|
Ringing the School Bell-Curve
By William C. Carlotti, Vermont
Sara Humphrey’s recent article [The Truth About Standardized Tests, February 21] accurately describes some of the consequences of using standardized tests and succinctly states that "their purpose is to rank." I would wholeheartedly agree with her assessment and would embellish it by saying that the historical and continuing intent of the tests has been and is to discriminate, stratify and segregate. Who the tests discriminate against has varied over time, but has always included the African-American and, according to the time of their use, also women, southern and eastern Europeans, native Americans and, most importantly, always the poor and moderate-income families of whatever ethnicity or "race."
The demographic disparities in the responses to standardized test questions were used to create the 1921 and 1924 immigration laws that deliberately restricted the immigrants from southern and eastern Europe. Laws that permitted the forced sterilization of individuals identified by their responses to standardized tests were typically passed by the states in the 1930s, and thousands of such forced sterilizations were performed throughout the country, principally relying on the tests. (Prof. Henry F. Perkins, at the University of Vermont, targeted the Abenaki and French Canadians for sterilizations using the tests.) Such test results were also used as part of the means to identify "mentally deficient" individuals for extermination by the Nazis in the 1930s.
There is a great mysticism surrounding standardized tests—not the least of which involves the enormous secrecy, comparable only to the secrecy of matters involving national security—that engulfs the means and methods that are used to select the form and the content of the questions. But there are clues regarding what is being done to create the tests because of our knowledge of similar activities, in spite of the secrecy of the Educational Testing Service, which produces the Scholastic Aptitude Tests and some 140 other standardized tests.
For instance, in the taking of polls, the form of the question and the content of the question are understood to be critical to the outcome of the poll. This is so much the reality that there are now descriptions of the entity that has commissioned a poll, because there is a realization that it is possible to spin the results to support or to oppose some particular point of view by how and what questions are asked. Pollsters are described as Republican or Democratic, conservative or liberal, etc.—because of the outcomes that their polls routinely support or oppose.
Further, having watched the returns on the evenings of elections, we have seen that the Big Media Networks are able to predict the outcomes of elections with an almost 100% accuracy after they have obtained responses to the exit polls of voters who have already cast their votes. Based on the historical voting pattern and demographic composition (age, income, ethnicity, etc.) of the districts where the exit polls are taken, the Big Media Networks accurately predict the election results using a 1% response to their exit polls.
Now, if we consider the process of the SATs, MCATs, LSATs, etc., every one of the applications to take the tests contains a detailed demographic profile of the applicant (i.e. age, income, religion, ethnicity, locale, etc.). What is also very important is the fact that one-third of the questions on every one of the tests are experimental questions for the exclusive use of the E.T.S., and the responses are not included in the test score of the applicant.
The consequence of this is that the E.T.S. not only knows the specific demographic profile of every response to a particular form and content of the questions whose results are included in the scores, but they also have the same information regarding the responses to the "experimental" questions that may be used in future tests. The information is backed up by millions of responses on each test, and is further backed up by tens of millions of historical responses. The tests, therefore, can be seen as no more than polls with a self-fulfilling prophecy where the disparities of responses (i.e., in the test scores) evident in demographic comparisons are so subtly built into the form and content of the questions that they can only be identified by the statistical responses. It is these prefabricated test scores processed into a statistical bell-shaped curve that have been used by every one of the challengers to the affirmative action programs, including the DeFunis, Bakke, and now the University of Michigan cases.
The Educational Testing Service sells results according to the requirements of any private or governmental buyer (including to the Department of Defense, the Central Intelligence Agency, and other branches of the government)—even to the extent of supplying tests when the buyer requests exclusion of specified categories in the demographic profile. Its income for the year 2002 was over $600 million. The Educational Testing Service develops and annually administers more than 11 million tests worldwide on behalf of clients in education, government and business in 181 countries.
What is especially interesting about the Educational Testing Service is that since the passing of the Bush-sponsored "No Child Left Behind Act of 2002" it has created a new division to handle the expected boom in K-12 test requirements that they project will provide them with an additional $100 million in income. Another interesting development is that E.T.S. has entered into a special agreement with Jeb Bush’s state government to provide online educational and career planning services for Florida residents.
In spite of Ms. Humphrey’s and many other objections to the standardized tests (there are some thousands of them on the Internet), unless we are able to develop an organized response to their use to match the organized lobbying of the companies that sell them, I’m convinced that the "science" of discrimination, stratification and segregation will continue to be ingrained in the United States' educational/ societal landscape.
|
|
