Greetings I am Mary Gray from the American University in Washington, D.C. and a member of Statistics Without Borders. Recently, I was involved in surveying Rwandan prisoners.
- Sample the appropriate stakeholders. Two years after the genocide that killed 800,000 Rwandans, primarily Tutsis, there were 80,00-90,000 imprisoned in a country of a few million and the prison population continued to grow by as many as 10,000 per month, the only release being death. In spite of international horror over the brutal loss of life, international notions of justice demanded due process and some semblance of a speedy trial for the accused. The post-genocide Rwandan government rightly claimed that the fragile judicial system, deprived of most of its personnel and much of its infrastructure, could not handle the prospective case load. Donor governments who had already constructed several large new prisons asserted that however horrible the crimes of which they were accused it was not acceptable to put suspects in prison and throw away the key. Why not, proposed representatives of the US and other nations, with the agreement of the Rwandan government, begin by selecting a sample of prisoners to bring to trial?
- With large populations, stratify the sample. Because conditions under which large numbers of suspects were arrested and imprisoned in different regions of the country, a stratified sample from four regions and the capital Kigali was used.
- Prepare for the worst, records may be inadequate. By the time of the survey the prison conditions were generally adequate but records were difficult to acquire. There were generally lists or card files that could be used for systematic sampling, but the information was generally limited. Usually the crime was listed only as “genocide” without the names of victims or of arresting officers and no reference to the time and place of the offense.
- Prepare for the worst, data may be missing. By the time of the survey little could be done about the missing data so expectations of what information could be gathered had to be revised.
- Educate your stakeholders when possible. The lessons learned are the usual ones about educating those involved about the importance of considering what data needs to be collected. An unfortunate outcome was that those authorizing the survey did not understand that a random sample might not include those whom they were most eager to bring to trial.
The American Evaluation Association is celebrating Statistics Without Borders Week. The contributions all week come from SWB members. Do you have questions, concerns, kudos, or content to extend this aea365 contribution? Please add them in the comments section for this post on the aea365 webpage so that we may enrich our community of practice. Would you like to submit an aea365 Tip? Please send a note of interest to email@example.com. aea365 is sponsored by the American Evaluation Association and provides a Tip-a-Day by and for evaluator.