Search results

Showing 1 – 6 of 6 results.
Curated

Eyewitness Identification: A Systematic Investigation of Lineup Composition and Fairness, United States, 2019-2022 (ICPSR 38761)

Released/updated on: 2025-06-26
Geographic coverage: United States
Time period: 2019-01-01--2022-12-31

The major objective of this project was to investigate photo array composition in order to improve eyewitness identification procedures. Photo array composition involves the fillers, or known-innocent individuals that police add to a photo array so that the perpetrator/suspect (referred to as the "target" in experimental design) does not stand out. An unbiased (fair) photo array contains fillers that match the description of the suspect provided by one or more eyewitnesses. In contrast, in a biased photo array, the suspect stands out from the fillers. Another popular procedure used by police is the showup, when the suspect is presented without any fillers. This project involved all three of these procedures.

This collection contains raw and aggregated data from 12 sets of experiments that investigated different aspects of eyewitness identification, including fair vs. biased lineups, lineup size, distinctive facial features, target-filler similarity, impact of sleep on eyewitness accuracy, memory strength, number of suspects presented, impact of courtroom instructions and expert expertise, and speed and confidence of eyewitness identification. Each experiment set followed a similar general design, with variations based on the purpose and hypotheses of the specific study. United States-based adult participants recruited via SurveyMonkey were asked to complete an online experiment in which they would be presented with a crime vignette and a suspect facial image (created from a faces database), given a distractor task, and then asked to select the suspect from a lineup and rate the confidence level of their decision.

The data were provided to ICPSR in Excel workbook format (41 data files, 3 codebooks) and are available for download as a zipped package. ICPSR has not modified the files from the format in which they were supplied. Data files are organized into subfolders that are named with a short content descriptor and citation of the relevant publication. Unless noted, data files contain a "codes" sheet that explains the variables and experimental condition groups. Articles and theses/dissertations that used each dataset are available under Data-related Publications. Please refer to the ICPSR README for more information.

Curated
Partially restricted

Influence of Eyewitness Memory Factors on Plea Bargaining Decisions by Prosecution and Defense Attorneys in California, 2010-2011 (ICPSR 32181)

Released/updated on: 2012-07-31
Geographic coverage: United States, California
Time period: 2010-01-01--2011-12-31
The purpose of the study was to assess how the strength of eyewitness evidence affects plea bargaining decisions by prosecutors and defense attorneys. Surveys were administered to 93 defense attorneys and 46 prosecutors from matched counties in California. On the questionnaire, each participant was asked four background questions and read four versions of a crime scenario in which two specific eyewitness factors -- (a) same- versus cross-race identification and (b) prior contact or not -- were experimentally manipulated in a factorial design. The scenarios described a store robbery in which identification by one eyewitness was the only evidence against the defendant. After reading each scenario, attorneys were asked to respond to five questions in light of the facts presented. The study contains 28 variables including background data on the study participants and responses to questions relating to four crime scenarios that differ in terms of the details of the eyewitness evidence.
Curated
Partially restricted
Simple Crosstabs

National Survey of Eyewitness Identification Procedure in Law Enforcement Agencies, 1994-2012 (ICPSR 34274)

Released/updated on: 2014-03-07
Geographic coverage: United States
Time period: 1994-01-01--2012-12-31
The data results from a study conducted by the Police Executive Research Forum (PERF) designed to obtain the first nationwide assessment of the state of the criminal justice field regarding eyewitness identification procedures used by law enforcement agencies. PERF designed and conducted a survey of 619 police departments across the United States. The study focused on the departments training and policy when conducting eyewitness identification; particularly the study examined the use of "blind" administrators and the use of simultaneous or sequential presentation to the witness. The number of lineup members, witness instructions, police training, number of witness viewings and recording of the witness statements were also examined. A pilot test of the survey was conducted prior to the study.
Curated
Restricted

Reduction of False Convictions through Improved Identification Procedures: Further Refinements for Street Practice and Public Policy, 1983-2010, in five countries. (ICPSR 34316)

Released/updated on: 2016-04-28
Geographic coverage: Canada, United States, Tucson, United Kingdom, South Africa, Germany, Arizona
Time period: 1983-01-01--2010-12-31

These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.

This study was a three part project which evaluated the procedural aspects of police lineups. The first part was a meta-analysis of existing laboratory data on comparative eyewitness accuracy rates for sequential versus simultaneous lineups. The second part was three experiments on the elements of current field lineup practices in simultaneous and sequential lineups. The third part was a field experiment in Tucson, Arizona, which tested double-blind simultaneous versus double-blind sequential lineups.

Curated
Restricted

Social Science Research on Wrongful Convictions and Near Misses, 1980-2012 (ICPSR 34522)

Released/updated on: 2016-05-31
Geographic coverage: United States
Time period: 1980-01-01--2012-01-01, 2012-02-01--2012-02-29

These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.

This study examined how the criminal justice system avoids wrongful convictions by comparing violent felony cases that ended in an official exoneration after conviction ("wrongful convictions") with those in which defendants had charges dismissed before trial or were acquitted on the basis of their factual innocence ("near misses"). Data were collected on a total of 460 cases (260 wrongful convictions and 200 near misses), and these cases were compared quantitatively and qualitatively on variables that might explain the different outcomes. These variables included the usual causes of wrongful convictions, such as eyewitness misidentification, false confession, and forensic error, as well as demographic, social, and procedural variables.

Curated

Use and Effectiveness of Hypnosis and the Cognitive Interview for Enhancing Eyewitness Recall: Philadelphia, 1988-1989 (ICPSR 9478)

Released/updated on: 2006-03-30
Geographic coverage: United States, Philadelphia, Pennsylvania
Time period: 1988-01-01--1989-06-30
This study investigated the effectiveness of hypnosis and the cognitive interview (a technique for stimulating memory) on the recall of events in a criminal incident. The data collected in the study address the following questions: (1) Does hypnosis or the cognitive interview mitigate recall deficits that result from emotionally upsetting events? (2) Does hypnosis or the cognitive interview improve recall when individuals recall events in narrative fashion? (3) Does hypnosis or the cognitive interview improve recall when individuals are required to respond to each item in a set of focused questions? (4) Does the cognitive interview improve recall better than motivated control recall procedures? For this two-stage study, subjects were randomly assigned to receive hypnosis, cognitive interview, or control treatment. Stage 1 involved completing unrelated questionnaires and viewing a short film containing an emotionally upsetting criminal event. Stage 2 was conducted 3 to 13 days later (the average was 6.5 days) and involved baseline information gathering about the events in the film, application of the assigned treatment, and post-treatment written recall of the events. Data were collected from the written narratives provided by subjects and from an oral forced recall of events in a post-experimental interview. Variables in File 1 include total information (correct, incorrect, confabulations, and attributions) as well as new information given in the post-treatment written narrative. The remaining variables in File 1 include score on Harvard Group Scale of Hypnotic Susceptibility, Form A (HGSHS:A), repressor status, and the number of days between viewing the film and completing the baseline and post-treatment interviews. Variables in File 2 were derived from the post-experimental oral forced recall interview and include total correct and incorrect responses and confidence ratings for correct and incorrect responses. The unit of observation is the individual.
Back to top