The power of well designed teams: Brian Uzzi

In this talk, Brian Uzzi presents his research on teams and collaboration, and shows: that established teams suffer from sunk costs, in that people want to leverage team specific resources to the max (such as knowing how each other work) to the detriment of including...

Controlling for confounding variables

Confounding variables are nothing new, so controlling form them are common. However, as this paper shows, a lot of these controls are flawed, calling into question the validity of the results. SEM is proposed as part of the solution.   Statistics: When...

How politicians poisoned statistics

In this article, Tim Harford uses the distinction put forth by the Princeton philosopher Harry Frankfurt, between those who lie with statistics, and those who simply do not care what the facts are, but use statistics to support their position. The latter can be...