Capturing human rights data in Analyzer

Human rights groups collect data containing details of human right abuses from various sources, including medical records, newspaper articles, witness testimonies, letters, interviews, and official reports and documents. Analyzer can be used to capture this data for analysis.

Data is coded according to the "Who did what to whom" model and entered into the capture set of Analyzer. Data about the source of the information is entered in the source tab, shown in Figure 1. (Note: the data in the figures shown here, unless otherwise indicated, are from a sample, not an actual, dataset.)

Figure 1 - Analyzer source tab

Each source document can include data about one or many victims and one or many perpetrators. Data related to the participants is entered in to the participants tab, shown in Figure 2. Each victim can suffer from one or several human rights violations.

Figure 2 - Analyzer participants tab

Data about the violations is entered into the acts tab, shown in Figure 3. All of the data entered into the source, participants and acts tabs, and the relationships between the pieces of information are carefully stored in the Analyzer database.

Figure 3 - Analyzer acts tab

When trying to calculate the magnitude of abuses, it is critical to distinguish where reports overlap so that violations are not over-counted. The overlaps must be identified by making judgments about which records are unique and which records describe the same participants and events. This process is called creating a "judgment layer" on top of many sources.

Analyzer separates the source and judgment layers within the database. Identifying sources that refer to the same people and/or events is called matching. Analyzer facilitates this process with an interface that enables data analysts to view fields from several records side-by-side. Decisions from the judgment layer link back to the source layer providing an audit trail for all of the records and the decisions made about those records in the database.

The data fields that appear, and the order in which they appear, in this screen are fully customizable. This enables the user to select the data fields that are the most relevant for making decisions about whether or not two records refer to the same person, which varies from project to project.

The source and judgment layers together can be used to calculate statistics, while the judgment layer is the basis for Multiple Systems Estimation (MSE). Analyzer can combine data from multiple data collection projects, facilitate matching within each project and then between the projects, to do Multiple Systems Estimation.

If you are interested in using Analyzer, please contact us at info@hrdag.org.

Consulting Services | About HRDAG | Resources | RSS | Home

© Copyright © 2002-2010 Beneficent Technology, Inc. (The Benetech® Initiative)
Benetech is a California nonprofit organization.

The Human Rights Data Analysis Group (HRDAG) applies information technology solutions and statistical techniques to help human rights advocates build evidence-based arguments.