27 August 2010
If you enjoy Australian politics, betting markets, and sharp statistical analysis, take a look at Simon Jackman's blog. He has been killing it lately.
You may think you have good reasons to not stop what you are doing and read Phil Schrodt's essay on the "Seven Deadly Sins of Contemporary Quantitative Political Analysis". But you do not. Not only does the piece make several astute points about the current practice of quantitative social science (in a highly enjoyable way, I might add), but it also reviews developments in the philosophy of science that have led us here. The entirety is excellent, so picking out an excerpt is difficult, but here is his summary of our current philosophical messiness:
I will start by stepping back and taking a [decidedly] bird's eye (Thor's eye?) view of where we are in terms of the philosophy of science that lies beneath the quantitative analysis agenda, in the hope that knowing how we got here will help to point the way forward. In a nutshell, I think we are currently stuck with an incomplete philosophical framework inherited (along with a lot of useful ideas) from the logical positivists, combined with a philosophically incoherent approach adopted from frequentism. The way out is a combination of renewing interest in the logical positivist agenda, with suitable updating for 21st century understandings of stochastic approaches, and with a focus on the social sciences more generally. Much of this work has been done last decade or so in the qualitative and multi-methods community but not, curiously, in the quantitative community. The quantitative community does, however, provide quite unambiguously the Bayesian alternative to frequentism, which in turn solves most of the current contradictions in frequentism which we somehow--believing six impossible things before breakfast--persuade our students are not contradictions. But we need to systematically incorporate the Bayesian approach into our pedagogy. In short, we may be in a swamp at the moment, but the way out is relatively clear.
His section on "prediction versus explanation" is also quite insightful and deserves more attention. The upshot:
...the point is that distinguishing scientific explanation from mythical (or other non-scientific, such as Freudian) explanation is one of the central themes for the logical positivists. In the absence of prediction, it cannot be done.