Computational power and digital data have created new opportunities to explore and understand the social world. A special synergy is possible when social scientists combine human attention to certain aspects of the problem with the power of algorithms to automate other aspects of the problem. We review selected exemplary applications where machine learning amplifies researcher coding, summarizes complex data, relaxes statistical assumptions, and targets researcher attention. We then seek to reduce perceived barriers to machine learning by summarizing several fundamental building blocks and their grounding in classical statistics. We present a few guiding principles and promising approaches where we see particular potential for machine learning to transform social science inquiry. We conclude that machine learning tools are accessible, worthy of attention, and ready to yield new discoveries.
One good paragraph: > Machine learning tools cannot solve these causal problems on their own. In general, prediction policy problems are only prediction problems under the assumption that the relevant causal effect is identified. Or better yet, the relevant causal effect may be known: we do not need an experiment to know that an umbrella will keep you dry if it rains. The more that we know about the causal effect, the more that the researcher can focus on the predictive side of the problem. In general, new tools for prediction are best deployed in tandem with careful attention to underlying causal assumptions.