Loudoun County, VA Disparities
(Sept. 5, 2017)
Prefatory notes: This subpage discussed discipline disparities issues involving Loudoun County, Virginia, an affluent area just west of Washington, DC, encompassing a number of suburban communities and Dulles International Airport. Loudoun County has the highest median household income of any county in the nation. As such, with regard to many matters, Loudoun County will exhibit patterns like those I discuss in “It’s easy to misunderstand gaps and mistake good fortune for a crisis,” Minneapolis Star Tribune (Feb. 8, 2014), and “The Mismeasure of Health Disparities in Massachusetts and Less Affluent Places,” Quantitative Methods Seminar, Department of Quantitative Health Sciences, University of Massachusetts Medical School (Nov. 18, 2015) (Abstract), whereby comparatively advantaged geographical areas, where adverse outcomes are comparatively uncommon, tend to have comparatively large relative demographic differences in adverse outcomes, but comparatively small relative demographic differences in the corresponding favorable outcomes, than less advantaged areas.
The subpage is related to the California Disparities, Colorado Disparities, Connecticut Disparities, Florida Disparities, Maryland Disparities, Minnesota Disparities, Oregon Disparities, Rhode Island Disparities, Utah Disparities, Beaverton, OR Disparities, Denver Disparities, Henrico County, VA Disparities, Los Angeles SWPBS, Minneapolis Disparities, Montgomery County, MD Disparities, Portland, OR Disparities, St. Paul Disparities, South Bend Disparities subpages of the Discipline Disparities page of jpscanlan.com. These subpages discuss data showing that when discipline rates were reduced in the referenced jurisdictions, relative racial/ethnic differences in discipline rates increased. This subpage is also related to the DOE Equity Report, Suburban Disparities, and Preschool Disparities subpages of the Discipline Disparities page. These subpages address the tendency for relative differences in adverse discipline outcomes be larger, and relative differences in the corresponding favorable outcomes to be smaller, in areas where the adverse outcomes are less common. In the latter regard, a pertinent item is my Letter to Boston Lawyers’ Committee for Civil Rights and Economic Justice (Nov. 12, 2015) explaining that Massachusetts would be expected to have comparatively large relative differences in suspension, but comparatively small relative differences in avoiding suspension, because suspension rates are generally low in Massachusetts.
Most of the aforementioned pages pertain to the widespread belief that generally reducing discipline rates will tend to reduce relative demographic differences in discipline rates and the proportions more susceptible groups make up of persons disciplined. In fact, the exact opposite is the case. Useful comparatively short background readings 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). The pertinent statistical principles are treated at much greater length in “Race and Mortality Revisited,” Society (July/Aug. 2014).
Very recent treatments of the subject may be found in my July 17, 2017 letter to the Departments of Education, Health and Human Services, and Justice, urging them to explain to school administrators that, contrary to what those agencies have led the public and school administrators to believe, generally reducing discipline rates will tend to increase relative differences in discipline rates and the proportions more susceptible groups make up of persons disciplined. The letter is discussed in my “Innumeracy at the Department of Education and the Congressional Committees Overseeing It,” Federalist Society Blog (Aug. 24, 2017), See also my August 25, 2017 letter the American Institutes for Research urging that organization to explain the matter to the Department of Education.
***
For at least several years, the leadership of Loudoun County Public Schools (LCPS) has apparently been concerned about discipline disparities. LCPS contracted for, and in June 2013 received, a report from Hanover Research titled “Ensuring Equitable Discipline: Practices and Policies.” Consistent with the research one could secure from a great many research groups, including the above–mentioned American Institutes for Research, the report reflected the view that School-Wide Positive Behavior Support (SWPBS) policies and like approaches to discipline that generally reduce discipline rates would tend to (a) reduce relative racial/ethnic and other demographic differences in discipline rates and (b) proportions more susceptible groups make up of persons disciplined. As explained in the above materials, exactly the opposite is the case. Generally reducing discipline rates tends to increase both (a) and (b).
