Online Program

Return to main conference page
Friday, February 15
Fri, Feb 15, 5:15 PM - 6:30 PM
St. James Ballroom
Poster Session 2 and Refreshments

Development of Improved Statistical Methodology for Eyewitness Identification (303802)

View Presentation View Presentation

*Alice J. Liu, University of Virginia 

Keywords: eyewitness identification, positive predictive value, negative predictive value, decision

Eyewitness lineups are frequently used in law enforcement and are crucial in many criminal cases. However, law enforcement do not have the time and resources available to conduct the much-needed research for the development and validation of more reliable practices. Research in the effectiveness of law enforcement practices for eyewitness identification (EWI) procedures remains incomplete. In fact, in the EWI research field, there is a divide on how to analyze the data, collected in either a laboratory setting or real data from law enforcement agencies, and how to present the results. EWI errors have severe consequences, both in convicting the innocent and in freeing the guilty to commit other crimes. A key aspect to this research is to produce more discerning statistical and analytical models for assessing EWI procedures, considering and maximizing both positive and negative predictive values (PPV and NPV, respectively), as both types of errors are important considerations in policymaking. Currently, the statistical analyses only take into consideration PPV, but not NPV. Other statistical approaches that properly incorporate several covariates will also be considered.