Incorrect estimates and false reports: How framing modifies truth

摘要

Nearly all framing studies to date presuppose unbiased estimates. If an expert says that "programme A will save 200 people", it is tacitly assumed that this prediction is correct. In real life this is rarely the case. In the present study people were asked to evaluate such claims that eventually turned out to be incorrect. Participants in five experiments were asked to rate how correct, and how true, are predictions and reports that either overstate or understate the facts. Overall, understatements were considered more accurate than overstatements, and pessimistic statements were better than overly optimistic ones. Thus predictions of an outcome that turns out better than expected should preferably have been presented in a positive frame (e.g., money saved), whereas outcomes that turn out worse than expected will appear less "wrong" if predictions were presented in a negative frame (e.g., money lost). Parallel studies were performed with wrong predictions and incorrect factual claims (lies). In all studies, correctness ratings were affected by the way the statements were framed.