The information in Table 1 below is based an April 6, 2015 presentation of the LCPS Special Education Advisory Committee (SEAC) titled “Discipline and Disproportionality, Data and Actions.” The table shows the black and white suspension rates for the five school years for which information was provided in the SEAC presentation. The columns “Black/White Susp Ratio” and “White/Black No-Susp Ratio” show the black to white ratio of experiencing the adverse outcome and the white to black ratio of experiencing the favorable outcome. In these circumstances the relative difference is 1 minus the rate ratio; that is, a ratio of 3.0 means that one group’s rate is 200 percent greater than the other group’s rate. The “EES,” for “estimated size,” shows the measure discussed in "Race and Mortality Revisited" and used in a number of the Discipline Disparities page and many of its subpage that plausibly quantifies the strength of the forces causing the favorable and adverse outcome rates of two groups to differ (sometimes characterized as the differences in the circumstances of the groups reflected by their outcome rates) in a way that is unaffected by the general level of discipline.
Table 1: Black and white suspension rates in LCPS for five school years, with measures of difference.
|
|
School Year
|
Bl Susp Rt
|
Wh Susp Rt
|
B/W Susp Ratio
|
Wh/Bl No Susp Ratio
|
EES
|
|
2009-2010
|
4.60%
|
1.30%
|
3.54
|
1.035
|
0.55
|
|
2010-2011
|
4.30%
|
1.30%
|
3.31
|
1.031
|
0.50
|
|
2011-2012
|
4.60%
|
1.30%
|
3.54
|
1.035
|
0.55
|
|
2012-2013
|
3.70%
|
1.00%
|
3.70
|
1.028
|
0.54
|
|
2013-2014
|
3.00%
|
0.70%
|
4.29
|
1.024
|
0.58
|
A substantial decline in suspensions for blacks and whites occurred between the 2011-2012 and the 2013-2014 school years. Possibly this occurred as a result of the same concerns that prompted LCPS to secure the assistance of Hanover Research, between the 2012-2013 and 2013-2014 years, the June 2013 report itself). As commonly happen in circumstances of a substantial reduction in suspension rates, the relative difference between black and white rates for the declining outcome (suspension) increased, while the relative difference between black and whites rates for the increasing outcome (avoidance of suspension) decreased. The EES indicates that when discipline rates substantially declined (as in the preceding years) the strength of the in forces causing the adverse (or favorable) outcome rates of black and white to differ changed very little.
The SEAC presentation makes a point of the fact that suspension rates in LCPS are lower than nationally and that racial disproportionality is larger than nationally. But, as is essentially universal among analyses of demographic differences in discipline, the presentation does not reflect an understanding of the relationship between low rates and high disproportionality.
With regard to the comparatively low suspension rates in LCPS, the SEAC presentation states (at 9): “The percentage of [LCPS] students who are suspended on a yearly basis has decreased over the past four years to 1.0%. The national suspension rate is 6.0% (EdWeek: February 20, 2014).” The 1.0% figure, which is the 2013-14 school year, is apparently a little later than the EdWeek figure. And overall suspension figures are not useful for comparing rates in different areas because overall rates are functions of the racial/ethnic makeup of students in the places being compared. But, as will be shown in Table 2 below, LCPS suspension rates are indeed quite low compared with national figures.
