Network News

X My Profile
View More Activity

Posted at 6:00 AM ET, 09/13/2010

Class size DOES matter after all

By Valerie Strauss

This is a response by veteran teacher Larry Ferlazzo to a post I published last week by Justin Snider of the Hechinger Report. Snider’s piece, which you can find here, argued that the country can no longer afford large class reduction initiatives and that they have not been shown to be effective in any case. Ferlazzo disagrees.

Ferlazzo teaches English at Luther Burbank High School in Sacramento, California. He writes a blog for teachers and is the co-author (with Lorie Hammond) of "Building Parent Engagement In Schools." He is also a member of the Teacher Leaders Network. This piece first appeared on his blog. Read Snider's post, and this piece, and let me know with whom you agree.

By Larry Ferlazzo

“...class-size reduction programs in California and elsewhere – especially Florida – look foolish.”

So says Justin Snider from the Hechinger Report in a guest post at the Washington Post’s Answer Sheet.

I find these kinds of pieces very irritating for a number of reasons (and I am also trying hard to not let my unhappiness with Hechinger’s financing of the research for the Los Angeles Times research on “value-added” teacher assessments get in the way of looking at his piece relatively objectively. The Times used it for their insulting series making teacher “scores” public).

First, I wish all “school reformers” who live in academia would read Corey Bower’s post “First Day Of School: Where Are You?” Corey is a former K-12 teacher who now is at a university and writes:

And yet, I now find myself up in the ivory tower consorting with others who regularly cast stones at the lowly teachers.... And you know what? Most of them couldn’t hack it in the classroom either.

Mr. Snider teaches writing at Columbia University, and says he was a former high school English teacher. I did a little research and, though I couldn’t find out how long he taught in a high school or where he taught, I did find that he was a course examiner for the International Baccalaureate program. [Course examiners for the IB program assess some of the work of students in the courses.]

Well, I have a large IB class, and it works quite well. Of course, these are just about the most motivated and disciplined students you’re going to find anywhere.

I’m very confident in my ability as a teacher. But if you gave me the same number of students for my mainstream ninth-grade English class in an urban school, I’d go nuts (and I don’t think it would be a great experience for the students, either). I can’t imagine what it would be like with first-graders.

Finally, I know some say the research questioning class size’s role on academic achievement is very convincing, but when I googled “research on class size” I found most of the research to be pretty positive. It would also be important to note Bower’s observation on class size research:

I think one of the problems with class size research is that there isn’t a whole lot of variation in class size across most schools, or after implementing most policies. Let’s say a district decreased class size from 26 to 24 — would we expect a huge, and easily measurable difference? Probably not. And yet researchers are trying to quantify these differences and finding out that there’s not much there. If, on the other hand, we reduce class sizes from 25 to 15, we would expect differences to appear. The Tennessee STAR study remains the most rigorous evaluation of large differences in class size and found large, positive effects of changes of this magnitude.

So, what do you think? Does class size matter? Or do you agree with Mr. Snider that it’s “a luxury ... we can no longer afford”?


Follow my blog every day by bookmarking And for admissions advice, college news and links to campus papers, please check out our Higher Education page at Bookmark it!

By Valerie Strauss  | September 13, 2010; 6:00 AM ET
Categories:  Class size, Guest Bloggers, Larry Ferlazzo  | Tags:  STAR and class size, class size, class size and tennessee, class size matters, justin snider, larry ferlazzo  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: If Rhee leaves ... if Rhee stays
Next: Student loan default rates increase -- especially at for-profit schools


This response seems particularly weak. "Here's some anecdotal evidence from my own experiences and here's my opinion of some stuff I Googled." If one of my students gave me a piece of writing with evidence this flimsy, I would give them back their paper and tell them to start from scratch.

Posted by: cc1221 | September 13, 2010 7:07 AM | Report abuse

People certainly do care about that type of anecdotal experience when it is their child in a large class.

The red flag should always be raised when we begin taking anything from people unable to defend themselves. When we take what we know to be a quality educational experiences from children we may want to re-evaluate the real cost:benefit relationship. Better to cut somewhere that may not cost us in the long run and where all involved are able to articulate their preferences easily.

Posted by: pennychristensen | September 13, 2010 7:26 AM | Report abuse

If I may do as I did on Larry's own blog, I'd like to point at a piece I did a decade ago, called Among School Students, Class Size Matters
It was written for the now defunct Prince George's Journal, and I think what I offer is relevant to Larry's post.

Simply put, if class size is not important, why do elite private schools advertise in this paper's education extra about their class size, why are wealthy parents so willing to pay for those schools? Why does Princeton Review (and I presume Kaplan) cap the size of their SAT prep classes?

