• The fraction of Documents identified as Non-Relevant by a search or review effort that are in fact Relevant. Elusion is estimated by taking a Random Sample from the Null Set and determining how many (or what Proportion of) Documents are actually Relevant. A low Elusion value has commonly been advanced as evidence of an effective search or review effort (see, e.g., Kleen), but that can be misleading because it quantifies only those Relevant Documents that have been missed by the search or review effort; it does not quantify the Relevant Documents found by the search or review effort (i.e., Recall). Consider, for example, a Document Population containing one million Documents, of which ten thousand (or 1%) are Relevant. A search or review effort that returned 1,000 Documents, none of which were Relevant, would have 1.001% Elusion, belying the failure of the search. Elusion = 100% – Negative Predictive Value. 1
  • An information retrieval measure of the proportion of responsive documents that have been missed. Most often used as a quality assurance measure in which a sample of non-retrieved documents is evaluated to determine whether a review has met reasonable criteria for completeness. 2


  1. Maura R. Grossman and Gordon V. Cormack, EDRM page & The Grossman-Cormack Glossary of Technology-Assisted Review, with Foreword by John M. Facciola, U.S. Magistrate Judge2013 Fed. Cts. L. Rev. 7 (January 2013).
  2. Herb Roitblat, Predictive Coding Glossary.
Print Friendly, PDF & Email