Statistics for eDiscovery: Size and Significance in eDiscovery Sampling
How big a sample do we need in eDiscovery?
The required size of a sample depends on the test being run, the desired confidence level, and the desired confidence interval. At reasonable levels, the number of documents that must be reviewed, for example, is often only a few hundred.
What is a statistically significant sample?
The phrase “statistically significant” when referring to sample size is misused. Samples are not statistically significant. The outcome of a statistical test can be statistically significant if it is unlikely to have occurred by chance. When considering sample size, the relevant dimension is called “power.” Basically, power refers to the sample size required to find significant an effect of a certain size.
Does our desired sample size depend on the size of the population/collection?
For all practical purposes in eDiscovery, the size of the collection does not matter, provided that the size of the collection is much larger than the size of the sample.
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 27th, 2012 at 2:22 pm. 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.
Comments are closed.
OrangeLT's Digital Evidence Services help law firms and corporations conduct the critical eDiscovery tasks of legal hold, data mapping and data collection for multiple sources from multiple locations without incurring the burdensome time, cost and equipment expenses.
To learn more about these services, click here.
The OneO® Discovery Platform is an integrated, web-accessible, forensically sound electronic discovery platform that enables online analytics, processing, and review of data from the security of a hosted centralized repository.
To learn more about OneO®, click here.
Information and Demonstrations from OrangeLT eDiscovery experts can help you quickly understand and begin to evaluate the benefit of our services.
To make an information request or schedule a personalized walk-though of the OneO eDiscovery Platform, click here.
The One Decision® Document Review Accelerator leverages advanced near-duplicate identification technology to enhance reviewer efficiency by grouping similar documents and allowing them to be considered together during legal document reviews as part of the eDiscovery process. This grouping and ability to propagate coding decisions throughout all near-duplicate documents results in more efficient and economical document reviews.
To learn more about One Decision®, click here.
OrangeLT's Complete Managed Review Service helps corporate legal departments and their outside counsel conduct document reviews by combining teams of expertly trained legal review attorneys with an integrated, forensically sound eDiscovery platform to allow for efficient, economical and defensible document reviews.
To learn more about managed review, click here.
The Intermountain eDiscovery Conference (IEDC) is a one-day legal technology conference designed to provide electronic discovery practitioners with information on the latest tools, techniques, and practices of eDiscovery. Hosted Annually by OrangeLT and presented this year in conjunction with the Mountain West Chapter of the Association of Corporate Counsel, IEDC combines presentations, panels and interactive discussions led by nationally recognized experts to create an intimate and educational professional learning experience.
To learn more about IEDC 2013, click here.