Partially Blind
Does “Jamal” Receive a Harsher Sentence Than “James”? First-Name Bias in the Criminal Sentencing of Black Men
Dushiyanthini (Toni) Kenthirarajah et al.
Law and Human Behavior, February 2023, Pages 169-181
Method: In an archival study, we obtained a random sample of 296 real-world records of Black male prison inmates in Florida and asked participants to rate the extent to which each inmate’s first name was stereotypically Black or stereotypically White. We then tested the extent to which racial stereotypicality was associated with sentence length, controlling for relevant legal features of each case (e.g., criminal record, severity of convicted offenses). In a follow-up experiment, participant judges assigned sentences in cases in which the Black male defendant was randomly assigned a more stereotypically Black or White name from our archival study.
Results: Controlling for a wide array of factors -- including criminal record -- we found that inmates with more stereotypically Black versus White first names received longer sentences β = 0.09, 95% confidence interval (95% CI) [0.01, 0.16]: 409 days longer for names 1 standard deviation above versus below the mean on racial stereotypicality. In our experiment, participant judges recommended significantly longer sentences to Black inmates with more stereotypically Black names above and beyond the severity of the charges or their criminal history, β = 0.07, 95% CI [0.02, 0.13].
Judicial Scarring
Karthik Srinivasan
University of Chicago Working Paper, February 2023
Abstract:
I document that experienced decision makers can be influenced by irrelevant events in a high stakes setting, felony sentencing in Cook County. Using a stacked difference-in-differences design, I estimate that judges hand down sentences that are 13% longer after sentencing a first degree murder. The effect is twice as large for defendants who resemble the murderer along the dimensions of race and charge severity. The bias affects 6% of defendants on an ongoing basis and temporarily increases the Black sentencing penalty by 91%.
And Nothing but the Truth: An Exploration of Perjury
Stephanie Crank & Drew Curtis
Journal of Police and Criminal Psychology, March 2023, Pages 83–92
Abstract:
Perjury is a deception that occurs within the legal system. Research indicates that legal personnel, including police officers, detectives, and secret service agents, detect deception slightly above chance levels (Bond and DePaulo 2006). Scant research has been conducted to examine the frequencies and motivations of perjury. Thus, we conducted two studies to explore the rates of perjury and incentives to engage in perjury. Study one examined the reported frequencies of perjury, finding that it occurs in 29.9% of the interactions with legal personnel. Study two was designed to examine perjury-like behaviors using the Ariely (2012) shredder paradigm. Over half (54%) of participants in study 2 engaged in perjury-like behaviors and those who received money as an incentive were equally likely to engage in perjury behaviors as those who did not. Implications for practitioners and researchers are discussed.
The Eye of the Beholder: Increased Likelihood of Prison Sentences for People Perceived to Have Hispanic Ethnicity
Erik Girvan & Heather Marek
Law and Human Behavior, February 2023, Pages 182–200
Method: We analyzed official state records of more than 220,000 unique sentencing decisions for nearly 200,000 individuals under state correctional supervision between 2005 and 2018, including demographic characteristics, statutory crime-seriousness and criminal-history scores from state sentencing guidelines, and sentencing outcomes.
Results: Even after controlling for crime severity and criminal history, we found that individuals who were labeled as Hispanic in criminal justice records were nearly twice as likely to be sentenced to prison as those who were labeled as White (odds ratio [OR] = 1.95, 95% confidence interval [CI] [1.86, 2.04]). By comparison, individuals who were labeled in criminal justice records as White but, on the basis of validated estimates, were predicted to self-identify as Hispanic had the same likelihood of being sentenced to prison as individuals who were accurately perceived to be White (OR = 1.01, 95% CI [0.94, 1.07]).
Machine Learning as a Tool for Hypothesis Generation
Jens Ludwig & Sendhil Mullainathan
NBER Working Paper, March 2023
Abstract:
While hypothesis testing is a highly formalized activity, hypothesis generation remains largely informal. We propose a systematic procedure to generate novel hypotheses about human behavior, which uses the capacity of machine learning algorithms to notice patterns people might not. We illustrate the procedure with a concrete application: judge decisions about who to jail. We begin with a striking fact: The defendant’s face alone matters greatly for the judge’s jailing decision. In fact, an algorithm given only the pixels in the defendant’s mugshot accounts for up to half of the predictable variation. We develop a procedure that allows human subjects to interact with this black-box algorithm to produce hypotheses about what in the face influences judge decisions. The procedure generates hypotheses that are both interpretable and novel: They are not explained by demographics (e.g. race) or existing psychology research; nor are they already known (even if tacitly) to people or even experts. Though these results are specific, our procedure is general. It provides a way to produce novel, interpretable hypotheses from any high-dimensional dataset (e.g. cell phones, satellites, online behavior, news headlines, corporate filings, and high-frequency time series). A central tenet of our paper is that hypothesis generation is in and of itself a valuable activity, and hope this encourages future work in this largely “pre-scientific” stage of science.
The Gender Gap in Supreme Court Legitimacy
Christopher Krewson & Jean Reith Schroedel
American Politics Research, forthcoming
Abstract:
Men and women diverge in their political behavior and attitudes. We test whether gender-based variation in political attitudes extends to perceptions of US Supreme Court legitimacy. Using a dataset covering the years 2012–2017, we show that one’s identification as a man or a woman predicts their diffuse support for the Court. In particular, women almost always extend less legitimacy to the Court than men do. This is true within both Republican and Democratic identifiers, and regression analysis shows the gender gap holds when controlling for partisanship, ideology, race, age, education, income, and Supreme Court approval. Additionally, we included a series of questions in a 2021 Cooperative Election Study (CES) module to explore why the gender gap in perceived legitimacy exists. We find that differences in perceptions of the Court’s representation of women and its fairness drive the gender gap in legitimacy.
