The reduction of disproportionality was a key concern of Congress in writing IDEA '97, but it is the Office for Civil Rights that appears to have taken the lead in addressing this issue.

I do not have the experience you are soliciting. In general, however, disproportionality has been cited as a major problem in the mentally retarded and emotionally disabled categories, with the former, however, drawing special concern. The federal statute said,

I previously submitted a question along similar lines to this Listserve; one respondent said that their school system had intensified efforts to increase the effectiveness of interventions in the intervention or assistance team prior to referral. My understanding is that the efforts resulted in reduced referrals, but unfortunately the proportionality held. Someone else indicated that in reaching a consent agreement, their district had agreed to reevaluate all the native Americans enrolled in their school system; the reevaluations, however, were confirming what had been found in previous assessments, so OCR's recommendations for reassessment were apparently not helpful in reducing the disproportionality for that group.

Wake County in North Carolina was investigated based on allegations of disparate impact in several areas (disciplinary actions, I believe, were also a concern) by OCR last fall, and I know they signed a consent agreement--primarily, I inferred, because some of its schools were experiencing significantly more disparate impact than others. I have not been able to find the details of the agreement on the Internet, nor do I know if the actions they undertook have been effective.

I do think it would be important to determine whether or not the over representation you cite was a consistent phenomenon across all schools in your district; or particularly problematic in some. If the latter were to be the case, I think it would be prudent to focus internally on what the teams were doing differently as a starting point. (I think that's what OCR would do.)

OCR has said that disproportionality does not in and of itself prove discrimination. But this is a major area for them, and providing a convincing defense that YOUR disproportionality is not discriminatory could be problematic. Unless a school system offers up an agreement that is satisfactory to the complainant, these investigations have been known to run on for months with no end in sight.

It is important to understand that disparities in student performance based on race, national origin, sex, or disability, alone, do not constitute disparate impact discrimination under federal law. Furthermore, nothing in federal law guarantees equal results. (OCR, Draft High-Stakes Decisions, 2000, p. 60.)

Theoretically, even if your practices do result in disparate impact, you would be able to justify it if the tests you are using serve a legitimate educational interest--assuming the complainant can't come up with a way to accomplish the same goals with less disparate impact.

Where the use of a test results in decisions that have a disparate impact on the basis of race, national origin, or sex, the test use causing the disparity must significantly serve the legitimate educational goals of the institution.168 This inquiry is usually referred to as determining the "educational necessity" of the test use or determining whether the test is "educationally justified."169 The test need not be "essential" or "indispensable" to achieving the institution's educational goal; 170 rather, the educational institution must show a manifest relationship between use of the test and the institution's educational purposes. (OCR, Draft High-Stakes Decisions, 2000, p. 3)

Though that sounds simple enough, this can be a tough row to hoe, as more than one district has discovered.

I wish I had a solution, because this is a nation-wide dilemma.

Guy

Reference: Making High-Stakes Decisions for Students