Guided discretion sentencing and proportionality review are methods used by death penalty jurisdictions to control for arbitrariness and discrimination in sentencing. Findings from this study show that very few death penalty cases are presumptively proportionate. By pooling cases of similar levels of culpability using the Barnett Scale, the authors conclude that at least eight of the 107 death sentences are presumed to be comparatively excessive, and a much larger number are suspect. Using logistic regression procedures, evidence was found of discrimination based on the race of the victim in the prosecutorial decision to proceed to penalty trial and in sentencing. Evidence from the Supplemental Homicide Reports also suggests racial disparities based on the race of victim and offender. The Barnett scale is a sufficient measure of case seriousness in some respects; however, results from the scale should be interpreted cautiously. Although a number of caveats appear when using the Barnett scale, a scale created with the use of logistic regression provided findings more consistent with previous research. 10 tables, 9 notes, and 56 references