This paper discusses the distinction between racial disparity and racial discrimination; summarizes what is known about racial bias in capital and non-capital cases; and points out the reasons that the serious limitations of the research methods in this field make it difficult to make valid or reliable inferences about whether bias is present. Racial disparities do exist in the criminal justice system: although blacks constitute less than 11 percent of the U.S. population, they make up nearly half of the national prison population. But disparities alone are not evidence of discrimination. Comparisons between groups in the rate at which they receive a penalty must first control for factors that can appropriately influence sentencing decisions. Only after such controls are instituted can there be an investigation of the unique role of race. However, even when controls are used, social science research procedures still fall short of providing a valid or legally sufficient assessment of whether race affects sentencing decisions. Statistical and research techniques are just not up to the task of providing definitive evidence of whether racial bias is present. Discrimination in capital or non-capital cases cannot be inferred simply because groups differ in the rate at which they receive a given penalty.