Sometimes, a simple bar graph is all you need. This one comes from a report about class-based alternatives to (unpopular and legally imperiled) race-based affirmative action programs for selective college admissions.

At the top 146 colleges – defined as Barron’s “most” and “highly” competitive categories, including the Ivies but also big public schools like UCLA and UVA and some liberal arts colleges – less than 10% of students come from the bottom half of the socioeconomic distribution.* Owch. I knew my own school – Michigan – had gotten that bad**, but I didn’t quite realize how pervasive it was across selective colleges.

* See Carnevale and Rose for details. It’s a composite measure of reported income and parental education/occupation to deal with some reporting bias, apparently.
* A full 30% of UM undergrads have parents making $200,000 or more. About 10% have parents making under $40,000 per year.

Nathan Yau at Flowing Data links to this lovely analysis of average dissertation lengths at Minnesota. Here’s the key chart:

Two random thoughts about this. 1. History, as befits its reputation, really is a bit of an outlier in having long dissertations, with Anthropology a close second, and 2. Sociology has one of the widest boxes, probably reflecting the diversity of methods employed in the field (quant dissertations being short like econ, historical dissertations being long like history). Economics has one of the tightest boxes. All of which suggests that a tight page range is one (kind of silly) measure of disciplinary uniformity (of a sort, anyway).

Sociologists often look at Economics and see a field dominated by rigid formal theorizing – utility-maximizing rational choice etc. Our image of economics often looks like an intermediate undergraduate textbook. But what’s actually going on at the forefront of the economics research frontier? This table from Hammermesh (2013)’s Six Decades of Top Economics Publishing: Who and How? shows some revealing trends in the methodology of economics from 1963 to present:

The big findings, as I see it, are the rapid decline in pure theory articles and the big increase in the use of proprietary data. Mid-20th century economics involved a lot of formal theorizing in both macro and micro (Samuelson’s program), and a lot of empirical research using newly available kinds of standardized data (e.g. the National Income and Product Accounts and their non-US equivalents). Now, empirical work is more likely to come from proprietary or experimental data than secondary sources. Finally, though there is an uptick in experimental studies, they are a pretty small fraction of papers still.

For more fun facts about econ journal publishing, see also Card and DellaVigna (2013) Nine Facts about Top Journals in Economics. Also, kudos to the AEA for releasing the full-text of JEL articles for free!

Posting has lagged a bit as I’ve been back and forth between collecting and processing archival data. This slower rate of posting will likely continue for a few weeks, at least, though I have a special post planned (an author Q&A) for sometime this month. In the meantime, I might post some shorter snippets that catch my eye… like this chart from Hallerberg and Wehner’s The Technical Competence of Economic Policy-Makers in Developed Democracies. Hallerberg and Wehner are interested in the conditions under which governments are more likely to have technically competent economic policy-makers – understood here as graduate training in economics, though the authors have more measures than just this one – as opposed to policy-makers with stronger political backgrounds (such as previously elected officials). One straightforward finding, to give an example, is that presidential systems are more likely to appoint technically competent finance ministers than parliamentary systems (where ministers are usually themselves elected MPs). This chart does a nice job of showing the relative proportions of advanced economics degree holders among prime ministers/presidents, finance ministers, and central bank heads:

Other findings include the sensible one that finance ministers and central bank heads appointed during financial crises are more likely to have technical competence and that left-wing governments are especially more likely to appoint economists during crises. Check out the full paper for details.

Note: VotD stands for “Visualization of the Day,” and may become a recurring (though certainly not daily) feature.