Mental Illness as a Sentencing Determinant: A Comparative Case Law Analysis Based on a Machine Learning Approach

• Published date: 01 July 2023
• DOI: http://doi.org/10.1177/00938548231170801
• Author: Mia A. Thomaidou,Colleen M. Berryessa
• Date: 01 July 2023
• Subject Matter: Articles

CRIMINAL JUSTICE AND BEHAVIOR, 2023, Vol. 50, No. 7, July 2023, 976 –995.

DOI: https://doi.org/10.1177/00938548231170801

Article reuse guidelines: sagepub.com/journals-permissions

© 2023 International Association for Correctional and Forensic Psychology

976

MENTAL ILLNESS AS A SENTENCING

DETERMINANT

A Comparative Case Law Analysis Based on a

Machine Learning Approach

MIA A. THOMAIDOU

COLLEEN M. BERRYESSA

Rutgers University, School of Criminal Justice

This study identifies factors that contribute to sentencing outcomes for criminally sentenced individuals experiencing mental

disorders, in two U.S. states with divergent sociopolitical ideologies. Recent case law (n = 130) from appellate courts in New

York and Kansas (from 2020 to 2021) was analyzed using regression and machine learning to predict sentence severity for

individuals experiencing mental disorders. Across both states, trauma-related and personality disorders led to the most severe

sentences, while paraphilia, addiction, and mood disorders had the lowest probability of imprisonment. Sentencing outcomes

in Kansas were significantly more severe as compared with New York. A classification analysis identified important patterns

of sentencing determinants that predicted which mental disorders were more likely to lead to incarceration. Findings and

implications are discussed in relation to punishment disparities as well as the potentials and pitfalls regarding the use of

machine learning approaches in criminal justice research and policy.

Keywords: criminal justice; sentencing; mental illness; political ideology; machine learning

INTRODUCTION

Diagnoses of mental disorders are often traditionally considered in courts as either aggra-

vating or mitigating factors to sentencing proceedings (Davidson & Rosky, 2015; Miley

et al., 2020; van Es et al., 2020). In the United States, differences in legislation and the

application of law between states often lead to differences in rates and lengths of

AUTHORS’ NOTE: The authors have no conflict to declare. The authors thank Yvan Fafchamps for his valu-

able consultation on statistical modeling and the machine learning analyses, and Esther van Ginneken for her

advisement on the first version of this paper. Supplemental materials: The data set and analysis scripts can be

accessed by contacting the corresponding author. Correspondence concerning this article should be addressed

to Mia A. Thomaidou, School of Criminal Justice, Rutgers University, 123 Washington Street, Newark, NJ

07102-3094, USA; e-mail: athina.mia@gmail.com.

1170801CJBXXX10.1177/00938548231170801Criminal Justice and BehaviorThomaidou, Berryessa / Mental Illness as a Sentencing Determinant

research-article2023

Thomaidou, Berryessa / MENTAL ILLNESS AS A SENTENCING DETERMINANT 977

incarceration for individuals experiencing mental disorders (Durante, 2021; “State-by-State

Data on Sentencing Rates,” 2019). Particularly, it has been suggested that sentencing legis-

lation and guidelines in more conservative states in terms of political ideology may pose

major disadvantages for those that have received psychiatric diagnoses (Alexander & Link,

2003; Durante, 2021).

Yet it remains unclear from existing literature whether different types of mental disorders

are, in practice, differentially judged during sentencing, or how this could differ in states

with divergent sociopolitical ideologies. The populations of many Western and Eastern

states, such as New York, are found to be majority liberal in political ideology, while many

Southern and Midwestern states, such as Kansas, are shown to be majority conservative. As

electability is a major concern for lawmakers and most state-level judges in the United

States, a state’s predominate political ideology, to some extent, may also influence its crim-

inal-legal system’s responses to offending (Corrigan & Watson, 2003; DeLuca & Yanos,

2016; Silver & Silver, 2017). Comparative differences could lead to inconsistent sentencing

outcomes, which may raise concerns regarding uniformity and equality in sentencing for

those experiencing mental disorders (Berryessa, 2018b).

In the current study, we examine disparities in sentencing outcomes between different

categories of mental disorders, in a majority liberal state (New York) and a majority conser-

vative state (Kansas), based on recent case law from criminal case proceedings involving

individuals experiencing mental disorders. Particularly, using machine learning, we aim to

address how mental disorders may impact sentencing in two different states, with the

broader objective of identifying particular legal and extralegal factors, such as the serious-

ness and severity of offenses, individual-level characteristics such as race and gender, or the

jurisdictions’ and courts’ sociopolitical contexts, which may influence sentencing outcomes

for those that experience mental disorders in states with differing sociopolitical ideologies.

MENTAL DISORDERS AND CRIMINAL SENTENCING

Mental illness refers to a wide range of disorders that affect cognition, emotion, and

behavior (Njoku, 2022). While many may occasionally experience transient concerns

related to their behavior, a mental disorder is defined by ongoing signs and symptoms

that may cause frequent distress to the individual and affect their long-term ability to

function (American Psychiatric Association, 2013). Such mental disorders can be

broadly categorized into personality disorders, neurodevelopmental and traumatic, intel-

lectual, mood, and psychotic disorders (American Psychiatric Association, 2013). There

is no uniform definition of what constitutes mild or less serious mental disorders versus

serious mental disorders (Miley et al., 2020). However, the National Institute of Mental

Health has commonly distinguished less serious versus serious mental disorders by the

degree of functional impairment that substantially interferes with a person’s major life

activities.

As they most often influence a wide spectrum of behaviors (Appelbaum et al., 2015;

Batts, 2009; Tsimploulis et al., 2018), many mental disorders have been correlated with

criminal justice contact and involvement (Burd et al., 2010; Fast & Conry, 2009; Fazel

et al., 2016). A broad range of symptoms and secondary disabilities in people experiencing

mental disorders have been linked to antisocial behavior, aggression, and deficits in reason-

ing or in individuals’ ability to foresee the consequences of their actions (Douds et al., 2013;