Professor X grew up in the rural countryside and had a long-standing belief that country living was healthier than living in an urban setting. She decided that this might be a decent area of inquiry, and so developed a theory explaining why rural upbringings are especially wholesome compared to urban upbringings, and set out to conduct research to confirm her hypothesis. Professor X examined differences in crime exposure, safety, violence, etc., between rural and urban settings to test her theory. All her findings collected over a very long career in fact revealed evidence that there are many positives to growing up in rural as compared to urban settings. Should the question of where to raise wholesome children therefore be considered settled science?There are at least two reasons why the answer to this question is "no." First, Professor X relied on positive test strategies, that is, she only tested the benefits, but not the possible harms or costs of rural as compared to urban living. Second, Professor X focused on hypothesis confirmation to the neglect of falsification. For example, she only considered contexts where rural upbringing would perform better than urban upbringing (e.g., exposure to crime) without considering contexts where urban upbringing might perform better than rural upbringing (e.g., exposure to greater diversity)-that is, contexts that would be more likely to falsify her theory.The same problems emerge when one tests hypotheses about ideological differences using a constrained set of stimuli and with a goal to confirm rather than provide a strong test of a hypothesis. Someone might have the hypothesis, for example, that conservatives are more prejudiced than liberals, and tests this hypothesis by examining whether there are ideological differences in racial animus, only to confirm her hypothesis. Does this mean a tendency toward prejudice is hard baked into conservative thinking? Without testing the same hypothesis using a PROMOTING STRONG INFERENCES IN POLITICAL PSYCHOLOGY 3 wider variety of possible targets of prejudice with the explicit goal of possible hypothesis disconfirmation we cannot be certain. Strong inferences require testing hypotheses not only in the contexts most favorable for hypothesis confirmation, but in the contexts most favorable for hypothesis disconfirmation as well (Platt, 1964).The goals of this chapter are to first describe two common pitfalls of social psychological approaches to the study of ideological differences, specifically, a pre-occupation with explaining conservatives to the neglect of liberals and an over-reliance on positive test strategies. As guards against these potential pit falls, we recommend that researchers shift their orientation toward negative test strategies, something that can be facilitated by using a "grid" approach to hypothesis generation. The grid approach to hypothesis generation forces researchers to consider a set of competing explanations for liberal and conservative thoughts, feelings, and behavior, that vary in possible normative s...