[Interpretationandmethods] Interpretationandmethods Digest, Vol 43, Issue 1

[Patrick Thaddeus Jackson]

On Jul 6, 2008, at 6:11 PM, Judy Brown wrote:

> 3.  The student about to start will be examining the involvement of  
> accountants in the decision-making processes surrounding information  
> technology governance. In particular, she wishes to explore the ways  
> in which the ideas and attitudes of accountants and IT practitioners  
> might enable or constrain the involvement of accountants in IT  
> governance.  An interpretivist methodology will be adopted in order  
> to explore the impact of various "ideational" aspects on IT  
> governance design choice e.g. the ways in which IT governance is  
> conceptualised by accountants and IT practitioners, interpretive  
> schema and institutional 'habits', professional identity,  
> organisational culture and politics. I am meeting her later this  
> week, and will encourage her to join the list so she can tell you  
> more about her interests.


Judy:

One question I would ask the student is what, if anything, is  
"interpretive" about this research question. "The impact of various  
'ideational' aspects" sounds like pretty standard neopositivism to me:  
code a variable like "professional identity" or "organizational  
culture" and then test to see whether the variable is significantly  
correlated with an outcome of interest across cases. I see  
interpretive methods here, but no interpretive methodology. "Attitudes  
cause an outcome" is not the kind of proposition one needs anything  
complicated to evaluate; Mill's Methods ought to do the job just nicely.

Of course, one could also ask a research question here that *would*  
require a non-neopositivist research design. But "the impact of X on  
Y" wouldn't be it. For example, one could propose an ethnographic  
account of accountants working in IT governance, seeking to bring to  
light the interpretive horizon within which they work; that isn't  
hypothesis-testing, but it also isn't causal (so such a study couldn't  
be about "impact"). Or one might make a relational turn, and discuss  
IT governance as emerging someplace between a variety of conflicting  
parties and strategies, and hence examine the detailed transactions  
that take place between (e.g.) IT practitioners and accountants in the  
process of constituting "IT governance." That's causal, but it's not  
"X causes Y" (or statistical-comparative) causal.

I think it's great that you have a multiple-methodology sequence for  
your students; I agree, that's the best way to allow students to  
experiment, and to find the right methodological approach for the  
question at hand. But the flip-side of that, I think, is that if one  
has a neopositivist question then one should use a neopositivist  
methodology in an attempt to answer it.

PTJ
===
Patrick Thaddeus Jackson
Director, General Education Program, American University
Editor-in-Chief, Journal of International Relations and Development
http://profptj.blogspot.com | http://www.kittenboo.com
calendar: http://ical.mac.com/onyxdr/Patrick