Why would you complain about the author not answering a question that the article is not about? I'd like to quibble with the term prediction in this context to

Why would you complain about the author not answering a question that the article is not about? I'd like to quibble with the term prediction in this context too. I think it's too strong to say that social scientists use historical data to make predictions. They would use historical data to make inferences about the likely causes of events and use that insight to inform policy going forward. That is not prediction but gaining insight into probable outcomes if certain things are done in certain contexts. The closest one might come to prediction would be economic forecasting where one looks at various patterns in the data to make an, ahem, educated guesses about likely outcomes. This too is not exactly prediction because one is interested in the likelihood of a range of outcomes. Even those who do these things for a living admit that this is, at best, a guessing game. (As JK Galbraith once quipped, "The only function of economic forecasting is to make astrology look respectable")