Most of what I read was pretty nice. One of the two admissions officers called an essay I wrote “a bit cheesy for my tastes,” which turned out to be a huge understatement; the essay was included in my files, and it was so dripping with schmaltz that I couldn’t make it to the end. At one place, one of my readers must have confused me with somebody else, because she made a note in her text box...
A good study on MOOCs
Key findings: 1. Participation in HarvardX and MITx open online courses has grown steadily, while participation in repeated courses has declined and then stabilized. 2. A slight majority of MOOC takers are seeking certification, and many participants are teachers. 3. Academic areas matter when it comes to participation, certification, and course networks. 4. Those opting for fee-based ID-verified...
How to describe your p>.05 results
The solution is to apply the time-honoured tactic of circumlocution to disguise the non-significant result as something more interesting. The following list is culled from peer-reviewed journal articles in which (a) the authors set themselves the threshold of 0.05 for significance, (b) failed to achieve that threshold value for p and (c) described it in such a way as to make it seem more...
“Engagement isn’t necessarily a recipe for academic gains”
At schools around the globe, girls outscore boys, and bored students are better test takers than their more motivated peers. These topsy-turvy observations are the latest findings in a report from the Washington-based Brookings Institution, research that is part of a long-running series that aims to put a finger on the pulse of academics in the United States and abroad. The study by...
If you have ALS, what is a statistically significant result?
For people with amyotrophic lateral sclerosis, which attacks the body’s motor neurons and renders a person unable to move, swallow or breathe, the search for an effective treatment has been a crushing disappointment. The only drug available for the disease, approved two decades ago, typically extends life just a few months. Then in the fall, a small California biotech company named Genervon began...
Inside the academic fascist mind
A stunning and absolute must-read confessional from a recent college graduate. Some observations: She reads like a present-day Whittaker Chambers. The fact that she is afraid to use her real name speaks volumes about her comrades and their tolerance for dissent. I, too, have noticed that social justice warriors tend to be a miserable and unhappy lot. She provides some insight as to why. First...
Peter Thiel: higher ed is fucked up and everyone knows it
From a great interview of Thiel by Tyler Cowen: Well there are lots of things that are true that everyone agrees with me on. I think for example even this idea that the university system is somewhat screwed up and somewhat broken at this point. This is not even a heterodox or a very controversial idea anymore. There was an article in TechCrunch where the writer starts with “this is going to be...
Thinking about education, mobility and inequality
Is education the key to reducing inequality? Turns out, probably not: Brad Hershbein, Melissa Kearney and Lawrence Summers offer a simple little simulation that shows the limits of education as an inequality-fighter. In short, more education would be great news for middle and lower-income Americans, increasing their pay and economic security. It just isn’t up to the task of meaningfully reducing...
Do universal pre-K programs work?
There is another issue she does not mention: it can be be difficult, if not impossible, to successfully bring to scale those small interventions that work under carefully controlled contexts. The problem is that there’s no evidence that universal pre-K comes even close to its touted capacity to move the needle for disadvantaged children. Pre-K advocates widely cite two well-run demonstration...
How to interpret interaction terms
I keep referring to this issue in manuscript reviews, so I thought it worth a post. If you include an interaction term in a model, the statistical significance of the main effects and the interaction term tells you nothing about the interactive effect. This seems contrary to your intuition. Turns out the interactive effect may only significantly differ from zero for part of the covariate space...