Question: In general, how do personal information aggregators, such as Choicepoint and Accurint, compile data? I do a lot of research using AutoTrackXP and Accurint. Occasionally, I find errors in the reports. It would help me to understand such basics as the sources from which they obtain the data and how they link it; that is, how do they determine a relationship exists?

Sara R. Paul: The best resource available is The Canadian Internet Policy and Public Interest Clinic's survey of how consumers' personal information is collected and sold. This report will thoroughly answer your question!

I also use Accurint and Lexis quite frequently to locate people. After doing a bit of research on aggregators a couple of years ago, I now take the results with a grain of salt. For example, I have found that the real property records are generally correct (since these are often taken directly from government sources) and that the family and possible acquaintances part of the report are almost always wrong (the algorithm can never seem to get that right).

For more info on the accuracy of these services or the industry in general:

• Ackerman, L. and Pierce, D. (2005). Data Aggregators: A Study of Data Quality and
  Responsiveness. Privacy Activism.

• Brooks, N. (2005). Data Brokers: Background and Industry Overview. Congressional
  Research Service.