Joanne Jacobs chastises me for jumping on the class size bandwagon. I guess, if the bandwagon started in the 1990s. I wrote a paper in the late 90s that discussed the quantity v. quality problem with President Clinton’s class size reduction policy (I went to work for him anyway, I thought then, and now, that he was right about far more on education than wrong). As I noted in last week’s class size column at TIME the difference between then and now is that we know a great deal more about how class size policies play out in practice today – though you wouldn’t know that from the national debate. Even in the ’90s, though, it was understood that teacher effectiveness mattered more than class size, the Department of Education’s own literature review supporting the class size reduction policy noted this at the time and the Clinton policy was targeted in an effort to address that. Also, one aspect of this that doesn’t get a lot of attention is the idea of strategically lowering class sizes in key subjects, for instance English, to encourage more writing and critical discussion. In addition to smaller classes for academically at-risk students ideas like that make a lot of sense. Unfortunately, we’re not having that sort of conversation because of the back and forth and focus on across-the-board cuts or increases. To wit, class size jihadist Leonie Haimson told The New York Times the other day, “Unfortunately we’ve also seen the rise of a narrative that’s become dominant in education reform that insists that class size doesn’t matter.” Really? Who in the mainstream of the education debate is actually saying that?
Speaking of the 1990s and context, or lack thereof, it’s worth pointing out that test score fraud happened then, too. And before then as well. Yesterday’s USAT examination of the issue raised some important points – and a lot of this does go on – but it left out a little bit of context in terms of the longevity of the problem. This didn’t originate with No Child Left Behind (and the article did include some boilerplate hysterics about the No Child law’s consequences that really don’t reflect the actual policy). For example, here’s Historian Diane Ravitch in 1999 discussing a cheating scandal in New York City: ”To say that tests create cheating is wrong,” Dr. Ravitch said. ”What creates cheating is people who cheat. If we spent as much time teaching kids as showing them the answer, they might have learned to read.”
But, while this does go on, let’s hope examination of it doesn’t fuel the cynicism that seeks to debunk the results every time a teacher or school or school system accomplishes something extraordinary. That happens, too.
10 Replies to “What’s Old Is New…”
There was a time when Ravitch had a brain . . . .
I disagree with the idea presented by Arne Duncan (as quoted on Jacobs’ blog) that class size should be linked to teacher quality. As many people have noted in the comments to the March 3, 2011 “Class Size” post, the variables of STUDENT and SUBJECT are most important in determining appropriate class size. No one can reasonably assert that a great teacher can teach 35 kids in a woodshop class designed for 20 without an increase in the risk of accidents. No one can reasonably assert that a great teacher can effectively teach labs for 35 kids when there’s only equipment and tables for 20 kids. Even a great English teacher can’t grade 175 papers a night.
In my experience as a former teacher, I’d be more concerned with the types of students (special education, honors, ESL) and the type of class (science lab, English class) than the qualities of a teacher.
However, we should be wary of imposing class size caps on certain subjects or populations without looking at their effect on other subjects or students. When I taught in California, English classes for certain grades were capped at lower numbers (I believe 20 per class). Because more teachers weren’t hired in any appreciable amounts, the caps on some classes resulted in the unfortunate effect of VERY large classes for other subjects (history, PE).
I am so glad the press is finally taking a look at cheating on standardized tests, which is probably rampant and frenzied. This is how it’s often done, according to my experience:
The principal of a school encourages his teachers to “familiarize yourself” with the test days before administration. Teachers then look at the items and drill their students on them right before they give the test without proctors. Some teachers do not see this as cheating. One teacher said to me, “How will the kids know how to do it if I don’t teach it?” These teachers may not have taken courses in Testing and Evaluation and so they don’t understand the concept of sampling. However, most are wise enough not to put their pretesting drills in writing (as did the man described in the USA Today article) so there is no proof of irregularities.
