In this commentary, the author argues that there is a considerable degree of similarity between research in the hard sciences and education and that this provides a useful lens for thinking about what constitutes "rigorous" and "scientific" education research. He suggests that the fundamental property of hard science research is its predictive power, a property that can equally be applied to large- and small-scale and quantitative and qualitative research in education. Although variables may differ and methods of collection may not be the same, researchers do their best to measure and/or control those variables that matter, and design experiments and subsequent tests to ensure that those that can neither be measured nor fully controlled are unlikely to change the results in significant ways. He concludes that although fields like physics or chemistry are mature sciences, the "cutting-edge" work in these fields is often "messy," as researchers struggle to determine which variables are important. He suggests that education research often resembles the patterns seen in cutting-edge research in the "hard" sciences, as researchers are struggling to identify variables that are important to the problem.