Horowitz, I. A., & Seguin, D. G. (1986). The effects of bifurcation and death qualification on assignment of penalty in capital crimes. Journal of Applied Social Psychology, 16, 165-185.
A total of 708 subjects were used as jurors, comprising 59, 12-person juries. Subjects were recruited from eligible jurors in Toledo, Ohio, and Jamestown, N.Y. The Toledo, Ohio, juries participated in the moot courtroom of the University of Toledo Law School. The New York State juries sat in a college classroom designed as a courtroom. Trial situations were presented through audiotapes. Forty-four 12-person juries assigned to 1 of 5 trial conditions returned verdicts indicating that death-qualified, bifurcated juries rendered the most severe verdicts and the highest proportion of guilty verdicts. Nondeath-qualified, trial-only juries returned the least severe verdicts and the highest proportion of not-guilty or hung outcomes. Thirty-seven penalty-phase juries, of which 20 had previously assigned guilt, were distributed among 5 penalty conditions. The most severe sentences were given by the death-qualified, bifurcated juries. The least severe sentences were given by the nondeath-qualified, bifurcated juries. Juries impaneled for the penalty phase only and those death-qualified meted out moderate sentences.
Bifurcated juries tended to convict more than juries empaneled for trial only; death-qualified juries convicted more and opted for death penalty more than non-death-qualified juries. Death-qualified, bifurcated juries had highest conviction rate and most death sentences.