It was hypothesized that juror-defendant similarity would lead to grea
ter leniency toward a criminal defendant when the evidence against tha
t defendant was weak or inconclusive; but when evidence was strong, it
was expected that this relationship would be reversed. In Study 1, re
ligious similarity was found to be simply and positively related to ev
aluation of the defendant and leniency, a relationship unaffected by t
he strength of evidence. This pattern of results was attributed to (a)
insufficiently strong evidence against the defendant and (b) the lack
of anticipated jury deliberation, problems addressed in Study 2. In t
hat study, when evidence was strong against the defendant, juror-defen
dant racial similarity did increase the likelihood of conviction, but
only when jurors anticipated being in the racial minority in their jur
y. Implications of the findings for psychological theory and for voir
dire were discussed.