The study used a 2 x 2 factorial design in which one of the factors was whether or not the crime was reported and the other factor was the crime's seriousness. Students were randomly assigned to 1 of the 8 experimental groups (10 subjects per group). Fictitious Canadian newspaper accounts were used to convey the crime's nature and the witness's behavior. Crimes described included vagrancy, receiving stolen goods, rape, and murder. In accordance with predictions, the seriousness of the crime and whether it was reported differentially affected the judgments. Reporting a crime was seen as a more morally right action than failing to report it; this difference increased with the increasing severity of the crime. Nonreporters of serious crimes were perceived as the least moral individuals. In addition, data indicated that for certain nonserious crimes, reporters were less liked than nonreporters. Such findings may help to explain why observers, confronted by lack of clarity in ongoing events, often categorize the acts being committed as nonserious and fail to report them. Tabular data, footnotes, and 15 references are given. (Author abstract modified)