Prefatory note:This subpage is related to the California Disparities, Maryland Disparities, Los Angeles SWPBS, Denver Disparities, Minneapolis Disparities, St. Paul Disparities, DOE Equity Report, and the Suburban Disparities subpages of the Discipline Disparities page of jpscanlan.com.The former six subpages address studies showing that when discipline rates were reduced in the referenced jurisdictions, relative racial/ethnic differences in discipline rates increased.The seventh subpage addresses a Department of Education study showing that relative differences in expulsions are smaller in districts with zero tolerance policies than in districts without zero tolerance policies. The eighth subpage addresses the fact that relative racial differences in discipline rates tend to be greater in suburbs than in central cities simply because discipline rates tend to be lower in suburbs and in central cities.
In a March 26, 2014 Education Week commentary titled “Using Evidence to Impose Discipline Fairly,” the superintendent of the Beaverton , Oregon School District discussed that his district had partnered with five other districts in the Portland area and REL Northwest (also known as Education Northwest) to create the Oregon Leadership Network Research Alliance and that the alliance was using research and data to document and address disparities in student discipline, as well as in graduation rates.The main theme of the article was that the high quality data developed in this project make addressing disparities more scientific.
While not getting into the matter in detail, the article reflects the common view that generally reducing discipline rates will reduce racial differences in discipline rates.It noted:
“Next, REL analyzed the districts' discipline data by race/ethnicity, special education status, and grade level to help each district identify the students who are most frequently and severely disciplined. For me, this clarified how big an issue discipline disparity is in Beaverton's 51 schools. We learned, for instance, that, in February 2013, 7.4 percent of black middle school students in Beaverton had been suspended or expelled, compared with 2.7 percent of white students. In February 2014, the percentages were reduced to 6.5 percent of black students and 2.2 percent of white pupils, but there's more work to do.”
The referenced black and white suspension or expulsion figures are set out in Table 1.The table shows the common pattern found when discipline rates generally decrease.That is, the ratio of the black discipline rate to the white discipline rate increased, while the ratio of the white rate of avoiding discipline to the black rate of avoiding discipline decreased.Also, as commonly occurs when an outcome in the rate ranges at issue decreases, the absolute difference between rates decreased.The EES figure, indicates that, to the degree that we able to measure the strength of the forces causing the discipline rates to differ, the strength of those forces remained essentially unchanged.It is doubtful that the slight increase in the EES was statistically significant.
Table 1.Black and white suspension or expulsion rates in Beaverton OR, with measures of difference [ref b5326a2]
B/W Adverse Ratio
W/B Favorable Ratio
Beaverton appears to be a suburban community, though, as the article notes, with significant demographics shifts in recent years.I am not in a position to say whether the suspension and expulsion rates are particularly low or the relative differences in the adverse outcome are particularly high.I note, however, that administrators in suburban school districts will no doubt often be surprised at the size of the relative racial and ethnic differences in their districts, while typically failing to recognize that connection between large relative differences in an outcome and infrequency of an outcome.See the Suburban Disparitiessubpage and my “It’s easy to misunderstand gaps and mistake good fortune for a crisis,” Minneapolis StarTribune(Feb. 8, 2014)
In a comment I posted on the Education Week article on May 25, 2014, I noted that it was pointless to talk about evidence-based decisions, or to examine evidence, unless one understands how to analyze evidence (which would, of course, mean understanding the ways measures tend to change because the prevalence of an outcome changes).This criticism could be made as well with respect to the health and healthcare disparities research of the Agency for Healthcare Research and Quality, the Centers for Disease Control and Prevention, the Institute of Medicine, Harvard Medical School, Harvard School of Public Health, and Massachusetts General Hospital (as I do at pages 26-32 of the Federal Committee on Statistical Methodology 2013 Research Conference (FCSM) paper “Measuring Health and Healthcare Disparities”)and remainder of research institutions around the world that employ standard measures of differences between outcome rates without recognizing that the measures tend to be systematically affected by the frequency of an outcome.See also the October 2012 Applied Statistics Workshop at Harvard’s Institute for Quantitative Social Science ““The Mismeasure of Group Differences in the Law and the Social and Medical Sciences” and the September 2013 University of Kansas School of Law Faculty Workshop page “The Mismeasure of Discrimination.”