Kristian Lum, Chesa Boudin and Megan Price (2020). The impact of overbooking on a pre-trial risk assessment tool. FAT* '20: Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency. January 2020. Pages 482–491. https://doi.org/10.1145/3351095.3372846 ©ACM, Inc., 2020.
Kristian Lum, Chesa Boudin and Megan Price (2020). The impact of overbooking on a pre-trial risk assessment tool. FAT* ’20: Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency. January 2020. https://doi.org/10.1145/3351095.3372846 ©ACM, Inc., 2020.
Text in English
Para evaluar afirmaciones sobre la reducción de la violencia letal en Colombia
En marzo de 2007, el Grupo de Análisis de Datos de Derechos Humanos (HRDAG por sus siglas en inglés) publicó un estudio con el título de "Para Evaluar Afirmaciones Sobre la Reducción de la Violencia Letal en Colombia." Los autores de dicho estudio evaluaron aseveraciones que la violencia en Colombia disminuyó tras la desmovilización de los paramilitares. Demostraron que tales afirmaciones se basan tanto en una sobreinterpretación de datos no ajustados como en inferencias causales infundadas. Los autores concluyeron que se requieren múltip...
In December 2006, Human Rights Watch (HRW) issued a report documenting torture and unlawful killings committed by Bangladesh's Rapid Action Battalion (RAB), an elite anti-crime and anti-terrorism force. According to HRW, many of the deaths for which RAB is responsible resulted from summary executions or extreme physical abuse. The statistical analysis presented in the report, "Judge, Jury, and Death: Torture and Executions by Bangladesh's Elite Security Force," was conducted by Romesh Silva, a statistician who worked within the Benetech Human Rights Program at the time, and who now works with HRDAG.
While researching these incidents, HRW compiled ...
Some of the earliest large-scale human rights information projects happened in El Salvador. One was developed by Patrick Ball at the Salvadoran non-governmental Human Rights Commission, also known as Comision de Derechos Humanos de El Salvador (CDHES-ng). Between 1977 and 1990, more than 9,000 testimonies were taken in an effort to document the nature and scope of the bloody conflict between the army and the Farabundo Marti National Liberation Front (FMLN). Starting in 1991, Patrick worked with CDHES staff to organize the information in an early computer database. They linked reported human rights violations with the career structures of individual ...
HRDAG associate Miguel Cruz has an epiphany. All those data he’s drowning in? Each datapoint is a personal tragedy, a story both dark and urgent, and he’s privileged to have access.
At HRDAG, we worry about what we don't know. Specifically, we worry about how we can use statistical techniques to estimate homicides that are not observed by human rights groups. Based on what we've seen studying many conflicts over the last 25 years, what we don't know is often quite different from what we do know.
The technique we use most often to estimate what we don't know is called "multiple systems estimation." In this medium-technical post, I explain how to organize data and use three R packages to estimate unobserved events.
Click here for Computing Multiple Systems Estimation in R.
Principled Data Processing is a way to prove to someone, usually yourself, that what you did was right.
I got an email from my superheroic PhD adviser in June 2006: Would I be interested in relocating to Palo Alto for six months in order to work with Patrick Ball at the Human Rights Data Analysis Group? (She'd gotten a grant and would cover my stipend.) Since I'd spent the last several months in New Haven wrestling ineffectually with giant, brain-melting methodological problems, I said yes immediately.
The plan with my adviser was simple: I'd digitize the ancient, multiply-photocopied pages of data from the United Nations Truth Commission for El Salvador, combine them with two other datasets, match across all the records, and produce reliable ...
Multiple Systems Estimation
What is MSE?
What do you mean by statistical inference?
What is an overlap, and how do we know when lists overlap?
How does MSE find the total number of violations?
How was MSE originally developed?
How does the Benetech Human Rights Program use MSE?
1. What is MSE?
A: Multiple Systems Estimation, or MSE, is a family of techniques for statistical inference. MSE uses the overlaps between several incomplete lists of human rights violations to determine the total number of violations.
Return to Top
2. What do you mean by statistical inference?
A: ...
We are saddened by the passing of Steve Fienberg yesterday in Pittsburgh, at the age of 74. He is perhaps best known around the world for bringing statistics to science and public policy and was a beloved professor at Carnegie Mellon University. At HRDAG we are in awe of and grateful for the work Steve did formalizing multiple systems estimation. His work on that front blazed a trail and essentially enabled all of our most important analytical work at the intersection of human rights and statistical science.
If we are to reduce the amount of human violence in the world, the first task is to determine the scope of the violence, to know how much of ...
“Usually, the thing you’re trying to predict in a lot of these cases is something like rearrest,” Lum said. “So even if we are perfectly able to predict that, we’re still left with the problem that the human or systemic or institutional biases are generating biased arrests. And so, you still have to contextualize even your 100 percent accuracy with is the data really measuring what you think it’s measuring? Is the data itself generated by a fair process?”
HRDAG Director of Research Patrick Ball, in agreement with Lum, argued that it’s perhaps more practical to move it away from bias at the individual level and instead call it bias at the institutional or structural level. If a police department, for example, is convinced it needs to police one neighborhood more than another, it’s not as relevant if that officer is a racist individual, he said.
Reports of torture and disappearances in Syria are not new. But the Amnesty International report says the magnitude and severity of abuse has “increased drastically” since 2011. Citing the Human Rights Data Analysis Group, the report says “at least 17,723 people were killed in government custody between March 2011 and December 2015, an average of 300 deaths each month.”
In this NPR story, HRDAG’s Patrick Ball comments on first-of-its-kind projections.
