**Principal investigator:** **Benjamin Lauderdale** London School of Economics and Political Science Email: b.e.lauderdale@lse.ac.uk Homepage: http://benjaminlauderdale.net/ **Sample size:** 3600 **Field period:** 4/15/2011-7/13/2011 **Abstract** Some partisan disagreements about politically relevant facts have been ascribed to psychological processes that lead citizens to believe whichever facts would rationalize their political preferences. This survey experiment is designed to generate new partisan disagreements about political facts, without misinformation or direct partisan cueing, by providing the kind of factual information that requires rationalization to be made compatible with partisans' prior beliefs. It provides experimental evidence that rationalization can create partisan disagreements about political facts at the population-level. Revealing a political fact---whether a substantive fact or an actor's position---related to a second political fact only via its relevance to a political choice, can increase expressed partisan disagreement about that second political fact. **Hypotheses** Is it possible to induce partisan disagreement about political facts by forcing citizens to rationalize other political facts. **Experimental Manipulations** A 2x2 design, involving provision of information about President Obama's positions on two issues, and about two substantive facts related to the same issues. **Outcomes** Citizens' stated beliefs about President Obama's positions on two issues, and about two substantive facts related to the same issues. When Obama's positions are the manipulation, the substantive fact is the outcome, and vice versa. The quantity of interest is the difference in treatment effect magnitude for Democrats and Republicans. **Summary of Results** At least some respondents were sufficiently unsure about the facts such that they were responsive to treatment. Citizens' partisanship shaped their interpretations when they were presented with facts, not necessarily in terms of the direct implications of these informational treatments, but rather in how they reconciled them with their other beliefs. Democrats were more inclined to believe the best about Obama; Republicans were more inclined to believe the worst. Democrats tended towards profiles of responses in which their own position was supported by the facts, and that position matched Obama's position. Republicans tended towards profiles of responses in which their own position was supported by the facts, but that position did not match Obama's position. **References** "How to Generate Partisan Disagreement about Political Facts Without Misinformation", presented at the 2012 annual meeting of the Midwest Political Science Association, Chicago, IL, USA.