Not so sophisticated, afterall

I’m a fan of Vox, and I’ve enjoyed their coverage on gun violence in the US.

Their video on the “biggest” problem with lax gun control rules really puts most headlines in perspective. The old shark-attacks-make-better-headlines-than-cardiovascular-diseases argument can be applied here, too. Gun rampages capture the public’s attention more than suicides.

I laughed a bit when I read this, though:

“A recent, highly sophisticated study found that, once you control for general crime rates and other confounding factors, “each 1 percentage point increase in proportion of household gun ownership” translated to a 0.9 percent increase in homicides.”

Highly sophisticated? What makes this study so highly sophisticated, one might ask?

Fixed effects estimation and control variables.

It’s not that there’s anything wrong with using fixed effects – I used it in my undergraduate dissertation – but… I used it in my undergraduate dissertation. It’s commonly taught in introductory econometrics courses. It’s great but it’s a vary basic technique used to analyse panel data.

Another thing that supposedly makes it sophisticated is using ‘the largest-ever number of statistical controls for other variables in this kind of gun study’. Control variables? Sophisticated?

‘Comprehensive’ would have been a better word.

Not so sophisticated, afterall

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s