James P. Scanlan, Attorney at Law

California Disparities

(Mar. 29, 2014; revised Apr. 26, 2015)

This subpage is related to the Los Angeles SWPBS, Denver Disparities, Florida Disparities, Maryland Disparities, Connecticut Disparities,  Maryland Disparities , Minnesota Disparities,  Rhode Island Disparities, St. Paul Disparities, Minneapolis Disparities, Beaverton (OR) Disparities, Portland (OR) Disparities, Montgomery County (MD) Disparities, and Henrico County (VA) Disparities,

and DOE Equity Report subpages of the Discipline Disparities page.  All but the last subpage address reports indicating that when discipline rates in public schools the referenced jurisdiction reduced, relative differences in suspensions increased.  The last 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. 

Articles explaining why, contrary to the near universal belief that generally reducing discipline rates, will tend to increase relative differences in discipline, generally reducing discipline rates will tend to increase relative differences in discipline rates include “Race and Mortality Revisited,” Society (July/Aug. 2014), “Things government doesn’t know about racial disparities,” The Hill (Jan. 28, 2014), “The Paradox of Lowering Standards,” Baltimore Sun (Aug. 5, 2013), “Misunderstanding of Statistics Leads to Misguided Law Enforcement Policies, ” Amstat News  (Dec. 2012).  Letters of March 20, 2015, and April 1, 2013, attempting to explain this to the Senate Committee on Health, Education, Labor and Pensions may be found here and here. 

***

On January 29, 2014, the California Department of Education issued a press released titled “State Schools Chief Tom Torlakson Reports California See Significant Drops in School Suspensions and Expulsion” noted reductions in suspensions of public school students in almost every ethnic group.  The announcement stated that 2012-13 school year data showed that suspensions were down from the year before by 9.5 percent for African Americans, 10.2 percent for Hispanics, and 10.8 percent for whites.  But it also noted that racial/ethnic disparities still existed, citing data indicating that African Americans and Hispanics were suspended 6.5 times and 2.6 times as often as whites, and that African Americans were suspended 2.5 times as often as Hispanics.

The report did not discuss implications of the comparative sizes of the reductions.  But, allowing that changes in numbers of students enrolled may affect things somewhat, the comparative sizes of the reductions typically would mean that the disproportionality between whites and both minority groups had increased from the prior year, as had the disproportionality between Hispanics and African Americans.

Material added April 14, 2015:

On January 14, 2015, the California Department of Education issued another press released titled “State Schools Chief Tom Torlakson Reports significant Drop in Suspensions and Expulsions for Second Year in a Row.”  This press release reported that in the 2013-14, the numbers of suspensions were down from the prior year by 14.1 percent for African-Americans, 15.2 percent for Hispanics, and 16.6 percent for whites. 

The above observation that “changes in numbers of students enrolled may affect things somewhat” is of some pertinence to Hispanics, since Hispanic enrollment has been increasing in recent years while the African American and white enrollments have been decreasing.  The table below shows the actual percentage reductions in out-of-school suspension rates (rather than numbers of suspensions) for each group from the 2011-12 school year to the 2013-14 school year.  (The table is based on data made available on the California Department of Education website that show somewhat different figures from those in the press release.)  While the African American rate showed the least percentage reduction, the Hispanic rate showed a larger percentage reduction than the white rate.  Thus, relative difference between African American rates and white rates and between African American and Hispanic rates increased, while the relative difference between the Hispanic rate and the white rate decreased.   The final column, presents information on the change in terms of the estimated effect size, using the measure as I did in Table 7 of “Race and Mortality Revisited,” Society (July/Aug. 2014) an on the Subgroup Effects subpage of the Scanlan’s Rule page of jpscanlan.com.  These show that, to the extent that changes can be effectively measured, the African American and Hispanic rates changed more than the white rates. 

Table 1.  Out-of-school suspension rates for African American, Hispanic and white students in California schools for the 2011-12 and 2013-14 school years, with percentage reductions and EES (for estimated effect size)

The reader should be mindful that the EES figures are for the changes experiencing by each group, not the changes in the disparity between each group.  Thus, for example, the EES for the black-white differences was .87 in 2011-12 and .81 in 2013-14.

Ideally, at some point I will refine the above analysis to include in-school suspensions and expulsions.  Out-of-school suspensions is an intermediate outcome for which the EES approach (or any other measurement approach) is problematic.  That is, one can effectively analyze rates of experiencing in-school suspension or worse, out-of-school suspension or worse, or expulsion but one cannot effectively analyze either in-school suspensions or out-of-school suspensions by themselves (just as one can analyzes differences in rates of receiving grades C or below, D or below, and F, but not rates of receiving grades of C, grades of D,  or grades of C or D.  See discussion of this issue on the Discipline Disparities page and its Connecticut Disparities subpage.  But the expulsion category has few enough events that the analysis in Table 1 would not be materially affected.