If you’ve downloaded enough data from the IPEDS Data Center using the “Compare Institutions” interface, you’ve probably realized that, depending on what you’re downloading, the data provided is rarely in a format ready for analysis. Here, via a specific example, I describe what makes the IPEDS data format impractical, and how to use R to resolve that.
Reading in the Data I first downloaded Fall 2012 to Fall 2018 distance education headcounts for every college and university in the IPEDS Data Center.
The Curse of Knowledge in Everyday Life Several years ago my friend Lauren asked me for my recipe for BBQ seitan. I love food-related conversation, so I wasted no time. “Start by sauteing some chopped onion in oil…”, and as quickly as I began, she cut me off. “Hold on,” she interjected. “What kind of oil do you use? How much? How high do you turn the heat?”
Dissecting the conversation, what happened was that I implicitly made the absurd assumption that knowledge that is in my head must be in hers (i.
Data: It’s become a cliche to say that it’s everywhere and in quantities that are unimaginable. But data in its raw form, whether in a structured database or on the internet, is of limited use until a human does something to it: Gather it, clean it, visualize it, model it, write about it, and so-on.
The amount of data that those in institutional research encounter requires powerful tools to work with.