Statistical Teaching (bleg)

Ok, in my research methods class, we are hitting an overview of statistics in the closing weeks of the semester. As such, i would prefer to include some empirical examples to visualize the things we’re going to talk about that are fun / outside my typical wheelhouse. So, do you have any favorite (read: typical, atypical, surprising, bizarre, differentially distributed, etc.) examples of univariate distributions and/or bivariate associations that may “stick” in their memories when they see them presented visually? I have plenty of “standard” examples i could draw from, but they’re likely bored with the one’s i think of first by this point in the term. So, what are yours? It’s fine if you just have the numbers, i can convert them to visualizations, but if you have visual pointers, all the better.

(cross posted)

how many, indeed?

From class to news to research question. So, this morning in class I taught an article using the network scale-up method. It’s a great technique that’s been used to explore a number of interesting questions (e.g., war casualties, and HIV/AIDS).

I came back from that class to this article pointing to a debate on voter ID laws, and I couldn’t help but think that there has to be a meaningful way to throw this method at this question to estimate plausible bounds for the actual potential impact of these laws. And furthermore, it seems especially important because people without IDs are likely quite hard to accurately enumerate on there own (as are those who’ve engaged in voter fraud).

So, has this study already been published and i just missed it? Else, does someone have the data we’d need for that? I’m hoping it’s a solved question, as i assume its something it would be better to have known a few months ago than a few weeks from now. Anywho, just puzzling over a salient question that linked together some events from my day.

(Cross-posted)

Help / Discussion lists for R packages

If you want to learn a methodology, there may be an email list you should be on.  The two big network analysis packages in R  Statnet and igraph each have one (sign up: Statnet, igraph, Mixed Models).  If you join them, you can ask questions when you get stuck.  But you may end up learning even more from other people’s questions.  Jorge M Rocha stimulated Carter Butts to write a mini-essay on exponential random graph models which I received permission to repost.  Dave Hunter also adds some thoughts at the bottom.

False dichotomies (and synonymies)

It’s been a while. Let me allow you a few moments to catch your breath over the surprise of me posting (a real post) here again.

…[twiddles thumbs] [taps foot] [checks watch]…

OK, feel better? On with it.

Back in the middle-stages of grad school, i started to hear a lot of harumphing* about the frustrations of qualitative folks and all of the “quant-shop” requirements of our particular program. At the same time, i couldn’t help but notice the dismissiveness of some of the most quantitatively oriented folks towards brilliant qualitative work we’d occasionally discuss. Now, i know that the quant-qual divide was not unique to our program, is not something new to sociology, and has been pointed out as a false dichotomy by many (more qualified) folks who’ve passed through these ranks before.

But what strikes me today is that for some reason, in my limited experience at least, most folks also assume a necessary overlap between “mathematical” sociology and “quantitative” sociology…

The Credibility Revolution in Econometrics

Angrist and Pischke are on a tear.  They’re bringing econometrics to the masses with their new book, and the editors of the Journal of Economic Perspectives have seen fit to publish a debate around their article assessing the state of econometrics.  A&P claim, and I more or less agree, that microeconometrics has undergone an inspiring “credibility revolution.”

The best summary I’ve found of their article is by Austin Frakt, here.  Arnold Kling comments here.  Andrew Gelman reviewed their textbook positively and constructively here.

Angrist’s website gave ungated links to most of the comments on his paper:

Michael Keane, Edward Leamer, Aviv Nevo and Michael Whinston, Christopher Sims, and James Stock

Added 6/3:

Austin Frakt reviews the Mostly Harmless Econometrics.

Mostly Harmless Econometrics has a blog!

Social Network Packages Poll

Gabriel Rossman is running it here.

Do we want to be unified?

Teppo Felin already blogged this at Org Theory but I thought I’d raise one question about Herb Gintis’s proposal for the unification of the behavioral sciences (paper and lecture).  My question is this: would unification of the behavioral sciences discourage methodological and theoretical innovation?

As an interdisciplinary scholar, I am often frustrated but my fellow social scientists lack of regard for the insights gained in sister disciplines. Unification would seem to fix that problem, but some might argue that more unified academic standards would discourage  innovation.  The idea is that each discipline is currently like a separate experiment, and unifying them would be putting all our eggs in one basket.

I’d be interested to hear what other people think about this argument, but I’m inclined to believe we can pursue unification and intellectual diversity at the same time.  (post edited for clarity)

Biased in favor of your null hypothesis

Another great post on orgtheory.  I think the takeaway is that when it is really hard to prove something conclusively people fall back on their personal and disciplinary biases.  Rafe Stolzenberg has more than once reminded me how economists’ strong theory is a double-edged sword.  Economists need to be reminded that even if the data you’re currently analyzing can’t reject every possible rational choice model doesn’t mean you shouldn’t take alternative models seriously.  In fact, placing such a heavy burden of proof on alternative models seems quite irrational to me.

Exploratory vs. Confirmatory Data Analysis

Seth Roberts vs. Andrew Gelman

In most respects they are actually in agreement (e.g. both think that visualizing data deserves more attention than it often gets) but Andrew focuses on points of disagreement (e.g. do statisticians drastically overvalue difficult but less useful research).  Highly recommended.

Hirschman on Rational Choice Theory

Dan Hirschman, grad student at the University of Michigan, has a great post on rational choice theory.  It is framed as a critique, but I consider it wholly compatible with my defense of RCT. Human behavior is complex and different aspects of it will be best understood with different theories/models/levels of analysis.  See some of my previous posts on modeling here.