A data note, for people just getting into empirical reserch

It's always good to know things like who is captured in certain surveys, the difference between household and individual surveys, standard definitions of things like "unemployment" or "income", etc. - just little things to know your way around data.

For student readers/people just getting into empirical research - I've run into another common one recently that's worth noting too: industrial and occupational categories change over time, so keep an eye on that. You don't need to memorize all the different classifications, but just know that it comes up. I lost a third of the IT workforce this morning from January 2010 to January 2012. The recession was bad but not that bad. I had a suspicion and sure enough a new occupational code went into effect in the interim. Most things don't change in this sort of recoding but if something that you're looking at changes it can make a big difference.

A second case - in revising the engineering chapter about two weeks ago I was adding a table on the industry that different engineering fields work in in 1980 and 2010, to look at any changes. There was one big change for marine engineers from Manufacturing (I was using big industrial sectors and very detailed occupational sectors) to Transportation.

A major change in the work of marine engineers?

Nope.

Turns out a handful of very detailed industries completely switched sectors over that thirty years. Specifically, shipbuilding taking place in drydocks was moved from Manufacturing to Transportation, while shipbuilding not taking place in drydocks stayed in Manufacturing.

Not a big deal for the most part - but when you're looking at marine engineers and half them do shipbuilding in drydocks and half do shipbuilding not in drydocks, it can make for some big swings in employment!