Spiegel Online has an insightful update called ‘No Convictions to Date’: Study Finds Ethnic Profiling Useless in Preventing Terror

The report, entitled “Ethnic Profiling in the European Union,” argues that profiling is both ineffective and counterproductive, pointing out that “stops and searches conducted under counterterrorism powers in Europe have produced few charges on terrorism offenses and no terrorism convictions to date.” At the same time, targeting specific communities alienates them, “contributing to a growing sense of marginalization in minority and immigrant communities.”

A study across Spain, Bulgaria and Hungary showed that ethnic profiling is popular even when it does not seem to be effective. In fact, the researchers noted that while minorities often were three times as likely to be stopped by police the groups were statistically were less likely to be offenders than a local majority. One might think this alone would change police behavior, but obviously they do not see the forest for the trees. What’s the solution? Accountability and transparency.

The simple act of asking police officers to record the details of their stops made a difference: The more paperwork involved, the fewer stops cops were likely to make in the first place. But applying more stringent rules to how and when police could stop people also changed their behaviour dramatically, and resulted in fewer innocent people being hassled. “Training officers to think about who they were stopping and why led to a reduction in stops and an increase in effectiveness,” Neild says.

First of all that sounds bad. I thought the point was to reduce the number of false positives. Requiring more paperwork is not a good solution if it also reduces the number of true positives.

Second, I do not think this research argues against using hunches, as the article concludes. It calls for the introduction of more reliable and repeatable security practices related to data. In other words the police should still go with their hunches, but they also should be more exposed to the facts. I would like to believe that if management could see how time was wasted, they would course-correct. On the other hand, this assumes a balanced perspective. It is not clear in the study whether they account for the fact that local majorities often expect and want police to harass minorities. That is a whole other problem.