A quarterly magazine of urban affairs, published by the Manhattan Institute, edited by Brian C. Anderson.
• • • • • • • • • • • • • • •
Sizing Up Classrooms « Back to Story
Showing 27 Comment(s) Subscribe by RSS
Thanks for writing this informative article. As a California high school teacher for the final 10 years of my working career, what you say makes sense to me.
I always cringe when people attempt to support polemics with cherry-picked research articles. Lacking the time to unpack Mr. Sand's methodological and summary assertions about the articles whose conclusions back up his other points (or that he presents as doing such), I have to offer simpler input.
First. I agree that classroom size per se has become a shibboleth in the Ed Biz. But it is not so much the loathed and despised teachers' unions that have pushed this, as the professional departments of education, particularly at public universities.
And if ever there were a professional cadre with the motive to pump up the need for smaller classrooms, it's education departments, which have burgeoned in the past 40 years. These departments have to justify their existence and come up with reasons to increase their enrollments. And given that they are generally among the weakest departments on any campus, with very low standards of admission and attainment, their focus is increasingly on novelty rather than quality. Then the products of these departments get tenured jobs teaching in them...and one has pretty much a self-sustaining mess of the first water.
And thus entire bodies of education research have departed from the hypothesis that smaller classes are better...when any of us who ever taught (especially those of us who don't have Ph.D.s in "pedagogy" or whatever) know that there are different methods for teaching different students in different settings, and any of them can work well IF THE TEACHER WAS TAUGHT HOW TO APPLY THEM.
In other words, the methods that work for smaller class sizes are different than those for larger classes...and guess which ones the ed departments teach?
My experience in very poor and largely black public schools, in a decaying Rust Belt ghetto, taught me that in one respect alone, for some of us, smaller class sizes work better. That is where the teacher function of maintaining order is concerned. When fully 2/3 of the kids in a class have effectively no family life, no home training, no preparedness for learning, no meals, no discipline or structure, poor health, often early life substance abuse problems, and IQs well below average, then 20 in a class is going to be far better than 60 in a class for those of us whose families gave us the culture and habit of learning. A teacher might be able to control four or five acting out boys or girls, and moderate their effects on other children's ability to learn, but triple that, and it becomes a farce.
Mr. Sand also mixes apples and shoe phones. The main reason for Korean and Japanese (and Finnish) educational attainment isn't teaching method or class size or the presence or absence of teachers' unions. It's a strong cultural value on schooling/education, imparted to all kids through strong family cultures.
There was a recent series of items in Finland's main news outlets regarding that nation's outstripping the rest of the world in standardized test performance and other measures of student achievement. It turned out that the reason Finland's kids do so well isn't some magical formula imparted in a Ph.D. ed program, like class size.
The reason--well known to Finns--is that Finnish teachers are well paid, deeply respected, and they stay with children for more than one year. In other words, a child's first grade teacher will also be her second, third, fourth, fifth, and sixth grade teacher. This means that a teacher establishes very personal relationships with each student, and becomes aware of their learning styles and limits.
The American Ed Biz biz--the schools of education--couldn't be farther from this model. In their view, schooling is a factory, each child is a cog, each teacher is a production manager overseeing just one part of the child's life. No wonder that the longer kids stay in school, the worse they do, till so many drop out!
Add to that the American model of Employment Uber Alles--where families are expected to pack up and move for a job every two to five years if they expect to be gainfully employed--and it's clear that our nation has far deeper structural problems that must be addressed before schooling can ever be anything better than warehousing for so many children.
Msknob: Regarding your 9:30 post – Clearly teacher quality is important and via Eric Hanushek, I acknowledge that. But it is important to note that teacher quality and quantity are tied together. As I mentioned, when small class fever gripped California, more and more teachers were hired, thus diluting the pool. Think baseball – as the leagues expanded and teams, and therefore players, were added, the skill level diminished.
