미연방양형지침서 하의 양형격차 : 강도죄의 양형에 대한 경험적 연구

Sentencing Disparity under The U.S. Federal Sentencing Guidelines : An Empirical Study of Sentencing Outcomes for a Robbery

This research studies the effect of extralegal factors on sentencing outcomes for defendants convicted of a robbery under the U.S. Federal Sentencing Guidelines in 1993-1994. The U.S. Federal Sentencing Guidelines direct and constrain judicial discretion in sentencing by prescribing legally relevant factors. However, the Sentencing Guidelines do not restrict plea bargaining practice and allow judges to depart from the guidelines in defined situations. This research hypothesizes that sentencing disparity would persist even under the Guidelines. First, defendant characteristics will directly influence the length of imprisonment. Second, guilty plea and departure will affect the length of imprisonment. Third, defendant characteristics will have effects on guilty plea and departure. Two stages of analysis are conducted to test these hypotheses. For the first stage, OLS regression is used for analysis of the direct effects of independent variables on the length of imprisonment. For the second stage of analysis, logistic regression is employed to examine the effects of independent variables on the processing variables - guilty plea and departure. This research finds that 1) defendants who are black, male, younger, and with lower income receive longer prion term, 2) pleading guilty and departure result in shorter length of imprisonment, 3) black defendants and older defendants are less likely to pleading guilty, and female defendants and the offenders who have many dependents are more likely to receive the sentences departed from the Guidelines. The findings indicate that there are direct as well as indirect effects of defendant characteristics on the length of imprisonment in robbery cases under the U.S. Federal Sentencing Guidelines in 1993-1994.