James P. Scanlan, Attorney at Law

Connecticut Disparities

(Feb. 21, 2015)

Prefatory note:  This subpage is related to the California Disparities, Maryland Disparities, Beaverton, OR Disparities, Los Angeles SWPBS, Denver Disparities, Minneapolis Disparities, Montgomery County, MD Disparities, St. Paul Disparities, Henrico County, VA Disparities, Portland, OR Disparities, DOE Equity Report, Suburban Disparities, and Preschool Disparities subpages of the Discipline Disparities page of jpscanlan.com.  The first ten subpages address studies showing that when discipline rates were reduced in the referenced jurisdictions, relative racial/ethnic differences in discipline rates increased.  The DOE Equity Report subpage addresses a Department of Education study showing that relative RACIAL differences in expulsions are smaller in districts with zero tolerance policies than in districts without zero tolerance policies.  The Suburban Disparities and Preschool Disparities subpages addresses the fact that relative racial differences in discipline rates tend to be greater in suburbs than in central cities, and in preschool than K-12,  simply because discipline rates tend to be lower in suburbs than in central cities and preschool than K-12.

Useful background reading for this page include “Misunderstanding of Statistics Leads to Misguided Law Enforcement Policies, ” Amstat News  (Dec. 2012), “The Paradox of Lowering Standards,” Baltimore Sun (Aug. 5, 2013), “Things government doesn’t know about racial disparities,” The Hill (Jan. 28, 2014), and “Race and Mortality Revisited,” Society (July/Aug. 2014).  All address the fact that contrary to the view promoted by the Departments of Education and Justice that reducing discipline rates will tend to reduce racial and ethnic differences in discipline rates, reducing discipline rates will tend to increase relative racial differences in discipline rates (though reduce relative differences in rates of avoiding discipline).  My recent “The Perverse Enforcement of Fair Lending Law,” Mortgage Banking (May 2014) addresses the same issue in the lending context.  The statistical issues have also been recently addressed in my November 17, 2014 amicus curiae brief in Texas Department of Housing and Community Development, et al. v.  The Inclusive Communities Project, Inc., Supreme Court No. 13-1731.

***

On February 12, 2015, Connecticut Voices for Children issued a report title Keeping Kids in Class: School Discipline in Connecticut, 2008-2013, discussing recent decreases in arrests, suspensions and expulsions in Connecticut public schools and changes in demographic differences in rates of experiencing these outcomes.  In the main, the report found that reductions in the above outcomes were accompanied by increased relative differences in rates of experiencing those outcomes by race/ethnicity or special education status.

The report noted, for example: 

            • Page 6:  “In 2013, black students were arrested at 4.7 times the rate of white students, up from 3.6 times the rate in 2011 (page 6).

            • Page 7:  “Special education students were arrested at 3.0 times the rate of general education students in 2013, up from 2.7 times the rate in 2011.”

            • Page 12:  “Black students were 4.9 times more likely [[i]] to be expelled than white students, up from 4.2 times more likely in 2011. 

            • Page 13:  “In 2013, special education students were 1.8 times more likely to be expelled than general education students, an increase from 2011, where they were 1.5 times more likely to be expelled.” 

There were a few exceptions from the pattern of increasing relative differences in rates of experiencing the outcome examined, though nothing remarkable on its own. 

At some point I may discuss some of the data with respect to the point underlying the structure of Table 1 the main Discipline Disparities page.  The point involves the problematic nature of analyses of outcome rates for what I will term “interim outcomes” until I think of a better usage.  It involves the same situation that would arise if one tried to analyze disparities in rates at which students received F grades and disparities in rates at which they receive D grades.  Analyses of disparities in receipt of grades can be valid, but the analyses of disparities in rates of receipt of D grades cannot, save when combined with an analysis of F grades (that is, in terms of rates of receiving D or F).  The same holds with respect to analyses of differences in rates of experiencing fair and poor health.  One can reasonably analyze disparities in rates of rates of experiencing fair or poor health (i.e., the two combined), as in fact is often done (though not necessarily through a sound approach, as discussed, for example, on the Reporting Heterogeneity subpage of the Measuring Health Disparities and the “Illogical Premises and Unfounded Inferences” section of the "Race and Mortality Revisited").  One can also reasonably analyze disparities in rate of experiencing poor health.  But one cannot reasonably analyze disparities in rates of experiencing fair health.  Similarly, one may reasonably analyze disparities in rates of falling above or below certain proficiency levels – basic level, proficient level, advanced level – as is often done (though never soundly, as discussed on the Educational Disparities page and its subpages).  But one cannot reasonably analyze rates of falling within the proficient category. 

Discipline disparities analyses are the only kinds of analyses where I have noticed separate treatment of disparities in interim outcomes, which commonly include analyses of disparities in out-of-school suspensions (OSS) separate from expulsions and analyses of in-school suspensions (ISS) apart from OSS and expulsions.  The only sound categories for analyses would be (a) expulsions alone, (b) expulsions and OSS, (c) expulsions, OSS, and ISS.