I am in my 16th year of teaching. I have previously taught in industry. The classes I taught in industry were always kept relatively small to be able to give personal attention to the students whose companies were paying hefty fees for the instruction being offered.

Were all I to do is lecture, then the size of the class would not matter. If I want to engage students, including in discussion that helps them deepen their understanding, having almost 40 students in a 45 minute high school social studies class, as I do in my 3 Advanced Placement Government classes this year, is contrary to the best interests of the students. And it means I have to kill myself to read and offer corrections on their written work and get it back by the next class day, especially if you consider I have 3 other, non-AP classes, with another 81 students.

Posted by: teacherken | September 13, 2010 7:57 AM | Report abuse

Can you do work in the workplace with 25 people constantly around you? Neither can students. Bear in mind that most classrooms are not designed for 30 kids and it is very cramped and, in some cases, dangerous.

Posted by: zeptattoo | September 13, 2010 8:28 AM | Report abuse

If you'vre followed this debate in the educational press, what becomes clear is that the question is not "does class size matter?" Clearly it does, especially for low-skilled K-3 students, writing-intensive high school classes, and of course special ed classes. The question now on the table is different: it's "how much does class size matter relataive to other factors?" and there is research that shows that class size changes (particularly those in the middle rather than the margins) don't matter as much as teacher quality. So Larry is right, because he teaches HS students how to write. But at some level the class size debate becomes moot, because we don't have the resources to put every student at every age level in a class of 12 or 15 (and the California experiment showed that very stringent class size limites for younger children led to less-qualified teachers).

Posted by: jane100000 | September 13, 2010 8:28 AM | Report abuse

PS -- I went to a high school with very small class sizes -- Here's the tradeoff: no gym, no auditorium, no instrumental music, no art facilities, no counselors, no interscholatic sports.

Posted by: jane100000 | September 13, 2010 8:32 AM | Report abuse

The smaller the class, the more the kids learn.

Posted by: celestun100 | September 13, 2010 9:41 AM | Report abuse

Research on class size could fall under what Feynman calls "cargo cult science." For an interesting article touching on the subject:

Classroom Research and Cargo Cults
by E. Donald Hirsch Jr.
All the trappings of scholarship, missing only real results

Certainly class size matters. A quick glance at certain military training classes limited to small numbers of trainees is such since mastery is essential. Quality instruction matters.

If larger classes are implemented in schools, what assurances are there so that students can still have their questions/concerns adequately addressed? Time for tutoring? And as others have clearly noted, certainly essays take time to properly evaluate/grade.

Streamlined teaching to only tested materials may shoot up test scores (multiple choice, T/F style), but not cultivate authentic learning. So, in that regard, class size won't matter as much; you may just need a well-programmed robotic teacher to provide instruction.

Kids aren't cattle and teachers are herdsmen.

Posted by: shadwell1 | September 13, 2010 11:07 AM | Report abuse

As long as the debate continues to hinge on average or mean test score results, researchers will keep finding little or no effect by reducing class class size from 26 to 24. But any classroom teacher can tell you the difference between teaching in a classroom where every kid has a desk and those, like at Fairfax High in L.A., where the L.A. Times reports class size has swelled to 50 ( and that kids are sitting on the floor, on window sills and file cabinets. Kid number 25 or 26, who don't get the kind of personal attention he or she needs and drops out, may even have an upward effect on the scores, but their loss is still a black mark on our society's commitment to public education.

Posted by: MickeyK | September 13, 2010 11:41 AM | Report abuse

shadwell1's comments about small classes in the military where mastery is essential reminds me of a column that appeared in the Washington Post many years ago. (Apologies to the columnist for not remembering his name.) The paper had just installed a new computer system, and the columnist described all the help the paper was giving the reporters to learn to use it: detailed instruction manuals, experts on call in the newsroom, etc. He pointed out that eventually every reporter would learn to use it, no matter how much the paper had to spend on the support system because the Post NEEDED them to know and couldn't function if they didn't learn. He also pointed out that in most school classes, the students got one or two explanations, a little practice, and then the entire class went on to the next exercise, no matter how many were lost.

In other words, if we really thought algebra, chemistry, clear writing, etc., was important, we as a society would reorganize our schools to make sure every student learned these subjects--included making sure classes were small enough that each student could get all the help he needed.

Posted by: sideswiththekids | September 13, 2010 12:02 PM | Report abuse

I'm trotting out my "Old Woman in the Shoe" analogy again:

If it was known that a single woman/man was trying to raise 25 kids in a small house with no help, that person would probably be reported to social services and the children taken away from him. Yet US society thinks it's just fine for 1 teacher to address the educational needs of 25+ students in one room; some of whom have special needs, are behind, or don't even know English.


