Final Project Report EZM PDF

Preview

Summary

Final project...

Description

Running head: Race and Pretrial Bond Amount

Exposing Ties between Race and Pretrial Detention: Addressing bias during the pretrial detention process

Edward Maza with Madeline Colbert and Patrick Sullivan Yale University

Correspondence to: Edward Maza Yale University Email: [email protected]

Abstract This study examines the way race effects the length of time people are held in pretrial detention and their assigned bond amount in the state of Connecticut. We explored the connection of three different racial categories (Black, White, Hispanic) to the pretrial detention system. Our independent variable was race and our dependent variables were amount of time held in pretrial detention and the bond amount assigned. By running two one-way ANOVAS we did not find a significant effect of race on the amount of time held in pretrial detention, but we did find a significant effect of race on the amount of bond assigned. The mean bond amount for white people was the lowest, the amount for Hispanic people was higher, and the amount for black people was the highest.

Introduction

Despite the alleged presumption of innocence until proof of guilt, every year the Connecticut bail system forces the detention of thousands of people before they go to trial. Pretrial detention can have devastating impacts on the lives of those affected, keeping the detained person away from their job and family. According to a recent study it is 13% more likely for a person to be convicted of a crime if they were held in pretrial detention, as such defendants are more likely to plead guilty (Stevenson, 2016). There is a strong demonstrated relationship between blackness and perceived criminality which even further effects the way black defendants are viewed by a jury (Eberhardt, Goff, Purdie, & Davies, 2004). Additionally, an article from the William and Mary Law Review demonstrated that “significant gaps emerge in the severity of pretrial release conditions that disadvantage black and Hispanic defendants”(Griffin, Sloan, & Eldred, 2014). A 2010 study done in Pennsylvania found that black defendants were less likely to be released without bail than white defendants (Freiburger, Marcum, & Pierce, 2010). However, the study did not conclude that there was a significant relationship between race and the amount the bail was set for. Additionally a 2015 study found that there was a substantial disparity in wealth between white households and black and Latino households (Amy Traub & colleagues, 2015). Our study explores the impact of race on the amount of bail set for pretrial detention in the state of Connecticut and how long those detained spend in jail. This study will only focus on black, white, and Hispanic detainees because the population sizes of the other two racial categories reported in the data set are too small and beyond the scope of research of this study. Demonstrating whether or not race is a factor in determining bond amount and days detained would reveal bias in the detention process that unfairly impacts people of color and would contribute to the pertinent dialogue surrounding issues of race and the criminal justice system,

especially with impending legislation on the bail system in Connecticut. We hypothesize that race, when controlling for other factors such as level of crime, gender, and age, effects the number of days someone is detained before trial. We also hypothesize that race, when controlling for other factors such as level of crime, gender, and age, affects the bond amount assigned to a defendant.

Method This study is based on data released by the Connecticut Department of Corrections, collected from the Department of Justice. Each day since July 1, 2016 the DOC updates the publically available data on pretrial detention to show each detained person’s bond amount, gender, race, and location of incarceration. All of this data is made anonymous using randomly generated identifiers to differentiate between detainees. We will be using a data sample from July 1, 2016 to April 13, 2016. The data was downloaded from the DOC through TrendCT.org. We prepared our data for testing by using R to create multiple categories: felonies and misdemeanors, violent and nonviolent, and general groupings of crimes such as robbery, assault, drug crime, etc. We assigned these categories because these factors commonly affect the amount of bail assigned. By creating multiple categories we were able to control for gender, level of crime, type of crime, and age. In our study, race is our independent variable. Our dependent variables are bond amount assigned and the length of stay in the correctional facility.1 Results The Department of Corrections in Connecticut collected data on how long black, white, and Hispanic detainees were held in pretrial detention. A one-way ANOVA using race as a between subjects factor showed that there was no significant effect of Race on the number of days people were held in pretrial detention controlling for gender, age, bond amount, level of crime, the violence of the crime, and the type of crime. F(4, 9750)=1.82, p=.122. Because there was not a significant interaction no post hoc follow up tests were necessary. 1

Our team spoke to Professor Gee who told us we should only report the significance of our IV and DV and that we did not need to report the significance of the variable we controlled for. She also asked that we make a note of our conversation with her.

The Department of Corrections in Connecticut collected data on the amount of bond assigned to black, white, and Hispanic individuals who were held in pretrial detention. A one-way ANOVA was conducted on the amount of bond assigned using race as the between subjects variable. Overall, the ANOVA accounted for 12.9% of the variance of assigned bond amount. The oneway ANOVA showed a significant effect of race on the amount of bail assigned to pretrial detainees, controlling for gender, age, level of crime, the violence of the crime, and the type of crime. F(4,9751)=23.23, p...