A letter to the IDEA Data Center (Aug. 11, 2014), while not addressing an expressly mistaken view on the part of the recipient, addresses the recipient’s failure to recognize the referenced pattern in guidance it provides on the measurement of disproportionality in suspensions and other matters.
Letters to American Statistical Association (Oct. 8, 2015) and several other entities with presumed statistical expertise urge the recipients to explain to the federal government that reducing the frequency of adverse discipline actions tend to increase (a) relative differences in discipline rates and (b) the proportion more susceptible groups make up of persons disciplined.
On February 26, 2016, the Oregon organization Education Northwest gave presentation titled “Reducing Discipline Disparities – What Teachers are Saying” at the NW PBIS Network Spring Conference. A theme stressed in the presentation (at slide 5) was “Reduce the use of exclusionary discipline overall and to eliminate disparities. Slide 7 showed an overall reduction in rates at which students received one or more suspensions from 7.1% in the 2007-08 school year to 3.3% in the 2013/14. Slide 8 showed that during that period the black rate decreased from 18.3% to 10.5% while the white rate decreased from 4.7% to 2.3%.
The PowerPoint presentation does not state anything about the measurement of the disparities, though possibly that was discussed by the presenter. In any case, Table 1 below shows that, as commonly occurs in the circumstances of a larger overall decrease in suspension rates, the ratio of the black suspension rate to the white suspension rate increased (from 3.89 to 4.57, i.e., an increase in the relative difference from 289% to 357%). The table also shows that the three other standard measures of disparity changed in accordance with the usual patterns in the circumstances. That is, the relative difference in avoiding discipline decreased, the absolute difference between rates of suspension (or no suspension) decreased, and the difference measured by the odds ratio increased. The final column (EES for “estimated effect size) indicates that to the extent that difference in the strength of the forces causing the suspension rates of blacks and whites to differ can be measured (see "Race and Mortality Revisited" and the ASA letter), it decreased very slightly.
Table 1. Rates at which black and white students were suspended one or more times in the 2000/08 and 2013/14 school years, with measures of difference