Public policy, it is widely believed, is affected, sometimes strongly, by public opinion, lobbying, protest, the party balance, macro-level characteristics of political units (including demographic composition, economic development, etc.), and other factors. But there are controversies about the impact of all of these. This paper is a research synthesis on the determinants of public policy that attempts to ascertain how strongly policy is affected by a variety of factors. It goes beyond past work in five ways: it considers a wider range of hypothetical determinants of policy; hypothesizes that the more information the hypothetical determinants provide to policymakers, the more likely they are to influence policy; hypothesizes that the impact of variables on policy will be affected by how policy itself is conceptualized and measured; considers not only whether the relationship between independent variables and policy is statistically significant, but the magnitude of the impact as well; and describes how much attention researchers pay to the variables most likely to affect policy. There is some support for the hypotheses; the work reviewed has almost nothing to say about how strongly independent variables influence policy; and the variables having the greatest impact on policy are very seldom the subject of research.