This study examined people's tendency to draw prescriptive, moral inferences from the results of empirical research -an inferential error referred to as the "naturalistic fallacy. " Students read descriptive summaries of social science research articles and were tested for comprehension, attitudes on the issues addressed in the research, and endorsement of statements containing naturalistic fallacy errors. Students were retested 7 weeks after their initial reading of the research. Cross-lagged regression and analysis-of-variance results indicated that the fallacious inference statements were frequently judged as true, that holding attitudes consistent with the direction of the studies' findings increased the likelihood of committing the fallacy, and that the impact of attitudes on commission of these errors increased as the time between reading the research and drawing the inferences increased.