reputation of the university before granting access. Does this then lead to self-censorship and selective reporting if there are areas where there is room for improvement?
At a higher level of ethical debate, the selection of standpoints from which to pose questions and to begin observations and investigations can raise some concerns. If I fail to incorporate the perspective of voices that are less often heard, am I guilty of perpetuating the status quo?
Lots to think about and to be perfectly blunt, I would be naive not to factor in the question of what asking the wrong person the wrong question might mean for my career prospects in the organisation. Fortunately I have a little time to think about some of these things.