Regarding your 9:34 post – I don’t disagree with what you say. I didn’t set out to write an in-depth policy piece comparing cultural attributes and education policies of education systems around the world. I simply tried to make the point that small class size and high student achievement are not related. The fact that the Asian countries you cite are doing good things with larger classes proves my point.
Would you rather have your child in a class of 30 with a very good teacher?
Or would you rather have your child in a class of 15 with a poor to mediocre teacher?
Anyone who loves their child will answer yes to the first question and no to the second.
The smaller the class size, the fewer the children who will have access to good teachers and be stuck in classes with poor to mediocre teachers.
Also, this analysis fails to account for a myriad of elements that contribute to student achievement.
Clearly, student test scores since the 1950s are less than stellar, but Mr. Sand fails to recognize the drastic differences in culture between the US and the Asian nations he cites. Both Korea and Japan embrace education as a valued tradition, and their academic systems reflect the importance of that tradition. Both countries also provide moral instruction as part of their core curriculum, where the concept of character building has become a passé, almost an unmentionable, concept in American education.
One should also consider the difference in the number of days of instruction between the US and Asian countries. Students in Japan, Korea and Taiwan attend around 225 days per year, while we average around 180. Our school days are longer (by an hour), but, in terms of yearly hours, we end up behind Taiwan, slightly ahead of Korea and far in front of Japan. Knowing that teachers in the US spend nearly the first grading period after summer break re-teaching the entire pre-summer curriculum (not to mention time lost re-establishing discipline), Our fall-spring school cycle is a huge obstacle to achievement. More time on task minus the summer hiatus would increase student retention.
The writer wants to push the point that “… bigger classes could benefit some children and the economy,” yet the support he offers to prove that point -- an “impressive” review of the large number of studies on class size – indicates that 72% of the 277 studies found no correlation between class size and student achievement. He uses the same data to emphasize that 13% of the studies indicated a negative impact on achievement in reduced size classes.
The profoundly obvious conclusion quoted from Eric Hanushek -- “If you eliminate the bottom five percent of teachers in terms of effectiveness, or if you replaced five to eight percent of the worst teachers with an average teacher, U.S. achievement would rise to somewhere between Canada and Finland” – also fails to deliver any meat to Sand’s position about class size. Eliminating or replacing weak teachers is indeed the key to stronger student achievement. Debating class size is an insignificant issue.
Are the related ratios teacher/student or student/teacher? They seem to morph around in here.
One thing that people need to realize is that the average student-teacher ratio is just that-- an average. So in my school we are mandated by the Federal gov to have special ed classes of 10 or less. We have one teacher that just has 1 student because of his handicap. So they take the total #students/#teachers and get the average. My classes of geometry students? 36 - 40 They said next year class sizes would be as much as 10 more per class! (It reports that we have 23 per class.)
Next- who decides who the "weaker" teachers are? Have you never had a Principal that disliked you?
Things are getting really bad in the public schools but lets not blame it ALL on the teachers. Here in Tx we have no unions.
In my dual-language school in L.A., the dual language teachers have smaller classes (19 students), and the problem students all get dumped into the English immersion classes. I had 27 students last year with 6 problem students. You have to look at each individual class, NOT just the average size.
So if we take this argument to it's logical conclusion, why not just bump all class sizes to 100 kids, throw in a volunteer from "teach for america", put some tents over the whole thing and a barbed wire around it and solve all our education problems in one go?
We're not taking into account, in our discussion, exactly what kind of class we're imagining.
I think that any class in a foreign language, including those at the university level, would ideally be small (at least, they should have a 'section' with graduate assistant that's small), just to allow for enough individually-generated responses. A lecture class in history might not need the same intimacy.
If we are thinking of young children who need to be taught to read in their own language, again, the more opportunities for individual recitation, the better.
I would have guessed that math (or learning to play a musical instrument) would similarly need to maximize the chances for each member of the class to recite on his own.
The critical commenters profess belief that individualized classroom attention is the key to learning. But the Hanushek study offers unusually comprehensive evidence that class size, and therefore individual time, has no effect on learning.
