Data inanities

I feel like half the time I read something from one of these data news explainer sites, I want to blog about how silly it is. So while I’ve been wrestling with what to write here regarding a series of terrible NYT op-eds (no, I won’t link to them, but you know which ones I’m talking about), I will take a minute to call out 538 for publishing this article complaining that giving students free lunch is going to make data analysis difficult.

It’s absolutely true that students receiving free lunches is a proxy for student poverty. In fact, in my own teaching, we talk about proxy variables by examining a data set of school characteristics and students achievement scores. We actually run regressions where I encourage students to think about socioeconomic status and poverty through school lunch programs (along with other measures). But it’s also a rather coarse measure. In the way that school lunch programs have traditionally been applied, if you fail to meet some income threshold, you get free lunch, and in some cases, free breakfast. In Colorado, for a family of four, it’s $44,123. While it’s useful for looking at broad categories, it doesn’t tell you anything about the heterogeneity within those categories. The number of kids qualifying for free lunch could be the same at two different schools, but if one school is in a relatively homogenous district with most families hovering around the cutoff point and the other pulls from one very rich area and one very poor area, looking at those schools as the same actually “muddies” the waters” more than diluting the program.

So, it’s not actually a great measure, anyway, which we’ve kind of already covered by calling it a “proxy.” So why not look for better measures? The article mentions education levels of parents; that’s a good one. Or economic variables of the surrounding districts could work. Property values, for instance, are widely available and could be linked to school district. This is a little more work perhaps, because often these variables aren’t automatically linked to school quality data.

It’s true; we don’t like change. And changing a commonly used measure of poverty means looking for new answers, and that trends over time will be a bit difficult to determine for awhile, but with a little hard work and ingenuity, the new answers should be better. Decrying the end of a poor measure of socioeconomic status when its expansion will actually help a lot of kids at the margin is just not very useful. Why not spend a little more time thinking about how we can make data better, answer questions more fully, and ultimately improve school experiences for kids?


Central Limit Theorem in action

I had my #lafecon213 students run Monte Carlo simulations in class yesterday using a program we wrote in Stata. After we’d done the general one, I told them to change something about it and see how it affected the sampling distribution of the coefficient estimates. One student decided to run 100,000 repetitions of the simulation, not realizing what a time suck it would be. It took most of the rest of my lecture (surprisingly long, now that I think about it; perhaps I should complain to IT? I just tried it, pretty sure he did 1,000,000 repetitions), but when he finally had a histogram, I put it up on the big screens in my awesome smart classroom, broke into a huge grin and exclaimed, “isn’t it pretty?!”

If they didn’t think I was crazy before, they definitely do now. It took at least three minutes for them to stop laughing at me.

You have to admit it’s really pretty, no?

Monte Carlo reps output

Note-taking, live-tweeting, and first week of class update

My twitter assignment for my econometrics classes has garnered a bit of attention over the last few weeks. As we had our first assignment, reading Charles Wheelan‘s Naked Statistics, along with the regular assignments, my students interacted with each other,

with Lafayette College communications and library accounts,

with Alison Byerly, president of Lafayette College,

with some of my colleagues, and even with Stata. My favorite post of the week came when a student tweeted at me to ask whether a follow from Stata earned him extra points.

I’ll let you guess what the answer was.

Lafayette College has a social media working group, a relatively informal gathering of social media practitioners on campus, who asked me and another professor to come by and give a short presentation on how we were using twitter in the classroom. Last semester, Chris Phillips had students live-tweet his seminar on Moby Dick, while my assignments are mostly out of class, at least at this point.

A few days later, a colleague tweeted an article on a new psychology paper showing that students process information better, and get more nuance, when they take longhand notes as opposed to typing verbatim a professor’s lecture. While this justifies my reluctance to give out class notes, it also got me thinking about whether live-tweeting would be more like longhand notes or like typing notes. One of the participants in our twitter in the classroom discussion asked Chris whether the students who live-tweeted did better or worse than the others. He didn’t feel he could really make a statement either way (he already knew one of the students, not to mention the statistical power issue with a small sample size), but it’s a question worth asking. Assuming we take the study at face value, does the power of note-taking come with the physical process of writing out letters? Or is there something particularly damaging about typing verbatim that limits processing? And which process does live-tweeting, where you are typing, but have to process and condense information fairly quickly, mimic more closely?

h/t @betsylevyp

Prof. Fletcher’s Guide to Tweeting in the Classroom

The past few months have been full of new research projects and new ideas for me. I’m exploring sexualized violence among Colombian ex-combatants, obesity during pregnancy, and female labor force participation in the American South. I just got a small grant from Lafayette’s Digital Humanities Mellon Foundation Grant to study the last of these and I couldn’t be more excited.

