In this dissertation, I examine the treatment and sentencing of American Indian defendants. This work contributes to research on cumulative disadvantage and the role race and social context play to influence federal sentencing outcomes. Disparities in federal sentencing for racial and ethnic minorities are an important concern to scholars and policy makers. Literature suggests that blacks and Latinos are sentenced more harshly than similarly situated white offenders. These findings are concerning because they suggest that minorities are treated unfairly by the criminal justice system, questions the legitimacy of how offenders are processed and treated, and defendants of color who are meted out tougher punishments face substantial social and economic difficulties thereafter. Although the black-white and Latino-white disparities have been identified and highlighted, less is known about whether disparities extend to other minority groups, and consequently little is known about the treatment of these neglected groups.

I investigate whether American Indian defendants experience cumulative disadvantages at multiple decision points, disadvantage over time, and the effect of social context on drawing on American Indian disadvantage, the focal concerns and minority threat perspectives. The focal concerns perspective is used to develop hypotheses about how American Indian defendants will receive harsher punishments at multiple decision points. I also use this perspective to predict that American Indian disadvantages will increase over time. Lastly, I examine social context and its effect on punishment decisions for American Indians using the minority threat perspective. I hypothesize that 
social context impacts how American Indian defendants are sentenced at the federal level.

Data come from the Federal Justice Statistics Program Data Series, the US Census, and the Uniform Crime Report, with a focus on data gathered from the Administrative Office of the United States Courts and the United States Sentencing Commission. A range of modeling strategies are used to test the hypotheses including multinomial logistic regression, ordinary least squares regression, and multilevel modeling.

The results suggest that cumulative disadvantages against American Indian defendants is pronounced, American Indian disparity over time is significant for certain outcomes, and social context plays a limited role in American Indian sentencing disadvantage.