A Pragmatist's Guide to Prediction in the Social Sciences

Making out-of-sample predictions is an under-utilised tool in the social sciences, often for the wrong reasons. Many social scientists confuse prediction with unnecessarily complicated methods, or narrowly predicting the future. This is unfortunate, because prediction understood as the simple process of evaluating a model outside of the sample used for estimation is a much more general, and disarmingly simple technique that brings a host of benefits to our empirical workflow. One needn't use complicated methods or be solely concerned with predicting the future to use prediction, nor is it necessary to resolve the centuries-old philosophical debate between prediction and explanation to appreciate its benefits. Prediction can and should be used as a simple complement to the rich methodological tradition in the social sciences, and is equally applicable across a vast multitude of modelling approaches, owing to its simplicity and intuitive nature. For all its simplicity, the value of prediction should not be underestimated. Prediction can address some of the most enduring sources of criticism plaguing the social sciences, like lack of external validity and the use of overly simplistic models to capture social life. In this paper, I illustrate these benefits with a host of empirical examples that merely skim the surface of the many and varied ways in which prediction can be applied, staking the claim that prediction is one of those illustrious `free lunches' that can greatly benefit the empirical social sciences.