Those supporting smaller classes do so on the grounds of a handful of studies, even as they dismiss hundreds of other studies with far more consistent results.
The emphasis on individual attention for each student arises from a pedagogy, taught in ed. schools, that seeks to "decenter" classrooms by minimizing top-down discipline, teacher authority, and lecturing. Yet children lacking authority figures in other parts of their lives might benefit even more from teaching styles that teach "self control and personal responsibility" through adherence to discipline in (slightly) larger classrooms, as Terence O'Flanagan phrases it.
The proponents of smaller classes are making a purely emotional case. I don't know what type of evidence DWPittelli is seeking, but dismissing a respected analysis of hundreds of studies while demanding more evidence and offering only "intuition" in return is not a compelling stance.
My teachers were catholic nuns in an elementary school I attended between 1938 and 1946. Our class had seventy kids, half boys and girls. Discipline was effective as was the teaching which rewarded students with more than the basics in grammar, math and history; it taught us self control and personal responsibility. The current plea for small class size is just one more phony issue from an effete educational system populated by incompetent teachers.
Is it counterintuitive to believe that class size doesn't matter much? Let's look at the math:
Based on my memories of elementary school, school started at 8:30 and went to 3, so we were on the grounds for 6.5 hours. Roughly 1 hour of that was taken up by lunch and recess, another hour went to the art/music/PE activities. Realistically, roughly half an hour was taken up with screwing around while going to and from lunch/recess/music/PE, leaving 4 hours for classroom instruction. Of those 4, probably half of them were taken up with general lectures and discussions that would take the same time no matter how many students were in the class. I see 2 hours left for "individual attention." In a class of 20, this would mean 6 minutes per student per day, while in a class of 30, it means 4.5 minutes. I'm not sure that reduction of 1.5 minutes matters so much in the grand scheme of things.
Now admittedly this was 20 years ago, so maybe I'm misremembering, or maybe things have changed to give students far more class time. It's also possible that the 6 minutes a day add up to something significant. However, I don't think it's necessarily counterintuitive to assume that the amount of "individual attention" any student can get is so small that reducing it doesn't actually matter all that much.
This is exactly why we need unions! It's hard to believe that any teacher, union supporter or not, would want more students in a class; especially at at time when teachers are being scrutinized over individual achievement on one week's worth of state testing. It's simple; more students, less teacher resource and less quality time. I love how these people who are not in the trenches can so easily criticize. Oh, and P.E, is not a determinent of state performance. Lucky you!
Noted that class sizes have been shrinking, not growing, over the long run, and that test scores do not seem to be improving, how do you know that things wouldn't be even worse if we still had classes of 30 kids?
exactly why is the burden of proof on those whose position is supprted by Ockham's Razor?
Larry Sand is right, and more than right. In 1946 I was a pupil in a first infant class in an impoverished inner-city area in the post-war UK - Bradford, Yorkshire. Our teacher was Mrs. Barker, who certainly would not come out of a modern B. Ed course with flying colours; she started us, at 9.00 of our first day, by writing an 'a' on the board and getting us all to say the sound and try to copy the letter in chalk on our slates. Stone-age stuff. To make it worse, the class was large: nearer 50 than 40.
Nonetheless, by the Christmas holidays every child could read, at least a little. The class dunce could at least spell out 'The cat sat on the mat' while the class Einsteins were already reading, for pleasure, independently. Every one of us (at our own level, clearly) drew, wrote, signed, addressed and posted a Christmas card for our parents.
Now, with classes of 20 and less, there are a fair few children who cannot read properly when they transfer to secondary school at age 11.
Come back, Mrs. Barker. We need you.
There are some very good points here.
However, you have to be careful how you count the teachers, for the student-teacher ratio.
