Socioeconomic marginality in sentencing: The built-in bias in risk assessment tools and the reproduction of social inequality
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This article develops a sociological analysis and critique of including socioeconomic factors such as education, employment, income and housing in risk assessment tools that inform sentencing decisions. In widely used risk assessment tools such as the Level of Service Inventory-Revised (LSI-R) (Canada, US), the Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) (US), the Offender Assessment System (OASys) (UK) and the Recidive InschattingsSchalen (RISc) (the Netherlands), socioeconomic marginality contributes to a higher risk score, which increases the likelihood of a (longer) custodial sentence for underprivileged offenders compared to their more privileged counterparts. While this has been problematized in relation to gender and racial/ethnic bias, the problem of socioeconomic bias in itself has received little attention. Given the already marginalized position of many justice involved individuals and longstanding concerns about such disparities, and the adverse effects of imprisonment on socioeconomic opportunities, it is essential to evaluate the unintended social consequences of assessing socioeconomic marginality as ‘risk factor’. Elaborating on earlier critiques, I conceptualize risk-based sentencing as a meaning-making process through which (access to) resources and recognition are distributed among offender populations. Through tracing in detail two cultural processes – stigmatization and rationalization – I analyse how risk assessment is likely to produce sentencing disparities as well as to reproduce, and possibly exacerbate, social inequalities more generally.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it