|Eyewitness Identification Research Laboratory
At the University of Texas at El Paso
Lineup Fairness – General Principles
There are three authoritative recent sources on the matter of lineup fairness: two are more general policy statements while the most recent is a more detailed analysis of lineup structure issues.
Malpass, Tredoux & McQuiston-Surrett (2007) discuss background as well as more technical issues in both the construction and evaluation of eyewitness identification lineups. Coverage is both broad and detailed, covering specific techniques, particularly with regard to lineup evaluation.
The report of the NIJ Technical Working Group on Eyewitness Evidence (1999) sets forth general recommendations on lineup construction, but does not discuss their evaluation. The technical working group was comprised of law enforcement officers, prosecuting attorneys, defense attorneys and scientists who were asked to construct this consensus document by Attorney General Janet Reno. The recommendations are broad but not detailed.
The article by Wells, Small, Penrod, Malpass, Fulero & Brimacombe (1998) is a "white paper" published by Law and Human Behavior, the journal of the American Psychology - Law Society (APLS, Division 41 of the American Psychological Association), with the endorsement of the Society. It provides a broad analysis of eyewitness identification and reviews research findings on many of the issues. The authors propose four model rules for construction and administration of lineups, and one of these relates to lineup construction and evaluation.
These documents set forth two general principles that apply to lineup fairness.
First, the lineup should contain fillers that are good alternatives to the suspect. Fillers are also called “foils” or “distractors”.
* The NIJ publication "Eyewitness Evidence: A Guide for Law Enforcement" recommends that investigators "Include a minimum of five fillers (non-suspects) per identification procedure." (page 29)
* The APLS White Paper proposes that the chances of an innocent suspect being identified in a lineup, even when it is constructed and conducted according to the rules in this article, should be less than 10%. The implication is that there should be at least 9 fillers (10 people including the suspect) in a lineup.
The lineup size principle: nominal size v. effective size.
The principle here is that in addition to the number of persons standing in the lineup (its nominal size) the effective size of the lineup should equal the same number: the lineup fillers should be good alternatives to the suspect. But if a filler is so distinct from the suspect that he could not possibly be mistaken for him, he may as well not be present. Then the effective size of the lineup would be one person smaller than the nominal size. Malpass, Tredoux & McQuiston-Surrett provide quantitative indicators for measuring effective size - the degree to which lineup members are good and effective alternatives to the suspect.
Second, the suspect should not stand out from the fillers.
* The NIJ publication "Eyewitness Evidence: A Guide for Law Enforcement" recommends as follows:
"The investigator shall compose the lineup in such a manner that the suspect does not unduly stand out." (p. 29)
* The APLS White Paper, in their 3rd rule states:
“The suspect should not stand out in the lineup or photospread as being different from the distracters based on the eyewitness’s previous description of the culprit or based on other factors that would draw extra attention to the suspect.”
Lineup bias – toward or away from the suspect.
A lineup may be structured so that choosing the suspect is favored over any of the fillers. Presumably if the fillers are selected effectively, this will not happen, but it will be influenced heavily by the filler selection criteria. Were this to occur the lineup will be biased against the suspect and the lineup, as a safeguard against false identification, will have failed.
It is also possible that the lineup may be constructed so choosing one or more of the fillers is favored over the suspect. Were this to happen the lineup would be biased against the interests of the state, since this could mean that the lineup would work against identification of a guilty suspect. If the fillers are selected effectively this is unlikely to happen, but once more it will be influenced heavily by the filler selection criteria used.
It is important that filler selection is done well, because striking a middle position between bias toward or away from the suspect is instrumental to fairness in the sense of the structure of lineups.Malpass, Tredoux & McQuiston-Surrett provide quantitative indicators for measuring the degree to which the suspect stands out from the other lineup members.