Reducing Political Bias in Political Science Estimates
by L. J. Zigerell, Illinois State University [@LJZigerell]
Political science researchers have flexibility in how to analyze data, how to report data, and whether to report on data. This article reviews examples of reporting flexibility from the race and sex discrimination literature to illustrate how research design choices can influence reported estimates and inferences. Biases in estimates caused by reporting flexibility might cancel out if randomly or evenly distributed across the field, but the overrepresentation of political liberals in the political science ranks creates the realistic potential that political science estimates are more biased toward the preferences of the political Left. These reporting biases can be reduced or eliminated through preregistration and preacceptance, with researchers committing to a research design before completing data collection. Journals and researchers taking such steps to remove potential reporting flexibility can raise the credibility of political science research.