If standardized tests are to be used for high-stakes decisions, they must be different from year to year and administered, collected and graded by disinterested parties (similar to administration of SAT). If a district can’t afford to do this, then they can’t afford to give the test.
I don’t care about class size. Give me 100 kids, and appropriate resources — enough desks, a room in which all the kids can see the board or screen, sufficient quantities of textbooks, ability to run sufficient copies, etc.
More important than any of that, though — let me kick kids out who are chronically disruptive. (Before you jump down my throat, of course they can come back. When they quit the nonsense.)
Every child has a right to an education, and when discipline problems are allowed to perpetually negatively affect that right, any teacher will be less effective. 100 students with no discipline problems is a far more productive classroom than 10 students with even one or two discipline problems.
Heck, I’d take the 100 kids even with limited classroom resources, if someone, somewhere, would promise me that discipline would be swift, severe enough to be deterrent, and consistent.
I agree with both Linda and Jennifer (about different topics). Linda makes a good point that allowing teachers who proctor a test to know the contents beforehand is kind of ridiculous, if the test is to be used for any kind of high-stakes consequences.
Jennifer also makes a good point: It’s not the class size as much as the students in a class that make it more or less difficult to teach. In my experience, it was easier to teach 30+ honors students (who had good independent study skills and classroom behavior) than to teach 8 or 10 special education students (or a class with several mainstreamed special education students), particularly if any of the special education students had emotional disabilities. Giving teachers the power to swiftly and decisively eject any disruptive student from the classroom would go a long way…
See the journal, Educational Evaluation and Policy Analysis (Vol 33, No. 1 – March 2011) for an interesting report on a study entitled “The Non-Cognitive Returns to Class Size.”
After calling me a “class size jihadist”, you question my statement that “Unfortunately we’ve also seen the rise of a narrative that’s become dominant in education reform that insists that class size doesn’t matter.”
You write: “Really? Who in the mainstream of the education debate is actually saying that?” Umm, how about the Gates Foundation?
in 2010, Allen Golston, the president of the Gates’ US programs, said to the National Urban League that policymakers had for many years made decisions ”actually based on myths,” including “the myth that class size alone would make a big difference in educational outcomes for students.”
After the Foundation polled DC “insiders”, and found that many of them continued to believe that class size does matter, they reported the results this way: “Washington insiders, like the broader population, continue to worry about class size…and believe the myth that reducing class sizes are a key to reform.”
Clearly, those who run the Gates Foundation and many of the groups they fund are engaged in a mission to convince the American public and policymakers that despite rigorous research on this question, class size doesn’t matter.
I might call their intense focus on this goal a “jihad” only that would minimize the huge institutional and financial influence they have on the educational establishment in this country.
Well, we know that the educational expert Bill Gates Jr tried promoting small high schools and gave up on that when it failed.
Two responses to your necrobump:
1) Rotherham’s post asked who in the mainstream is saying that class size doesn’t matter.
Your response offered evidence of how the Gates Foundation believes class size alone will not fix our schools.
That is not evidence of how they believe class size doesn’t matter, though. It is evidence that they believe it is an insufficient singular change toward the goal of improving outcomes. He even says as much immediately following that quote:
“But it turns out, that class size – without classroom improvements – barely made a difference in student outcomes after the third grade.”
2) That report you reference shows that the polled insiders think teacher effectiveness and academic standards were much more important than class size and that they had misconceptions of actual class sizes:
“Only 12 percent of respondents report that teacher effectiveness is the top priority in education policy, while nearly two-thirds (62 percent) said it should be the top priority. […11% think class size should be the top priority]”
“Respondents overwhelmingly overestimate the average class size, with 82 percent overestimating secondary class size and 92 percent overestimating elementary class size.”
So, no, it’s not as if the report misrepresented their data. There were definite misconceptions about actual class sizes, and strong opinions on which of a list of possible school factors ought to be considered most important.
It is funny to read Chris Smyr here.
Does he believe Leonie is trying to kill thousands of people as Andrew Rotherham suggests?