Data used for human rights analysis is almost always incomplete. Minorities and other vulnerable populations may be excluded from official data gathering exercises because of deliberate policy, lack of access or resources, or other reasons. Even when such groups are included, they may make up such a small portion of the sample that it is functionally impossible to use disaggregated data. When human rights groups collect their own data, the scale, timeline, or precariousness of events may confound complete data collection.
When using received data, it is important to understand the sampling and data collection methods and their limitations. Especially crucial here is knowing whether a specific data set is the product of a randomized collection strategy or is a convenience sample.
By using statistical methods like multiple systems estimation, it may still be possible to draw rigorous conclusions from incomplete data. Using such methods requires a high degree of analytical sophistication.