It was predicted that joinder would increase the likelihood of conviction, but that the magnitude of the effect would depend on the similarity of the offenses charged, the similarity of the evidence in the offenses, and judge's instructions designed to reduce the prejudicial effects. The 732 subjects viewed videotaped trial reenactments in which the variables under study were manipulated. Results indicate that joining multiple charges increases the proportion of convictions on a target charge relative to the same charge tried by itself. Convictions increased regardless of offense similarity or evidence similarity, but showed a marginally significant tendency toward more guilty verdicts when the evidence was dissimilar. A strong set of judge's instructions had no effect on verdict judgments. Psychological, theoretical, and legal implications of these findings are discussed. Thirty-nine references are listed.