August 21, 1998
P.O. Box 2111
Roswell, GA 30077-2111
URL: www.heir.org
email: info@heir.org
Dr. James L. Muyskens
Senior Vice Chancellor
Board of Regents
270 Washington Street, S.W.
Atlanta, GA 30334
Dear Dr. Muyskens:
This letter is a response to your letter of February 20, 1998 regarding our meeting with you and our analysis of the "C equivalency" of the SAT II cut scores Your letter and this response are posted on the HEIR website, along with all other correspondence concerning this matter. The following numbered paragraphs respond to the corresponding paragraphs from your letter of February 20:
1. Your use of the term "misrepresentation" in reference to our analysis of the new regulations is misleading. We have been careful to state clearly our assumptions, and to acknowledge possible factual objections to the analysis. We have publicized in good faith the issues and the abundant hard data that bear on the issues. It is clear that we disagree with both your methods and your conclusions, but in no instance have we misrepresented anything.
Your letter seems to imply that the HEIR web site and our analysis of the cut scores is responsible for the large numbers of home school parents expressing concern. This is flattering, but hardly factual. Our polling data, beginning in the spring of 1997, consistently indicates an overwhelming opposition to the new policy, even before we had begun to analyze the data. In fact, our analysis of the data was done in an attempt to convince the Chancellor's staff that the regulation needed to be modified. If you think we have manipulated the homeschool community, then we suggest that you convene a public hearing, such as that normally required for legitimization of such administrative regulations, so that the Board of Regents can clear the air on this issue once and for all with the benefit of broadly based input from members of the community who are most affected.
2. You state that we have "used the SAT I to predict the percent of students who would pass the various SAT II tests." This statement is factually incorrect. In our analysis, we have considered only the SAT II test results for Georgia students who actually took the SAT II tests, and attempted to estimate how those students would have fared against the new regulations. In that analysis, we have not used any information about correlations between SAT I and SAT II Subject Test scores. As a collateral comment, we have observed that the students who actually took the SAT II tests have SAT I scores over 250 points above the Georgia average. On that basis, we claim that it is a fallacy to assert that the success of those students who actually took the SAT II tests is indicative of the potential performance of the "average" student in Georgia, and even more fallacious to assert that the success of those students is equivalent to the performance of "C" students.
3. You have described the HEIR analysis of the six test score requirement as "particularly faulty." As we acknowledged in our February 4 meeting with you, we used only the currently available cut-score data. Perhaps our analysis could have been more complete if your office had set cut scores for the tests that students are likely to want to take, but you have not done so. While it is unfortunate that the Chancellor has promulgated a regulation without fully specifying its most important details, it hardly passes the straight-face test to permit the Chancellor’s office to wield defects in its own regulation as an excuse for calling HEIR’s analysis "faulty." As we have stated consistently, we urge you to conduct your own analysis of the cut scores. So far as we know, you have declined to do so.
As we noted several times during the February 4 meeting, we understand that there are correlations between the various SAT II Subject Test scores for a particular student. We have encouraged Dr. Fullerton and the ETS experts to provide their own estimate of the percent of the actual SAT II test takers who would have been allowed in the admission pool according to the cut scores your office has set. No such estimate has been forthcoming. In discussions with the ETS statistician, she indicated that the correlations likely would not double the pass rate that we estimated. Unless you intend to follow up on your discussion of the correlations by having ETS compute their own estimate of a pass rate, your claim that our analysis is faulty is nothing more than one individual’s unsupported opinion. Such an opinion is unlikely to be viewed by impartial observers as a rational basis for an administrative agency action that impacts roughly 60,000 Georgia students currently enrolled in home study programs and unaccredited private schools.
Furthermore, given the exceptional ability of the students who actually took the SAT II tests and the fact that the vast majority of them are likely to have graduated from an accredited program, one would naturally assume that virtually 100% of them would have at least a C in the corresponding CPC course. For these students, the pass rate on the individual SAT II tests should be closer to 100% than to 75%.
4. We find anomalous the juxtaposition of your admission that you do not know whether the cut scores are fair and your refusal to consider the data and analyses that we have presented. We do not know how to characterize your summary dismissal, without counter-analysis, of our original analysis of the cut scores.
You state that the Angoff method is "commonly used" yet you did not use the recommendations from the Angoff study. In fact, your decision to arbitrarily adjust the scores downward by one standard error of measurement is a clear acknowledgment that the results of the Angoff study are not valid. Why not adjust the cut scores upward? Anyway, the ETS experts have themselves admitted that the Chancellor’s use of the Angoff method for this purpose is not scientifically valid.
Your assertion that there is no empirical data to support a rational choice of cut scores is incorrect. The cohort of Georgia students who took the SAT II tests provides an excellent case study of the question "How would Georgia public school students succeed if they were required to "pass" the SAT II tests?" Without question, the "average" student would not be nearly so successful as the "very very good" students, who had SAT I averages 250 points above the Georgia average, and who passed these tests at rates "between 72% and 88%".
Until your office produces a statistically valid analysis that contradicts ours, HEIR stands by its analysis of the cut scores. We encourage your office to make a good faith effort at validating the "C equivalency" of these cut scores. We believe that we have made a credible showing that the cut scores arbitrarily chosen by your office are not "C equivalent." Surely with the expertise you have at your disposal, you can determine an objective method that you can use to support or refute this assertion.
Finally, both you and the Chancellor have suggested that there might be an alternative method for admissions. So far as we know, you have not established a process for developing any such method to be employed with respect to homeschoolers, nor have you charged any individual or group within your office to do so despite the fact that USG institutions do employ alternative methods to admit various other categories of "non-traditional" students. So far, you and Dr. Fullerton have rejected out of hand any alternative that we have suggested.
While we are losing hope of finding an administrative resolution to this issue, we continue to be willing to enter meaningful dialogue with your office. If you plan to follow up on this letter, please inform HEIR so that we may make this information available to the homeschooling community in Georgia.
Sincerely,
Leon F. McGinnis, PhD, PE Kurt
Schulzke, JD, CPA
Angela
Paul
President, HEIR
HEIR
BoR Committee
HEIR
BoR Committee
John Sewell
Cindy
Sewell
Vice President, HEIR
Treasurer,
HEIR
Copy: Chancellor Portch Supt. Linda Schrenko
Gov.
Zell Miller Hon. Jeff Williams
Hon.
Roy Barnes Hon. Ron Crews
Mr.
Guy Milner Hon. Dubose Porter
Hon.
Don Balfour Hon. Richard Marable
HEIR
website