Patterns in the use of best practices for eyewitness identifications in the field

ABSTRACT

Much of the empirical research in the field of eyewitness identifications focuses on factors affecting eyewitness accuracy. The current study focuses on when evidence-based practices for eyewitness identifications are used by police in the field. We specifically explored whether there are factors associated with which law enforcement agencies use best practices for lineups by analyzing secondary data from a large, national sample of law enforcement agencies (Police Executive Research Forum, 2013; N = 619). We focused on four evidence-based practices (double-blind lineups, witness instructions, documenting confidence, and increased training) and analyzed whether agency size and location were related to the use of these practices. We also specifically studied the role of training and its relationship to the use of best practices. Results showed agency size, but not location, was a significant factor with smaller agencies reporting using fewer best practices than larger agencies. Additionally, our results revealed consistent, strong associations between training and implementation of a variety of recommended identification procedures. Qualitative responses also confirmed the important role of training. Implications regarding law enforcement training about eyewitness identifications and the translation of eyewitness research into practice are discussed.

KEYWORDS:

• Eyewitness identification
• photo lineups
• best practices
• training
• agency size

Open Scholarship

This article has earned the Center for Open Science badges for Open Data and Open Materials through Open Practices Disclosure. The data and materials are openly accessible at https://doi.org/10.3886/ICPSR34274.v1 and through the online supplemental materials.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The fifth procedure recommended by the NRC report was video recording all identifications. We did not investigate patterns in use of this procedure as the dataset used here did not contain information on how often agencies video recorded their lineups.

2 A full description of sampling procedures is described in the Police Executive Research Forum’s (2013) report . Briefly, the full population of law enforcement agencies in the United States was gathered from the National Directory of Law Enforcement Agencies. This directory also provided demographic information for these agencies. The sample was stratified by census region (Northeast, Midwest, South, West), agency size (1-25 sworn officers, 26-50, 51-99, 100-499, 500 or more, unknown), and department type (state police, police departments, sheriffs office). The target sample size was 1,401 agencies. Of all agencies that received the survey, 24 reported that they did not conduct eyewitness identifications, leaving a final sample of 1,377. Of these agencies, 619 responded with completed surveys, a 45% response rate. Consistent with past research (Wogalter et al., 1993), response rate differed based on agency size with the largest agencies in the largest strata having the highest response rate (81%) and agencies in the smallest strata having the lowest (29%).

3 A copy of the original survey can be found here: https://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/34274/datadocumentation#

4 For the eight categories for the challenges question, Cronbach’s α ranged from .76 to 1.00 (α_M =.88). For the three categories for the innovative question, Cronbach’s α ranged from .94 to .98 (α_M = .97). For the nine categories for the guidance question, Cronbach’s α ranged from .83 to .99 (α_M = .91).

5 Only categories that contained responses from five or more agencies were included.

Additional information

Funding

This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under grant number DGE-1321846. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors(s) and do not necessarily reflect the views of the National Science Foundation. This project was funded by a research award from the Newkirk Center for Science and Society at the University of California, Irvine and the University of California Consortium on Social Science and Law.