I loved my time studying political science as an undergraduate. Data was critical to understanding and supporting positions but I was most inspired by the ideas. I was fortunate to have multiple internship opportunities and perhaps weigh that practical experience greater than my classroom education. However, I know that without the inspiration of the ideas, I would have never chosen the internships (and later the jobs in state government).

Amen, amen and amen! Could not agree more. In graduate school one reads the BIG works, e.g. Waltz, Morgenthau, Moore, Skocpol, etc. Those books tackled big questions and gave big answers that still shape the way we think about the international system or domestic politics. Nowadays, what gets published is so narrow and sterile that it makes me want to laugh or cry. Reading the top Political Science journals today is incredibly boring, both because the questions are narrow, uninteresting or obvious, and because the methods are so "sophisticated" that they are incomprehensible except to the few who practice them.

The figure showing the decline of policy recommendations is startling. I am in an applied science field where the point of many papers is to make policy recommendations (albeit to practitioners, not to government). The papers have to be scientifically rigorous, with statistical methods to support the policy recommendation over alternatives. The recommendation also needs to be highly relevant to current practice and the forces making the current practice suboptimal. Perhaps political science an do the same.

Some political scientists make and implement public policy within the government and its agencies instead of theorizing in journals. Condoleezza Rice, Henry Kissinger, Paul Wolfowitz, and Zbigniew Brzezinski played pivotal roles in war and peace. Some claim they had enormous control over "national security" policy and its implementation, so it is ironic to read:

As a society, we run into trouble when we lack policy-relevant academic perspectives. Indeed, there are instances — the war in Vietnam, the recent Iraq War — in which, had the majority consensus of scholars influenced policy, the country’s national interest would have been better served.

Here is a fictional political scientist (modeled on RAND Corp strategists (Herman Kahn?)) influencing outside the classroom. He encounters boy scouts.
https://www.youtube.com/wat...
http://www.tcm.com/mediaroo...

Excellent piece. Robert Putnam was the best professor I had in college.

I agree with most of what is argued in this piece, but it failed to raise an additional point: the Beltway is also quite interested in COUNTRY expertise, and so the shift away from foreign language and regional expertise in favor of large-N studies in which all countries are thrown into regressions to measure the central tendency is not very helpful.

Interesting that the chart graphing the character of articles in the APSR makes the author's point much better than the narrative he presents. There is no reason good policy analysis cannot be and is based on a methodology. My feeling is that those using methodology need to present the data as simplified as possible so that as many readers as possible can understand the point being made. This is exactly what the author does with Sigelman's graph..John D. Holm

The social sciences abandoned Darwin at the turn of the last century. It's the only theory that has successfully explained living things, including humans. It furnishes the paradigm for all of the social sciences, not behaviorism or rational choice. A return to Darwin and evolutionary theory will put the discipline of political science back on track.

This is an excellent piece on the topic of academia and its role in shaping political policy. I have one correction though. Leslie A. White was an American anthropologist not a sociologist. We anthropologists like to be clear on our theorists.

The author's presentation is a bit ironic. The most compelling argument he presents for is point of view is the Silegman data graph. However, the Silegman methodology might not be relevant to his point. Silegman only classifies policy relevance as making "policy recommendations". "Statistical" articles might well include research which could provide the basis for formulating policy or could be illuminating possible concerns regarding existing policy. The author seems to argue that unless policy recommendations are made the article does not have policy relevance. In short, the author would be well advised to make his case with more rigor, i.e. methodology.

I think quantitative analysis makes the field relevant and useful. The problem is that a lot of the top journals publish results that everyone already knew, but with a fancy new method (looks at regression discontinuity, propensity score matching, etc.). Or they do some experiment which tells us nothing interesting. Also, I think this new fetish for gender politics doesn't really help. That sort of undermined the humanities, IMO.

Qualitative research isn't use, IMO. Half the time those conclusions turn out misleading.

I think that what has happened is that political science has striven to be a science, and not just wise old nostrums and adages. The politics part of it has gotten farmed out to be a subset of public relations. My cognate interest for my Ph.D. was political communication, which I found infinitely more interesting that running calculus for the axiom's of Arrow's Theorem. I much preferred discussing which topics of discussion had passed through the Overton Window, and why. I was interested in how our image of the ideal candidate has changed, and about the limits of governance to effect social change. The author above is not only correct, but I think if we're honest, political science could stand to be more interesting, as well.