Jailbirds: A Machine-Learning Approach to Measuring Racial and Ethnic Disparities in Bail Setting Open Access

Abstract

In the United States, nearly half a million individuals are held in prisons for crimes which they have not yet been convicted for, simply due to an inability afford bail. While the overwhelming body of literature surrounding racial and ethnic bias in bail setting finds that minority defendants suffer harsher bail outcomes than non-minority defendants, the most recent literature suggests otherwise. This thesis analyzes data provided by the New York Division of Criminal Justice Services in order to (1) explain such inconsistencies in recent literature as well as (2) develop a better grasp of how race and ethnicity affect bail outcomes. To further investigate racial and ethnic disparities in the context of bail setting, this thesis introduces a novel machine-learning method: an unpooled alternative outcome model which predicts bail outcomes for defendants, had they been of another race or ethnicity. Estimates from the analysis show that, for both racial and ethnic minorities, the probability of having monetary bail assigned would be lower for minority defendants had they been non-minority. 

Table of Contents

Section 1: Introduction ... 1

Section 2: Literature Review ... 7

2.1 Racial and Ethnic Disparities in Bail . . . . . . . . . . . . . . 7

2.2 Pretrial Detention in New York ................. 12

2.3 Inconsistencies in Recent Literature . . . . . . . . . . . . . . . 15

Section 3: Data ... 17

3.1 Overview.............................. 17

3.2 Dependent Variables ....................... 19

3.3 Independent Variables ...................... 23

3.4 Concannon and Na’s Data .................... 25

Section 4: Methods and Results ... 29

4.1 Framework............................. 29

4.2 Measuring Racial and Ethnic Disparities in the Assignment of Monetary Bail........................... 31

4.3 Investigating Concannon and Na’s Results . . . . . . . . . . . 36

4.4 A Machine-Learning Approach to Measuring Racial and Ethnic Disparities in Bail Setting .................. 43

Section 5: Discussion ... 44

5.1 Key Findings ........................... 44

5.2 Research Limitations and Future Research . . . . . . . . . . . 47

Section 6: Conclusion ... 48 

About this Honors Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Emory College
Department	Quantitative Science
Degree	B.S.
Submission	Honors Thesis
Language	English
Research Field	Law Statistics
Keyword	Statistical analysis Digital humanities Bail
Committee Chair / Thesis Advisor	Glynn, Adam, Emory University
Committee Members	Burchfield, Emily, Emory University Cherribi, Sam, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Jailbirds: A Machine-Learning Approach to Measuring Racial and Ethnic Disparities in Bail Setting ()	2023-04-10 12:06:31 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions