This message has been cross posted to the following eGroups: Washington Statistical Society and ASA Connect .
-------------------------------------------
The biggest news in statistics over the last week is the Department of Justice's report finding a racially disparate impact of law enforcement practices in Ferguson, MO based on the higher proportion African Americans comprise of persons experiencing adverse outcomes in the criminal justice system than they do of the Ferguson population. The report blames the impact mainly on overly aggressive policing and unduly harsh court rules.
Two years ago, I explained in an Amstat News Statistician's View column [1] some of the law enforcement anomalies arising from the fact that the DOJ and other federal agencies have long been proceeding under the mistaken view that relaxing standards and thereby reducing the frequency of adverse outcomes (like rejection of loan mortgage applications or suspension from public schools) will tend to reduce relative racial differences in rates of experiencing those outcomes. In fact, while reducing the frequency of adverse outcomes will tend to reduce relative differences in rates of avoiding the outcomes (i.e., experiencing the corresponding favorable outcomes), it will tend to increase relative differences in rates of experiencing the adverse outcomes. Unaware that reducing the frequency of adverse outcomes tends to increase relative differences in rates of experiencing the outcome - indeed, believing the opposite - the federal government has encouraged entities covered by civil rights laws to relax standards, while at the same time continuing to monitor the fairness of practices on the basis of relative differences in adverse outcome. Thus, the government has been encouraging covered entities to do things that increase the chances that the government will sue them for discrimination.
A corollary to the pattern whereby reducing the frequency of an outcome tends to increase relative differences in experiencing it is a pattern whereby reducing the frequency of an outcome increases the proportion groups most susceptible to the outcome comprise of those experiencing the outcome. In Ferguson, this means that relaxing the standards that the DOJ regards as having a disparate on African Americans because of the high proportion African Americans comprise of persons experiencing adverse outcomes will tend to increase, not decrease, those proportions. I attempt to explain this to officials of DOJ and the City of Ferguson in a March 9, 2015 letter, which is available online.[2]
As the letter will show, this is not my first effort to explain this to the DOJ or other federal agencies and, on its own, I am not sure this effort will be any more successful than earlier ones. But eventually the DOJ and other agencies must recognize the ways in which their beliefs about the effects of relaxing standards on relative racial differences in adverse outcomes have been mistaken, as well as the havoc caused by that failure of understanding. The final pages of the letter discuss some of issues DOJ will then face with respect to correcting the misunderstandings that it has been promoting for so long.
I will shortly be requesting ASA to formally address with DOJ and other agencies their misunderstanding about the effects of lowering standards on relative differences in adverse outcomes and certain other measurement issues. Meanwhile, I hope ASA members who are part of DOJ will examine the letter and internally urge the agency to seek expert guidance on whether its practices are as perverse as the letter suggests.
1. "Misunderstanding of Statistics Leads to Misguided Law Enforcement Policies," Amstat News (Dec. 2012) http://magazine.amstat.org/blog/2012/12/01/misguided-law-enforcement/
2. http://jpscanlan.com/images/Letter_to_Department_of_Justice_and_City_of_Ferguson_Mar._9,_2015_.pdf
-------------------------------------------
James Scanlan
James P. Scanlan Attorney At Law
-------------------------------------------