About 77,000 people nationwide are put on trial each year after witnesses picked them out of lineups. Many jurisidictions have moved to overhaul lineups, the New York Times says, as DNA evidence has exposed nearly 200 wrongful convictions, three-quarters of them resulting primarily from bad eyewitness identification. In the new method, the police show witnesses one person at a time, instead of several at once; the lineup is overseen by someone not connected to the case, to avoid anything that could steer witnesses to suspects the police believe are guilty.
The long-awaited results of an experiment in Illinois have raised serious questions about the changes. The study, the first to do a real-life comparison of the old and new methods, found that the new lineups made witnesses less likely to choose anyone. When they did pick a suspect, they were more likely to choose an innocent person. Advocates of the new method said the Illinois study, conducted by the Chicago Police Department, was flawed, because officers supervised the traditional lineups and could have swayed witnesses. The results have empowered critics who had worried that states and cities were caving in to advocacy groups in adopting the new lineups without solid evidence that they improved on the old ones. Lineup methods remain an open debate: law enforcement officials in California and New York have resisted changes, arguing that the evidence in favor of the sequential approach is not firm.
Link: http://www.nytimes.com/2006/04/19/us/19lineup.html?hp&ex=1145505600&en=b7b6