I expect that a lot of the increase in the number of teachers is accounted for by relatively recent requirements for special ed, where a fully-qualified teacher (also fully qualified as a specialist) is responsible for only a subset of students. Whatever one thinks about the necessity of having such "extra" teachers, it is important to count them as "extra" (i.e., not standing at the front of a self-contained classroom of students).
Surely it's not only from dishonorable motives (like, wanting always to take it easy!) that teachers prefer a smaller class: if one is expected to respond to individual needs ( = differentiated instruction), obviously one can stay on top of all those needs somewhat better if there are fewer students.
The change in class size, from relatively larger ones in the 1950s to smaller ones today, cannot be the only reason for the drop in test scores over that time. The change in family structure, and the type of common culture we all share, where deviancy has definitely been defined down, must have a lot to do with it.
If only it would be possible for us to fire the lowest-achieving teachers, and have them replaced by those who are at least mediocre.
How would that work, in practice? Won't there be strikes, etc., a la Wisconsin, before anything like that can happen?
All of this is almost completely beside the point. The most important factor in education is two involved parents who put a value on education. EVERYTHING else is not only secondary, but almost beside the point. That's why certain ethnic groups outperform others - and by outperform we aren't talking about small differences, the differences are HUGE. Why else do Asian students perform so spectacularly other than a cultural emphasis on education and two involved parents? And why else do the statistics how such a vast difference in achievement between African American children whose parents stay together and those raised with by one parent?
I can speak from the experience of helping raise three children, including one with attention problems - homework - schoolwork - does not come naturally. ParentS need to be there to help children navigate through school, whether it is helping with homework, getting rides to school events, choosing courses, assisting with choosing college, financial aid, and for general encouragement. And as hard as it for for two parents, for one it is an incredible burden - and next to impossible in a culture that does not value educational achievement.
In short, the value of two involved parents dwarfs any conceivable benefit as a result of small classroom size or any other factor.
In an age of political correctness, I find it difficult to even put down that maybe as a society we should rethink our abandonment of the old saying that parents should stay together for the children. That statement has somehow become dated, so that to even advocate such an idea is somehow.... anti-feminist. But facts don't lie and children do much better when parents stay together and share the raising of children.
But, of course, that's not what the article is about - but classroom size is a minor issue compared with the value to children of being raised in a home with two involved parents. Or better yet - being raised in a culture that places a premium on parents staying together AND educating children. In the absence of that all of the nonsense about different teaching methods, rewarding teachers or students for achievement or the latest fad in educating students is just that - nonsense. And yes, even classroom size is relatively unimportant in comparison.
But, we hear little about the value of two parents at home - when it comes to education, this topic is for some reason avoided. That is inexplicable. I shake my head when I hear about "No Child Left Behind" or monetary rewards to teachers or students - these desperate measures are irrelevancies in comparison to the value of two involved parents. And none of these programs even faintly advocate parents staying together for the sake of their children's educations.
And yes, parents sometimes drift apart after children are born - I can hear the arguments
now: "children should not be raised in a home where parents don't love each other." My response is "boo hoo to the parents - you made the decision to bring children into the world so suck up to your responsibilities as parents." And of course, that has to be done on a society level as well. The alternative is that all too often the children of such parents become a burden on society anyway. At the very least such parents deprive society of the adults these children could be, if the parents stayed together and were involved with their children's education. How many great scientists, artists, educators and the like have we lost as a result of welfare being tied to the father NOT being in the home, as well as the encouragement of one parent families? I recall the articles in my daily paper in the late 60's about welfare "cheats' who were receiving welfare even though the father was home.
And while we are talking about teachers unions, isn't there something wrong with allowing such unions to make political donations to the very elected officials that determine their wages and benefits? The donations only go to one political party which in return for such donations, grants wage and benefit concessions. Am I the only person who thinks that this is very dangerous and corrupting? Especially, when you have the elections of the officials in off months, so that the very people who benefit become a powerful voting block?
In short, the whole system is in need of dramatic reform, as is the society that supports it. Unfortunately, much of the media that could call for such reform simply will not criticize the party that benefits from what can only be called corruption on an institutional level.
