232 results for search: sijihomi.blogspot.com/feed/rss2/publications


Report on Measures of Fairness in NYC Risk Assessment Tool

The report tries to answer the question of whether a particular risk assessment model reinforces racial inequalities in the criminal justice system.

Syria: No word on four abducted activists

Razan Zatouneh is an esteemed colleague of ours, and we are one of 57 organizations demanding immediate release for her and the three other human rights defenders still missing. A year on, no information on Douma Four The prominent Syrian human rights defenders Razan Zaitouneh, Samira Khalil, Wa’el Hamada and Nazem Hamadi – the Douma Four—remain missing a year after their abduction, 57 organizations said today. The four were abducted in Duma, a city near Damascus under the control of armed opposition groups. They should be released immediately, the groups said. On 9 December 2013, at about 10:40 pm, a group of armed men stormed into the ...

Reality and Risk in Our Mortality Study of the Peruvian TRC

HRDAG researchers and analysts at Peru's Truth and Reconciliation Commission (TRC) estimated conflict mortality due to violence using Capture-Recapture methods.

Casanare, Colombia

Text in English [popup citation="Tamy Guberek, Daniel Guzmán, Megan Price, Kristian Lum and Patrick Ball. (2010). Benetech/Human Rights Data Analysis Group database of lethal violence in Casanare."] Estimaciones de Homicidios y Desapariciones en Casanare Casanare es un departamento extenso y rural de Colombia con 19 municipios y una población de casi 300.000 habitantes. Ubicado entre las faldas de los Andes y las planicies orientales, Casanare tiene una larga historia de violencia. Diversos grupos armados han hecho presencia en Casanare, entre ellos paramilitares, guerrillas y el ejército colombiano. Muchos habitantes del Casanare han sido vícti...

Reflections: Minding the Gap

How might we learn what we don’t know? HRDAG associate Christine Grillo hits the wayback machine and recalls her first exposure to People Against Bad Things, ideas about bias and correlation versus causation, and truth.

Protecting the Privacy of Whistle-Blowers: The Staten Island Files

HRDAG built a machine-learning tool to strip the raw data of any potentially identifying information such as names and court case numbers. There was no "acceptable error rate."

Learning a Modular, Auditable and Reproducible Workflow

The modular nature of the workflow and use of Git allowed us to work on different parts of the project from across the country.

Why raw data doesn’t support analysis of violence

This morning I got a query from a journalist asking for our data from the report we published yesterday. The journalist was hoping to create an interactive infographic to track the number of deaths in the Syrian conflict over time. Our data would not support an analysis like the one proposed, so I wrote this reply. We can't send you these data because they would be misleading—seriously misleading—for the purpose you describe. Here's why: What we have is a list of documented deaths, in essence, a highly non-random sample, though a very big one. We like bigger samples because we think that they must be closer to true. The mathematical justif...

Welcome!

As of today, the Human Rights Data Analysis Group (HRDAG) is an independent* non-profit! It's been a long time coming, and we're delighted to have gotten to this point. HRDAG is a non-profit, non-partisan organization that applies rigorous science to the analysis of human rights violations around the world; for more information, see our About Us page. Benetech has spun out the scientific and statistical part of the Human Rights Program to HRDAG. The spinout includes (as staff) me -- Patrick Ball -- and Dr Megan Price, as well as our many part-time scientific and field consultants (a list is here). The software and technology component of our work -- ...

How Review of Police Data Verified Neglect of Missing Black Women

Sloppy recordkeeping by Chicago police has compromised missing persons cases. HRDAG is working with Pulitzer Prize-winning Invisible Institute to shed light on these stories.

Multiple Systems Estimation: Collection, Cleaning and Canonicalization of Data

<< Previous post: MSE: The Basics Q3. What are the steps in an MSE analysis? Q4. What does data collection look like in the human rights context? What kind of data do you collect? Q5. [In depth] Do you include unnamed or anonymous victims in the matching process? Q6. What do you mean by "cleaning" and "canonicalization?" Q7. [In depth] What are some of the challenges of canonicalization? (more…)

In Pursuit of Excellent Data Processing

With help from HRDAG, Roman Rivera built the data backbone for the Invisible Institute's Citizens Police Data Project.

Reflections: Some Stories Shape You

The first time I met anyone at HRDAG, I was a journalist. It was 2006. I was working on a story about a graduate student at Carnegie Mellon who’d collaborated with the organization on a survey in Sierra Leone, and I contacted Patrick Ball to discuss the work. At the time, I found him challenging. But I thought his work—trying to estimate how many people were killed, or, in that study, otherwise injured, during wars—was fascinating. Over the next few years, I got to know other researchers working on similar questions. In 2008, as the war in Iraq ramped up, I spoke with epidemiologists from Johns Hopkins University, the World Health Organiz...

Perú

In 2001, President Alejandro Toledo, called for the establishment of the Comision de la Verdad y Reconciliacion (CVR) (Truth and Reconciliation Commission) to investigate human rights abuses in Perú between 1980 and 2000. Dr. Patrick Ball was invited to work with the Commission to evaluate the CVR technical work and make recommendations on the information management process and analytic strategies. HRDAG consultant Jana Asher worked with Dr. Ball and CVR staff members David Sulmont (Director Informations Systems) and Daniel Manrique (Database Expert) to present evidence of the violations in a report to the CVR. The work included new estimates of ...

Human Rights and the Decentralized Web

Our partners were eager to learn and talk about emerging decentralized technology.

Multiple Systems Estimation: The Basics

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. In this blogpost, and four more to follow, I’ll answer both conceptual and practical questions about this important method. (In posts to follow, questions that refer to specific statistical procedures or debates will be marked, "In depth.") (more…)

Scraping for Pattern: Protecting Immigrant Rights in Washington State

With HRDAG's help, the University of Washington Center for Human Rights team has been able to analyze the scraped text and search for key words such as “jail” in order to gain insight into where immigration arrests are being made.

Sri Lanka

Ten years after the war ended in Sri Lanka, we still don’t know to the nearest ten thousand how many people perished. The estimates for the death toll for the last five months of the war alone vary between 7000 and 147,000. In 2011, the UN said it thought approximately 40,000 civilians had died; then in 2012 an internal UN report estimated it was at least 70,000. Population data from World Bank and UN sources indicated that more than 100,000 Tamils living in the conflict areas in the north have not returned home after the war. HRDAG has provided technical assistance to a broad range of non-governmental human rights organizations in Sri ...

Counting the Dead in Sri Lanka

ITJP and HRDAG are urging groups inside and outside Sri Lanka to share existing casualty lists.

Predictive Policing Reinforces Police Bias

Issues surrounding policing in the United States are at the forefront of our national attention. Among these is the use of “predictive policing,” which is the application of statistical or machine learning models to police data, with the goal of predicting where or by whom crime will be committed in the future. Today Significance magazine published an article on this topic that I co-authored with William Isaac. Significance has kindly made this article open access (free!) for all of October. In the article we demonstrate the mechanism by which the use of predictive policing software may amplify the biases that already pervade our criminal ...

Our work has been used by truth commissions, international criminal tribunals, and non-governmental human rights organizations. We have worked with partners on projects on five continents.

Donate