Leaving Social Science to Social Scientists

(Well, I tried to post this as a comment, but apparently it was too long, so here we go - this is my comment to Namalie's post)

Are we doing that 'well it is the end of the semester, so I don't need to be politically correct' blog posts now? Please say yes, because that's what I have in mind right now!

I have a 'mild' position when it comes to answering questions such as what is social science, what is reality, and how can a research claim 'truth' (or make truth claims)? It is probably because I spent a couple of years at a sociology department at a 'technical' university (yes, I have a bachelor of science). The department was working day and night to differentiate itself from 'humanities' and to prove that sociology is a science.

Anyhow, I consider myself to be (or aspiring to be) a social scientist. And I was highly offended by Flyvbjerg (F9) book. I also find it difficult to call F9 a social scientist, not because of his ideas but because of his field of study. At the end of the day, this book sounds like the ideas of a non-social scientists about what he believes to be social science research. I have three main reasons for that: (i) his inconsistency in explaining the debates about social sciences, (ii) his interesting incorporation of Dreyfus&Dreyfus's learning styles into the debate, and (iii) his 'last' recommendation about activism.

Before the discussion in class, I had no idea what F9's main argument was. He starts with a discussion of contextual vs. context-independent information, then moves to a qualitative vs quantitative debate, and ends up with activism vs. pure research (?) debate. Firstly, these debates are not related with each other at all. Looking for context-independent knowledge does not necessarily require the use of quantitative data or to ignore activism. Besides, his understanding of context-independent was 'underdeveloped'. Context-independent does not necessarily mean ignoring human interactions, traditions etc. (see p.38-9).

Expert level in research... I don't know where to start. F9 gives the example of pilots: Novice pilots "fly their planes," while..experienced pilots "they fly" (p.19). Yes, when you gain competency in a skill, the way you do things change. The step-by-step thinking is no longer taking place conciously. If you guys were ever in a position to talk about your own research, you must have realized how difficult it is to explain it - but that does not mean there is a system that we are following! When I face a new case, I don't go in and look around randomly (I am sorry, the correct word is -I believe- intuitively, not randomly). I have methodological assumptions, analytical tools, methods to guide me. When I want to write a paper, I have an idea about how it should be structured. Maybe I don't follow strict guidelines to see what I should include in each part (abstract, introduction, etc) but that doesn't mean it is all intuition. Long story short, social scientists follow rules and procedures in reading, writing, teaching. When these rules and procedures are retracted to the background, that does not necessarily mean we try to comprehend everything as a whole (like the soccer player who score with a never-practiced play before), we know where to look because of our training.

Activism... is good (by the way, I am going to overlook consultancy here. Let's assume you are a sole activist). It is nice - sometimes it helps us change things. There we have another inconsistency - as Namalie quoted "we cannot find ultimate answers to these questions or even a single version of what the questions are". If you get this quote and put it on top of activism, what will keep any """researcher""" (I had to use triple sarcastic quotation marks to make my point) from manipulating data or results? Especially given the fact that there is no structure for these processes.

Long story short, some of you might know my dislike towards KKV. [And my dislike is mostly based on the fact that they boil down the methodological debate to quantitative/qualitative(actually to case studies/case analysis) and upgrade the latter with former so that we can use cases to create hypotheses.] But if I were in a two-party system with KKV and F9, I would definitely vote for KKV. As social scientists, we should look for generalizable (either across cases or across time) knowledge. We should be able to make validity claims (assuming that we except external social reality). A strong relativist approach (surrounded by I am not a relativist claims) is not the way to make social science matters. Robust research designs, sophisticated techniques, and actually easy-to-understand visual representation of data, in my humble opinion, are the ways to make sure social science matters again.


PS: Just for fun, this is the federal definition for 'research' (coming from IRB definition for human subject research):Research is a systematic investigation, including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge.