It has been suggested that jurors in criminal trials are less likely t
o convict when the penalty is more severe or the charge is more seriou
s. This was explained by Kerr (1975) in terms of a perceived increase
in the cost of a Type I error (convicting an innocent person) that res
ulted in a criterion shift in the amount of evidence jurors required t
o vote guilty. The previous research found only weak support for the p
rediction regarding severity but consistent support for the predicted
effect of seriousness. However, in the case materials used in these st
udies, more evidence was legally required to prove guilt on the more s
erious charges. This article presents studies in which the amount of e
vidence needed to prove guilt was equated for all charges. Under these
circumstances, there was no effect on verdicts of seriousness of char
ge or severity of penalty and no evidence of a criterion shift due to
either variable. There may still be reason to believe that these facto
rs affect real juries, but this belief is not supported by the systema
tic evidence from mock jury studies.