Teaching-wise, I’m also ramping up the innovation. In particular, I’ve been thinking a lot about how to have my students read and interact more next semester. I’ve settled on twitter. In the past, I’ve asked them to blog, and it’s been fantastic, though I think it works better with smaller classes. This Spring, I’m trying out tweeting to see if I can’t engender some networking skills while focusing on brevity.

Here is Professor Fletcher’s Guide to Twitter for my students, specific to this semester. It includes some assignments and some general guidelines, assuming they either know the basics or can figure them out fairly quickly. Comments, thoughts, ideas, are much appreciated.

Happy New Year!

Random thoughts on gender and professorships

I was recently privy to a discussion among some new faculty about what to have students call you, where “you” is a young, relatively freshly minted Ph.D. and new professor. There were lots of varying viewpoints, many of which I’ve heard before, ranging from “women just can’t have their students call them by first names and maintain distance and respect” to “I felt so much more comfortable when my professors let me call them by their first name and I went to office hours, so I let my students call me by my first name,” to “I like the ego trip that comes with being called professor so-and-so.” The gender issue is one that is particularly sticky and I’ve discussed it with many female colleagues at every stage of their careers. I come down on the side of staying Professor Fletcher until a student graduates. Having been subject to lovely gender- and age-based attacks such as “my male professor in another course says everything you’re teaching is crap,” and “you’re unprofessional” has only heightened this resolve.

During this discussion, one colleague framed the answer to this question in a way that I thought was incredibly insightful and a great use of his privilege, as a male in the classroom, to even the playing field. He said that a female member of his doctoral dissertation committee had told him under no circumstances to allow students to use his first name. The reason, she said, was that she couldn’t, and so he shouldn’t.

It seems rather simple, but I’d honestly never thought of it. Given the differential treatment women often face in academia, male professors exercising the privilege of letting students use their first names and subsequently seeming “more accessible” is actually a detriment to women’s careers. Another female colleague, who is up for tenure this year, told me that she felt that being hard on her students early in her career had hurt her student evaluations, and thus her chances for tenure. Though she doesn’t like to make such distinctions and would never say it, that’s a conversation I’ve never had with a man.

I do think it’s getting better, but until it’s actually better, we should all do our part. And having everyone be addressed the same way is a good way to level the playing field.

As a side note, the New Faculty Program through the Center for the Integration of Learning, Teaching and Scholarship at Lafayette College is pretty great.

Primes and Probability

At the suggestion of a colleague, I recently started reading Charles Wheelan’s Naked Statistics: Stripping the Dread from the Data. So far, it’s a fun, demystifying sort of book, the kind I hope my students will enjoy (watch out Lafayette Economics, I’m coming, and I will make you read). It rests on the twin ideas that statistics can be fun and statistics are incredibly useful to explain, to tell stories.

The book was high in my mind this morning when I read this deliciously accessible Slate piece by Wisconsin professor Jordan Ellenberg about an advance in prime numbers by a University of New Hampshire mathematician. Economists like to make lots of bad jokes about how they are failed physicists, who are in turn failed mathematicians, so while it interests me, I wasn’t expecting to really understand the discovery when someone riffed on twitter about primes.

What’s so wonderful about this really intense mathematical discovery, at least according to the mathematician author of this piece, is that it’s really about statistics, which I can totally get my head around. The theory goes that primes come in infinitely many ‘twin pairs,’ like 3 and 5 or 17 and 19, and the intuition lies in that we can think about primes as random numbers.

And a lot of twin primes is exactly what number theorists expect to find no matter how big the numbers get—not because we think there’s a deep, miraculous structure hidden in the primes, but precisely because we don’t think so. We expect the primes to be tossed around at random like dirt.

Zhang didn’t quite prove the twin pairs theorem, but he made an important step towards proving it, it seems, and understanding probability and statistics is key to getting there.

Why study social sciences?

Monkey Cage Blog has a great post up in response to criticism of his exhortations to study social sciences. He makes a broad argument about the validity of social science research because it has effects on the way that people live their lives. To be selfish for a moment, he highlights some important questions that I examine every day:

Families.  What makes families more or less successful?   What makes marriages more successful?  What makes them fail?  What are the effects of divorce?  Does it hurt the children of divorce?  How much, in what ways, and for how long?  A medical doctor can treat the effects of family dysfunction and divorce—say, with anti-depressants or therapy and so on—but we can learn and know more about how to prevent some of this dysfunction from doing social science.

The post is really about funding for social science research rather that defending my everyday work. It’s also not really about teaching undergrads social sciences, but clearly, we have train undergrads in social sciences if we eventually want some of them to do research in the social sciences. I think there’s a point to be made about how learning about these wide-reaching social phenomena—families, schools, economies, politics, attitudes, networks and norms—forces students to think about cause and effect in a nuanced way. When it’s not clear how X might affect Y or how Z has effects on X that in turn effect Y, it takes creativity and imagination and critical thinking to sort it out. It’s not that social sciences can do this exclusively, but the nature of the topics student lends itself to varied analysis and the development of skills that are useful in many careers.