Comments on Homework #1

Ahead of the upcoming test, I have some comments on Homework #1. Question 1: everyone did well, it’s straight forward. Question 2: also no comment. Question 3: some people were 100% correct, but don’t forget to include the value of intrinsic costs. Many people correctly mentioned time with family/children/spouse as an opportunity cost, but didn’t […]

Preview: Visualizing Transitions to the Workforce

The Iowa Department of Education and Iowa Workforce Development have teamed together to track students from community college into the workforce. While we’ve looked at returns to degree and in-state retention rate, we wanted to see where kids were working. In short, we looked at majors and tracked them into the industry. We aggregated majors […]

James Heckman on NPR

James Heckman, the Nobel-prize winning economist from the University of Chicago, spoke briefly on National Public Radio. My favorite part was Michael Martin asking why an economist (and there are quite a few of us) is interested in early childhood education: MARTIN: I hope the question doesn’t bore you. But there are…many people who will […]

Baseball Stats

ESPN’s interesting article on pitch count also contained some ugly graphs. The articles thesis is on starting pitchers now regularly pitching fewer than 100 pitches. Exhibit 1: Transparency and Area Charts Area plots don’t work well when they overlap. The above chart had to use transparency, which makes it difficult to associate colors with the […]

Selection Bias and Online Dating

New York Times has an article on the science behind online dating sites, like eHarmony and Chemistry.com, that base matches on intensive surveys. The article notes: Gian Gonzaga, senior research scientist at eHarmony, said that studies his company had conducted of married people who met through eHarmony and a control group who met in other […]

To standardize or not standardize?

Andrew Gelman compares standardized and unstandardized coefficients. Gelman has previously written a paper which addresses most of the shortcomings to unstandarized coefficients (e.g., raw coefficients, the stuff you see as a default) and addresses the issues brought up in the post. Unfortunately, he didn’t address the problems I was most interested in seeing. He writes: […]