You have to be kidding me. Research like statistics can be manipulated as I am sure you know and have experience with. Common sense tells me that when you compare a class of 20 and one of 34, the child in the smaller class has almost twice as much chance of having my attention in a small group. I also have almost twice as much time to work with them specifically. I know you have taught for 28 years... a few more than me. How long exactly did you teach in the grades K through 3rd? What kind of pedagogy did you adhere to?
Do you have any idea of how to teach reading? Do you know the difference between a 1st grader who has graphaphonics issues or even syntax issues vs. one who has vocabulary issues, and do you know the difference in teaching strategies you would have to use to best address them? I'm sure you can get a teacher desperate enough for attention to agree with you, but they will be few and far between regardless of whether they agree with their union stance. This issue really isn't about unions. I do not espouse to know about engineering and what it takes to be good about it. I do, however know what it takes to teach elementary students how to read, so please put your emotions and politics away. The fact you quote Dylan in your blog is heresy right?
In the end of your "article" you try to use hyperbole by pretending what you are claiming is about the difference between 1 or 2 students. Well, of course I would rather have a good teacher and 21 kids then a poor one and 20, but you and I both know that is not your original hypothesis. This is not what is happening. We are talking about the difference between 20 and 34 in kindergarten through 3rd grade, not and extra few students.
I do agree that a larger class size in middle school and perhaps upper grades in elementary starts to lose its value compared to other potential resources that money could be spent on. BUT, do not tell me that schools are better off now then they were 15 years ago. It simply is not true. You have to know that. There are too many variables that budget cuts account for to debate here, but I hope you as a former teacher would agree we are certainly not living up to what we could be, and some of that happens to be related to the amount of funds the state is putting towards education.
I agree with the idea that class size by itself cannot be blamed for scores staying flat, without also considering whether there are other factors at work. In other words, if we could say with the data, "all other things being equal, smaller class sizes have not improved proficiency," then we could look at increasing class size more clearly. What Mr. Sand is talking about, though, is just increasing classes by 1 student, on average. This would enable schools to get rid of the bottom performing teachers. The problem is there's no mechanism now for schools to get rid of bottom performers. As DWPittelli said, the teachers with the least seniority, who tend to be the better teachers, will be let go first.
What Sand is saying, though, is that, by the same virtue, it cannot be said that smaller class sizes have *improved* the situation. Class sizes have gotten smaller, but the situation hasn't improved. Busting the myth is the first step in looking at the matter differently. It's not the answer in and of itself.
California's student-to-teacher ratio is 21.3. Average class size is 25.4. That's because not all teachers are teaching a class.
Ten years later and the conclusion remains unchanged. See http://hfienberg.com/clips/sage1.htm and http://hfienberg.com/clips/nctimes.htm ....
I find this article unpersuasive. I think that strong (counterintuitive) claims should be backed up with strong evidence.
1) Noted that class sizes have been shrinking, not growing, over the long run, and that test scores do not seem to be improving, how do you know that things wouldn't be even worse if we still had classes of 30 kids?
2) Have any of these studies looked at a random assignment? (Certainly comparing us to Japan is pretty useless.)
3) What makes you think that if we fired 5% of teachers, it's the worst 5% who would be let go? I think we all know that it's the 5% with the least seniority who would be laid off. And if we somehow moved to a merit system and got rid of the dead wood, then any improvement wouldn't really be due to larger class sizes.
People can learn second languages in classes of any size. Or, did you neglect to add a "better" in there somewhere? Additionally, people learn second languages best by cultural immersion, by interacting with people that speak the chosen language. That's quite besides the point, though. We should stop teaching all second languages except Latin.
He may have a point in this article, however you must ask yourself this; if this was your little girl or boy and you CANNOT afford private school, would you rather your child be in a smaller classroom, or a classroom with 30 + other kids ?
I believe there is evidence that people learn second languages in smaller classes. I'm not so sure about history, literature, etc.