In the first study, subjects received accounts of sentencing hearings that appeared, during one week, in Toronto newspapers. These stories were generally short and provided little information about the offense, the offender, or the judicial reasoning underlying the sentence. Ratings of the stories indicated that subjects were generally confident about judgments of the reported sentences, the majority of which were rates as being too lenient. Subsequent experiments demonstrated the importance of the context in which sentences are placed. In the second study, accounts of the same sentencing hearing by different newspapers had variable effects upon subjects' ratings of the sentence. Finally, in the third study, comparisons of ratings by subjects who read a transcript summary with the reactions of others who read news media accounts of the same sentence revealed that those who read the news media account had more negative views of the judge, the offense, the offender, and the sentence. This effect was replicated with different offenses and other subject populations in Ontario. One uniform conclusion from all of these studies is that sentencing is a complex phenomenon in need of careful analysis. The mass media, in reporting sentences handed down in individual cases as if the issues involved were very simple, do not appear to present sentences in a manner that allows members of the public to draw reasonable conclusions about sentencing. 1 appendix, 3 tables, 3 footnotes, and 29 references (Author abstract modified)