Note added March 31, 2015:I recently explained this issue to the Portland Board of Education by letter of February 25, 2015, and by letter of March 20, 2015, to the Senate Committee on Health, Education, Labor and Pensions by letter of March 20, 2015.The latter item pertains to the legislation discussed on the Keep Kids in School Act subpage.
A February 16, 2014 Oregonian article titled “Portland expels, suspends fewer students, but still disciplines African American students at higher rates” that discussed released data on discipline rates in Portland’s Public Schools began as follows:
“Roughly 3 percent of Portland Public Schools students were suspended or expelled last school year, a drop from nearly 5 percent the year before, according to data released by the district.
“But African American and Native American students are still being sent out of class at far higher rates than their white peers--almost four-and-a-half times more often for African American students.”
The article’s title and the “But” beginning the second paragraph reflect the common mistaken belief that reducing discipline rates will tend to reduce relative difference in discipline rates. As discussed in the prefatory note, the opposite is the case.
While the data cited in the article and the underlying report show some variation from the standard patterns by which measures tend to change as the frequency of an outcome changes, with respect to the black-white and Native-American disparities noted in the article’s second paragraph, the table in the article on rates of suspension or expulsion shows that the ratio of the black rate to the white rate rose from 4.36 in 2011-2012 to 4.57 in 2013-2014 and that the ratio of the Native-American rate to the white rate rose from 2.3 in 2011-2012 to 3.22 in 2013-2014.
There will be variations from the pattern described in the prefatory note (as in the case of the decline in black/white ratio between 2012-2013 and 2013-2014that can be derived from the table in the article). But in general the further reductions in discipline rates planned by school administrators can be expected to increase relative differences in discipline rates rather than reduce them.
In the case of the substantial decreases in exclusionary discipline at the Boise-Eliot/Humbolt K-8 discussed in the article as a success story (from 15% in 2012-2013 to 7 % in 2013-2014), the 51st page of the underlying report shows that the ratio of the black rate to the white rate increased from 1.2 to 2.6.