With regard to comparatively high racial disproportionality in suspension in LCPS, the SEAC presentation states (at 19): “Black [LCPS] students are 3 times more likely to be suspended relative to their percentage of the overall population. The national statistic is 2 times more likely relative to percentage of overall population (EdWeek: September 2014).” The statement regarding LCPS (which is based on 2013-2014 figures showing that blacks made up 6.7% of students and 20.9% of suspended students, actually means that “blacks LCPS student make up 3 times the proportion of suspended students that they make up of the overall student populations,” which is something different from likelihood of suspension. As discussed in Section C of my “The Mismeasure of Discrimination,” Faculty Workshop, University of Kansas School of Law (Sept. 20, 2013), one can never effectively quantify a demographic difference based on the proportion a group makes up of persons potentially experiencing an outcome the proportion the group makes up of persons actually experiencing the outcome. Further, as shown as the penultimate column in Table 2 of the IDEA Data Center Disproportionality Guide subpage of the Discipline Disparities page of jpscanlan.com, for any given pair of black and white suspension rates, the smaller the proportion blacks make up of the student population, the larger will be the relative difference between the proportion they make up of the student population and the proportion they make up of suspended students. Because blacks made up about 7% of LCPS students compared with about 16% of students nationally, for any given pair of suspension rates, the relative difference will be larger for LCPS than nationally. (The matter is more complicated with regard to the absolute difference between the two proportions, as shown in the final column of the table). See also the Addendum to the subpage). Nevertheless, the size of the relative difference between black and white rates will contribute to the proportion blacks make up of suspended students.
Table 2 is based on a table from the aforementioned letter to the Boston Lawyers’ Committee. It was taken from a report where the Lawyers’ Committee was concerned that, while Massachusetts had comparatively low suspension rates, it had comparatively large relative racial differences in discipline rates.
Table 2: 2011-2012 suspension rates in Massachusetts, nationally, and in LCPS nationally, with measures of difference
|
Area
|
Bl Susp Rt
|
Wh Susp Rt
|
B/W Susp Ratio
|
Wh/Bl No Susp Ratio
|
EES
|
|
Massachusetts
|
10.0%
|
2.7%
|
3.70
|
1.08
|
0.65
|
|
National
|
16.4%
|
4.6%
|
3.57
|
1.14
|
0.71
|
|
LCPS
|
4.65
|
1.3%
|
3.54
|
1.04
|
0.55
|
Even though LCPS had the lowest suspension rates of the three areas (which would typically be associated with a comparatively large relative difference in discipline rates), LCPS also had the lowest relative differences in suspensions. And as necessarily happens in such a case, LCPS also had the lowest relative difference in rates of avoiding suspension and the lowest EES. Possibly by the 2013-2014 school year, when the black/white suspension ratio at LCPS had risen to 4.29, LCPS could have a substantially higher relative difference in suspension rates than either Massachusetts or the nation (though, with decreasing suspension rates in those areas, their ratios may have been rising as well). In any case, to the extent that the forces causing suspension rates of black and white students to differ can be measured, it is smaller in LCPS than in Massachusetts or nationally.
Table 3 presents that same information for Hispanics and whites that table 1 presented for blacks and whites.
Table 3: Hispanic and white suspension rates in LCPS for five school years, with measures of difference
|
School Year
|
Hi Susp Rt
|
Wh Susp Rt
|
Hi/W Susp Ratio
|
Wh/Hi No Susp Ratio
|
EES
|
|
2009-2010
|
2.80%
|
1.30%
|
2.15
|
1.015
|
0.32
|
|
2010-2011
|
2.60%
|
1.30%
|
2.00
|
1.013
|
0.29
|
|
2011-2012
|
2.70%
|
1.30%
|
2.08
|
1.014
|
0.30
|
|
2012-2013
|
2.20%
|
1.00%
|
2.20
|
1.012
|
0.32
|
|
2013-2014
|
1.40%
|
0.70%
|
2.00
|
1.007
|
0.26
|
In the case of Hispanics, the general decline in suspension between the 2011-2012 and the 2013-2014 school years was not accompanied by an increase in the relative difference between Hispanic and white suspension rates. There was in fact a slight decrease over the two-year period (though a small increase in the first year). Consistent with that modest departure from the frequency-related pattern in the circumstances, there was a small decrease in the EES.
Thus, in contrast to the situation with blacks and white, some factor other than a general change in the frequency of suspensions may have had some role in the matter, though it was likely a small role.