How Social Science Research is Done

Arthur C. Brooks is the Louis A. Bantle Professor of Business and Government Policy at Syracuse University and the author of Who Really Cares. In the New York Times, he explains how he goes about doing research:

“When I started doing research on charity,” Mr. Brooks wrote, “I expected to find that political liberals — who, I believed, genuinely cared more about others than conservatives did — would turn out to be the most privately charitable people. So when my early findings led me to the opposite conclusion, I assumed I had made some sort of technical error. I re-ran analyses. I got new data. Nothing worked. In the end, I had no option but to change my views.”

In other words, the steps are (1) start with a liberal hypothesis, (2) find data, (3) if data doesn't agree with liberalism, try to change the data, and (4) accept that the liberal hypothesis is wrong only as a last resort.