The Stigma of Wrongful Conviction Differs for White and Black Exonerees
Lakia Faison et al.
Law and Human Behavior, February 2023, Pages 137–152
Method: In Experiment 1, we unobtrusively measured non-Black participants’ behavioral reactions to an anticipated meeting with a Black or White exoneree or businessman. In Experiment 2, participants completed measures that assessed their motivation to appear unprejudiced and then, in a separate session, evaluated a Black or White exoneree and reported their beliefs about the legal system bias faced by the exoneree. Experiment 3 was a partial replication of Experiment 2. In Experiments 2 and 3, we examined data from both non-Black and Black participants.
Results: Non-Black participants in Experiment 1 stigmatized the White exoneree, d = −0.31, 95% confidence interval (CI) [−0.72, 0.10], but not the Black exoneree, d = 0.44, 95% CI [0.04, 0.83]. Experiments 2 and 3 replicated this finding, showing that the effect was mediated by the belief that Black exonerees faced greater legal system bias than White exonerees (Experiment 2: B = 0.21, SE = 0.06, 95% CI [0.11, 0.33]; Experiment 3: B = 0.35, SE = 0.09, 95% CI [0.19, 0.55]). Our results also suggested that Black individuals react more favorably to Black than White exonerees, potentially because of their beliefs regarding legal system bias.
Understanding racial disparities in pretrial detention recommendations to shape policy reform
Jennifer Skeem, Lina Montoya & Christopher Lowenkamp
Criminology & Public Policy, forthcoming
Abstract:
Federal pretrial services and probation officers assess defendants and make influential recommendations that defendants be either released or detained, based on their threat to community safety and risk of flight. To inform efforts to reduce disparities in pretrial detention, we examined officers’ decision making about 149,815 defendants across 81 districts. Overall, the probability of a detention recommendation was 34% higher for Black than White defendants. Racial disparities were most pronounced in ambiguous cases that invoked substantial officer discretion -- including cases where the defendant had little or no criminal record. Nevertheless, mediation analyses revealed that up to 79% of the racial disparity in detention recommendations operates through institutionalized factors (i.e., pretrial policy) rather than personally mediated factors (e.g., implicit racism or classism). The lion's share of the disparity operates through one institutionalized factor alone: criminal history.
Asking Versus Telling: The Supreme Court’s Strategic Use of Questions and Statements During Oral Arguments
Maron Sorenson
Political Research Quarterly, forthcoming
Abstract:
Supreme Court oral arguments are often characterized as the Court rapidly firing questions at attorneys who struggle to keep up; however, nearly half of the Court’s utterances come not as questions but as statements. I ask whether patterns of questioning and commenting behavior during oral arguments can predict case outcomes and justice votes. To answer this question, I develop a theory of strategic communication that accounts for the differential ways justices -- and other strategic actors -- use queries and comments during arguments. Using transcripts from 1981 to 2019, I code for use of questions and statements, finding the two theoretically and empirically distinct: where questions increase a party’s chances of winning, statements increase their chance of losing.
Televised Oral Arguments and Judicial Legitimacy: An Initial Assessment
Ryan Black et al.
Political Behavior, forthcoming
Abstract:
What happens to the perceived legitimacy of appellate courts when they allow cameras into their courtrooms? We implemented two experiments that exposed people to real video clips from two courts. In the first experiment we varied the modality (video or audio), contentiousness (neutral or contentious), and camera angle (static or dynamic) of exchanges between an attorney and judge and then measured people’s views toward judicial legitimacy. We found that static angles do not appear to influence legitimacy but using dynamic angles might have a limited effect. Watching a neutral exchange might increase judicial legitimacy -- compared to listening to that exchange -- but watching a contentious exchange might decrease it. In a second experiment we examined whether the presence of judicial symbols interacts with these effects. Evidence here is suggestive that these symbols could mitigate the negative effect of exposure to contentious content. Our results, though initial and limited in a number of ways, underscore both the complicated nature of cameras in the courtroom as well as the strong need for additional studies on a topic of great importance.
Testing the Forensic Confirmation Bias: How Jailhouse Informants Violate Evidentiary Independence
Baylee Jenkins et al.
Journal of Police and Criminal Psychology, March 2023, Pages 93–104
Abstract:
Research has demonstrated that primary confessions corrupt perceptions of forensic evidence, such as handwriting evidence. Additionally, research on secondary confessions indicates that statements made by jailhouse informants influence juror decision making to the same degree as primary confessions. The goal of the current study was to investigate whether jailhouse informant statements bias perceptions of forensic evidence. Participants were presented with a brief case summary about a bank robbery along with confession evidence from a jailhouse informant, in which both reliability and incentive presence were manipulated. Participants were then asked to examine a pair of either matching or mismatching handwriting samples before making case-relevant judgments. Results indicated that participants exposed to the reliable jailhouse informant were more likely to believe the samples were matching as well as rate them higher in similarity. These findings suggest that participants fell prey to the forensic confirmation bias.
The equity value implications of court ideology: Evidence from federal judge turnover
Stefano Cassella & Emanuele Rizzo
Journal of Corporate Finance, April 2023
Abstract:
We use exogenous variation in federal courts' composition, for instance due to a judge's death, as a quasi-natural experiment to study the equity value implications of court ideology. This design allows us to establish that firms experience an equity value loss when federal court ideology shifts in favor of plaintiffs and against corporations. The value loss stems from: (i) an increase in the expected costs of litigation; (ii) the deterioration of market-based governance mechanisms; (iii) and a perverse managerial incentive to decrease corporate disclosure in a heightened litigation environment.