Evidence-Based Sentencing and the Scientific Rationalization of Discrimination

By Sonja B. Starr. 2013

261 citations recorded by the source corpus.

Subfields: AI Ethics, Fairness & Society, Social Sciences & Humanities

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In AI Ethics, Fairness & Society

In Social Sciences & Humanities

Citations in the corpus

Interventions over Predictions: Reframing the Ethical Debate for Actuarial Risk Assessment (2017, positive): “These mechanisms correspond with different stages in the model design process such as identifying the research question, collecting training data, labeling examples within the training set, feature selection and doubleencoding bias in proxies for protected classes (Barocas and Selbst, 2016; Gong, 20”
Fairness in Criminal Justice Risk Assessments: The State of the Art (2017, background): “…and Stevenson 2016), government agencies (National Science and Technology Council 2016:30-32), foundations (Pew Center of the States 2011), and academics (Berk 2008; Berk and Hyatt 2015; Demuth 2003; Hamilton 2016; Harcourt 2007; Hyatt, Chanenson, and Bergstrom 2011; Starr 2014b; Tonry 2014).”
Algorithmic Decision Making and the Cost of Fairness (2017, background): “We note that despite such aggregate benefits, some have argued that statistical tools do not provide sufficiently precise estimates of individual recidivism risk to ethically justify their use [36].”
"Wrong side of the tracks": Big Data and Protected Categories (2014, methodology): “[16], which seeks to exclude some variables as inputs altogether.”
The false promise of risk assessments: epistemic reform and the limits of fairness (2020, background): “endorsements of evidence-based sentencing are similarly grounded in the goal of reducing incarceration [115, 148].”
A Primer on Fairness in Criminal Justice Risk Assessments (2016, background): “…Department of Criminology Department of Statistics University of Pennsylvania Introduction There are widespread concerns about fairness when actuarial risk assessments are used to inform criminal justice decisions (Harcourt, 2008; Tonrey, 2014; Starr, 2014; Berk and Hyatt, 2015, Crawford, 2016).”
RISK, RACE, AND RECIDIVISM: PREDICTIVE BIAS AND DISPARATE IMPACT*: RISK, RACE, AND RECIDIVISM (2016, background): “According to Starr (2014, 2015), it is legitimate to consider an offender’s criminal history in determining a sentence, but risk assessment instruments also include such “socioeconomic” variables as marital history, employment/education, and financial background.”
Rethinking algorithmic regulation (2015, background): “As the legal scholar Sonja B. Starr (2014) has shown, data-driven analytics are now entering the world of criminal sentencing.”
The Principles and Limits of Algorithm-in-the-Loop Decision Making (2019, background): “For although calls to adopt machine learning models often focus on the accuracy of these tools [14, 46, 59, 66], accuracy is not only attribute of ethical and responsible decision making.”
An impact assessment of machine learning risk forecasts on parole board decisions and recidivism (2017, background): “The most salient concerns are typically ethical (Harcourt 2008) and jurisprudential (Starr 2014).”
Algorithms in practice: Comparing web journalism and criminal justice (2017, background): “For example, the American Law Institute recommended a broader use of risk-assessment tools in their highly influential Model Penal Code (Starr, 2014).”
Risk Assessment in Criminal Sentencing. (2016, background): “Opponents of risk assessment in sentencing assert that " criminal history has become a proxy for race " (Harcourt 2015)—as have a number of other risk factors (e.g., employment status, education, and neighborhood are proxies for race and poverty; Starr 2014).”
Machine Learning Risk Assessments in Criminal Justice Settings (2018, background): “Current practice is far more sophisticated and nuanced, but some can still reasonably ask whether actuarial justice is a good thing (Feeley and Simon 1994; Buruma 2004; Robert 2005; Monahan 2006; Harcourt 2007; Boonin 2011; Tonry 2014; Starr 2014; Ferguson 2017; Huq 2019).”
Algorithmic Risk Assessment in the Hands of Humans (2019, unknown): “Algorithms don’t include race as a predictor, but they often include direct socio-economic markers like employment, housing or education status, which are correlated with race (Starr, 2014).”