A fundamental divide persists between Social Science, which prioritizes explanation and causal understanding, and Data Science, which emphasizes prediction and forecasting. Separately, each approach is limited – producing either explanations with uncertain predictive value, or predictions that lack interpretability and robustness under real-world policy change. What if the strengths of the explanatory and predictive approaches could be combined to enable research that is both scientifically rigorous and practically relevant for decision-makers?

Mission: To develop a framework combining the explanatory strengths of Social Science with the predictive capabilities of Data Science, improving how evidence is generated, evaluated, and used for theory development and decision-making. In this framework, explanation and prediction reinforce one another producing models that deepen our theoretical understanding of social processes and can forecast future outcomes. By integrating theory-driven explanation with data-driven prediction, Predict to Explain strengthens both scientific insight and real-world decision-making.