E. D. Hirsch, author of several books worthy of mention, provided a column for the page opposite the editorial page of the New York Times on 22 March to which I’d like to call attention. Under the headline “Reading Test Dummies,” Professor Hirsch argues that the problem with contemporary, high-stakes tests isn’t that they test knowledge, but that they test the wrong knowledge.
Professor Hirsch leads by quoting President B. Obama’s expression of concern for developing assessments that “don’t simply measure whether students can fill in a bubble on on test.” Then he suggests that instead of discarding the tests, we should change their content.
These much maligned, fill-in-the-bubble reading tests are technically among the most reliable and valid tests available. The problem is that the reading passages used in these tests are random. They are not aligned with explicit grade-by-grade content standards. Children are asked to read and then answer multiple-choice questions about such topics as taking a hike in the Appalachians even though they’ve never left the sidewalks of New York, nor studied the Appalachians in school.
There is much good in Professor Hirsch’s analysis. He is right that there is nothing really wrong with testing, that an important problem lies in what we test.
But there is something a tad disquieting about the analysis, too. Let me develop this by starting a bit afield with some personal impressions.
My informal observation is that when teachers complain about state-mandated competency tests, they are concerned that the tests focus too heavily on fact knowledge. This seems to be especially true when I talk with people who are concerned with the teaching of social studies, science, and similar content.
Folks espousing these concerns often profess constructivist views and express disdain for testing in general. The melding of these several components—process-over-fact, anti-testing, and constructivism (and perhaps others)—makes it very difficult to parse the argument and have an intellectually clear discussion.
However, given my interests in whether students can read, solve problems, and write clear statements, it is only obvious to me that the we should be able to test competence. It is a relatively simple matter to assess facility in reading aloud, for example. Of course, one would want to have students read aloud from an unfamiliar passage, even though they may have previously encountered nearly all the words in other contexts. Similarly, one would want to assess students’ competence in computation by determining whether they can correctly apply processes to solve novel calculation problems.
Perhaps it is this sort of thinking that, alas, has lead to development of tests that assess novel content in comprehension. Given, however, that there is ample evidence that comprehension is not simply an assembly of strategies (it requires decoding, fluency, and vocabulary as well), constructing novel content must be constrained by an understanding on the test creators’ part that passages should assess what students would reasonably expect to know. One of those constraints should be that the students have, in fact, been taught how to extract meaning from passages.
Professor Hirsch rejects teaching comprehension strategies, however. In part his analysis is correct. As he notes, teachers probably are not teaching reading comprehension. In the late 70s and early 80s Dolores Durkin reported that (a) observations in classrooms, (b) assessments of basal reading manuals, and (c) reviews of textbooks used in teacher preparation all showed that simply giving students passages to read and asking them to answer questions based on those passages was the modal method for teaching comprehension. That’s testing masquerading as teaching.
Major findings included the fact that almost no comprehension instruction was found. The attention that did go to comprehension focused on assessment, which was carried on through teacher questions. (Durkin, 1979, p. 481)
This is where I come back to Professor Hirsch’s column. I agree that it would be beneficial to have tests sensibly aligned with curricula and that those curricula should be predicated on a rational set of goals and objectives. But I hesitate to discard the idea of teaching reading comprehension. I simply think we need to do a much better job of it.
Professor Hirsch ends his column with a good turn of phrase: “We do not need to abandon either the principle of accountability or the fill-in-the-bubble format. Rather we need to move from teaching to the test to tests that are worth teaching to.”
That turn of phrase is just about on the proverbial money, but (thanks to Barb Bateman) I’d like to take it one step higher: “If something is worth testing, then it is worth teaching. If something is worth teaching, then it is worth testing.”
Link to Professor Hirsch’s editorial.
References:
- Durkin, D. (1979). What classroom observations reveal about reading comprehension instruction. Reading Research Quarterly, 14, 481-533.
- Durkin, D. (1981). Reading comprehension instruction in five basal reader series. Reading Research Quarterly, 16, 515-544.
- Durkin, D. (1986). Reading methodology textbooks: Are they helping teachers teach comprehension?. Reading Teacher, 39, 410-417.
0 Responses to “Hirsch on reading tests”