The Federal Reserve Bank of New York has come up with an innovative empirical method to lead to the theoretically predicted conclusion that big cities do a better job of matching workers to appropriate jobs, since they have deeper and more specialized labor markets. But I suspect the study will be best-known for the finding Brad Plumer highlights, namely that only 27 percent of college graduates have a job related to their college major.
But is that really true?
According to the paper, they're measuring relatedness by using the National Center for Education Statistics' Classification of Instructional Programs "occupational crosswalk" function. So I thought I would look up my own major, philosophy, and see what the federal government has to say about my career choices. They think that the only job I'm suited for is as a post-secondary teacher, teaching philosophy or religion classes at a college. In other words, I am one of the 73 percent.
But I would dispute the claim that my job has nothing to do with my college degree. My view is that undergraduate philosophy majors get a crash course in persuasive writing and logical argumentation. Any kind of liberal arts degree where you need to read a lot of texts and then learn to write persuasively based on said texts is a decent preparation for working in journalism. But philosophy is a particularly good one to study, because for better or for worse, journalists are typically asked to be generalists. As a philosophy major you read and discuss people who are not only great thinkers but people who managed to make meaningful intellectual contributions to the world without obtaining tons of new empirical information. Particularly in my little subfield of Web-based punditry, that sort of thing is crucial. You of course don't learn shoe-leather reporting as a philosophy major, and that's an important part of the journalism ecosystem. But large swathes of journalism have relatively little to do with shoe-leather reporting, and philosophy is a damn fine thing to major in if you want the skills you need to succeed in those swathes.
My larger point here is that it's really important to pay attention to data quality. If existing labor markets do a poor job of matching college graduates to things the NCES CIP "occupational crosswalk" function says are major-appropriate jobs, is that a fact about the labor market or a fact about the statistical series?