E.D. (Don) Hirsch in this morning’s How to Stop the Drop in Verbal Scores column:
… the best single measure of the overall quality of our primary and secondary schools is the average verbal [SAT] score of 17-year-olds.
Like Hirsch, I am worried about the narrowing of the curriculum that often comes from hypertesting policies, but this is a fairly atrocious bit of illogic. The SAT was not developed to measure “the overall quality of our primary and secondary schools,” and you would be hard-pressed to use one datum (the mean derived scale score on any assessment) to distinguish the possible effects of changing curriculum from all sorts of other things, including the exposure to English of new immigrants (at various ages of when teenagers taking the SAT may have come to the U.S.), the early childhood experiences of different cohorts, etc.
If you read the longer version of Hirsch’s piece that appears on the Core Knowledge Blog this morning (the Times edited down to 550 words) you will see that he is referring to verbal scores on standardized tests at large, not the SAT exclusively. To wit:
“….verbal scores on the standardized tests taken by 17-year-olds may be the closest thing we have to a crystal ball or a canary in a coal mine. Some firm correlations of life chances with verbal skills have been established over many years of research on the large data sets of the National Longitudinal Study of Youth. Verbal scores of 17-year-olds predict the students’ future and our collective future. An ability to read, write, speak and listen competently correlates with a students’ capacity to learn new things readily, to communicate with others, and to work at a job effectively. It predicts their future income levels. As the verbal competency of each new generation declines or stagnates, so too will our general economic effectiveness. The single most urgent need of our schools is to raise our children’s verbal scores.”
http://blog.coreknowledge.org/2011/09/19/educational-reform-slow-but-sure-vs-fast-and-fail/
There’s that, and then there’s the College Board’s response to the drop – and for once I buy their argument: that the drop in scores has to do with the increase in the sheer number of test takers, which tends to lower the averages. And most of the increase has come from students of color, who tend to score lower anyway.
But there is some truth to Hirsc’s observation.
Consider, for example, the NY State Regents Examinations, which were first introduced in 1866 as a way to allocate scarce funds to deserving academies, before there were public high schools.
With these “preliminary” Regents exams (which were graded in Albany at first), the state could approximate what primary schools that they visited were graduating what we would call prepared or qualified students. The academies accepting these students received funds, those accepting all students, and hoping for funds from the state, were denied them. This historical example supports Hirsch in a general way.
The immediate result, of course, was a significant drop in the number of students accepted by the academies.
You are correct. The SAT and ACT were developed to see if one could predict how well soon-to-be high school graduates would do in their first year of college, long before — and quite independently of — any interest in State self-assessment of public school performance. They have not, AFAICT, changed any of their main criteria on the math side other than a one-time “renorming” of the SAT score. No legislation has ever changed the type of exam used or what it tries to measure.
Perhaps this makes those tests more objective. Is it possible that the results support your concerns about the narrowing of the curriculum to meet a very specific State test?
Certainly one could reasonably argue that tests that measure college preparedness measure whether high schools are preparing students for college at a time when many assert that this should be their sole or primary purpose. Nonetheless, I disagree with Hirsch’s assertion for an entirely different reason that I will post separately.
I would argue that the math ACT or SAT score is a better measure of overall quality of a school, although only if everyone took it.
Why? Because I think that math skills depend much less on poorly controlled cultural factors like the presence of books and newspapers in a child’s home. From what I see in our local school system, we teach the mechanics of reading very well but this breaks down when a broader knowledge base is needed to interpret what is being read. Math remains more mechanical at higher levels, but is very sensitive to bad teaching and bad curricula. Math is not immune to active anti-intellectualism, because students can be conditioned to not want to learn anything, but less so than other topics.
I don’t disagree with Robert’s observations (I’ve seen many times the importance of high verbal skills in engineering and the sciences), but verbal scores measure how well a school makes up for limitations in the home environment in addition to how well it teaches those skills in isolation. You see that quite clearly in immigrant populations where there can be a wild mis-match between verbal and math scores.
CC,
What about the origins of CEEB?
According to Wechsler, The Qualified Student: A History of Selective College Admission in America, entrance exams were written, administered and proctored by department heads — and the lack of consistency (and transparency) were big enough problems that, when divisions consolidated the depts, and complaints mounted (not just from the schoolmasters, but against the drudgery involved), and Columbia went to uniform entrance examinations.
I highly recommend Wechsler on this.
Glen
I also highly recommend Harold Wechsler… both his writings and him as a person. Very nice guy.
What about it? I don’t see how that history contradicts my first observation that an exam developed and maintained to measure college readiness — in any sense of the term — is likely superior to state-by-state exams that change over time and are subject to political influence in their content and interpretation. Indeed, it seems to support my point.
I missed this — thought the historical point was about the use of testing: NY State Regents Examinations were first introduced in 1866 as a way to allocate scarce funds to deserving academies, before there were public high schools.