Over the last few years I have posted a good deal of material here regarding/reflecting my efforts to influence the way researchers and others analyze differences between outcome rates. Issues I have raised principally (though not entirely) involve (a) the failure of analyses to consider the ways measures of differences between outcome rates tend to be affected by the prevalence of an outcome and (b) the mistaken belief that generally reducing an outcome will tend to decrease, rather than increase, relative demographic differences between rates of experiencing the outcome and the proportions groups more susceptible to the outcome make up of persons experiencing it. Some of the more comprehensive discussions may be found in my November 14, 2016 comments to the Commission on Evidence-Based Policymaking and my October 8, 2015 letter to the American Statistical Association.[1,2] See also references 3, 4, and the rather more succinct reference 5.
Below is a recent blog post discussing prospects for government actions regarding these issues by an administration that, rather than being inclined to support longstanding policies, will be inclined to aggressively question those policies.
“Will Trump Have the First Numerate Administration?” Federalist Society Blog (Jan. 4, 2017) http://www.fed-soc.org/blog/detail/will-trump-have-the-first-numerate-administration
Compare the item with the appraisal of prospects for change in the “Looking Forward” section of reference 4 (at 343-345).
Of course, any change requires an understanding of the issues. And the misunderstandings I criticize have persisted for decades in both executive and legislative branches regardless of the party in control. So any change may take some time.
It would be well for the scientific community to address these issues before the government does.
References
1. http://www.jpscanlan.com/images/Comments_of_J_Scanlan_for_Comm_on_Evidence-Based_Policymaking_Nov._14,_2016_.pdf
2. http://jpscanlan.com/images/Letter_to_American_Statistical_Association_Oct._8,_2015_.pdf
3. “The Mismeasure of Health Disparities,” Journal of Public Health Management and Practice (July/Aug. 2016) http://www.jpscanlan.com/images/The_Mismeasure_of_Health_Disparities_JPHMP_2016_.pdf
4. “Race and Mortality Revisited,” Society (July/Aug. 2014) http://jpscanlan.com/images/Race_and_Mortality_Revisited.pdf
5. “Misunderstanding of Statistics Leads to Misguided Law Enforcement Policies,” Amstat News (Dec. 2012) http://magazine.amstat.org/blog/2012/12/01/misguided-law-enforcement/
------------------------------
James Scanlan
James P. Scanlan Attorney At Law
------------------------------