“Process and Sample” or “Sample and Process”?

Statistics for eDiscovery: “Process and Sample” or “Sample and Process”?

If I have to process all my documents in order to do sampling, doesn’t that make sampling very expensive?

Sampling applies to the documents from which you sampled. For example, you may not have processed ESI from custodians who are unlikely to have relevant information. Your sample is not representative of documents that were unavailable to your sample. Another scenario involves the situation where you want to assess whether documents there were documents among those that you did not process. Sampling itself does not depend on whether you processed files or not, only on whether they could be included in your sampling frame. You could just as easily sample and then process as process and then sample. For example, you could choose a random sample of custodians, then choose a random sample of PST files from those custodians, and then process only those PST files and choose a random sample of emails from them. Each custodian, each PST and each email then had an equal chance of appearing in the sample, so the standard statistical estimates would apply as normal.

Source:  Statistics For eDiscovery (PDF) - Used by Permission of Herbert L. Roitblat, Ph.D.

Series:  Statistics for eDiscovery – Blog Posts – Orange Legal Technologies.

Orange Legal Technologies helps corporate legal departments and their outside counsel conduct the critical electronic discovery task of document review by providing advanced predictive coding technologies and expert reviewer assistance to accelerate the electronic discovery process. To learn more about our automated predictive review services, click here.

This entry was posted on Monday, February 20th, 2012 at 11:51 am. It is filed under metadata, resources and tagged with resources, services. You can follow any responses to this entry through the RSS 2.0 feed.