Posted by: PLMichaelsArtist-at-Large | September 13, 2010 12:02 PM | Report abuse

I would encourage those in favor of smaller classes to do a bit of reading on something called Baumol's cost disease.

This economic phenomena describes quite aptly what happens when labor force productivity gains in one area begin to fall substantially behind productivity gains in other industries.

By moving to smaller classes, we are (economically speaking) arguing for a reduction in teacher productivity. Over time, absent some significant change in education (on-line learning?) the results are pretty predictable:

- a reduction in teacher pay as compared to other professional fields

- a reduction in teacher quality as compared to current teacher standards

- a general decline in educational quality

Posted by: FYIColumbiaMD | September 13, 2010 2:36 PM | Report abuse

FYIColumbiaMD, on what planet does your economic phenomenon have any meaningful bearing on the relationships between teachers and students that are at the heart of the teaching and learning process?

Certainly not this one.

Posted by: TeacherSabrina | September 13, 2010 3:54 PM | Report abuse

The quality of a student's education depends on the quality of their interactions with their fellow students and their teachers. Those interactions become fewer, and degrade in quality, the more students you cram into a room.

Posted by: TeacherSabrina | September 13, 2010 3:58 PM | Report abuse

'FYIColumbiaMD, on what planet does your economic phenomenon have any meaningful bearing on the relationships between teachers and students that are at the heart of the teaching and learning process?'

As long as we can find an endless supply of independently wealthy, well educated individuals interested in teaching for free or near to it, we can afford to apply your model.

If teachers expect to be paid competitively with other professionals with college degrees, then the economics tend to matter...

Posted by: FYIColumbiaMD | September 13, 2010 5:32 PM | Report abuse

One year I monitored the gains of my middle school classes on local assessment tests throughout the year. My classes of 20-25 students increased their average scores gradually by 4% over the course of the year.

HOWEVER, that same year, I was lucky enough to get two very small classes of 15. Those classes saw a steady increase in their average test scores of 12% per quarter. Same test. Same level of kids. Different class sizes.

I think class sizes makes a big difference for several reasons:
-More time to conference with individuals
-More written feedback (graded work)
-Stronger relationships with the kids
-Fewer classroom distractions

It's not rocket science. Anyone who has been in a classroom of 25 vs a class of 15 would know which is better.

Posted by: lhughe1 | September 13, 2010 6:21 PM | Report abuse

'It's not rocket science. Anyone who has been in a classroom of 25 vs a class of 15 would know which is better.'

That, unfortunately, is the wrong question.

The question is 'With a fixed budget what is the best investment that we can make to improve education?'

If we believed that class size was paramount, we could easily cut class sizes in half and teacher salaries in half and solve the problem.

But then, people would point out that we need high quality teachers - so we could double teacher pay and double class size.

And the argument repeats.

Show research that demonstrates that investment in reducing class size is a better decision than investing in higher quality teachers (or the other myriad of options that exist to improve overall education) and then perhaps it will be worth listening...

In trying to do so, you will probably demonstrate the workings of the economic phenomena of Baumol's cost disease.

Posted by: FYIColumbiaMD | September 13, 2010 7:04 PM | Report abuse

I teach biology for MCPS- a course with an end of the year HSA. This year class sizes increased and I have 33 students in a late afternoon biology class. It's chaos. The scores are lower, the students get less attention, and the behavior problems are worse. On the other hand, my first class of the day has 25 or so students and is the exact opposite.

Perhaps we don't need smaller class sizes across the board but it makes a noticeable difference when dealing with students who have come to believe that they aren't good at school!

Posted by: MCPSteacher | September 13, 2010 8:03 PM | Report abuse


I presented the data I found in my own classroom that supports class size as a worthwhile investment in education. But we teachers can't fund the research we need to promote such a change.

And you're only looking at the numbers. I highly doubt that ANYONE would suggest doubling both pay and class size.

Perhaps if our country placed a higher value on education and scholars and scientists instead of athletes, Hollywood, or the music industry, we'd find the money necessary to invest in the most efficient education system.

Posted by: lhughe1 | September 13, 2010 8:45 PM | Report abuse

I think some people here are missing the point...
Lets take the 33 student bio class. You might have lets say 60% of your students pass the HSA. That isn't good.

If class sizes were cut in half, and you get your pass rate up to 80%, that's better, right?

Maybe... if the teacher that got the other half of your kids gives out worksheets and reads the internet all day long, and his pass rate is 30%, then of that original 33, only 55% of the students have passed. This is clearly a less desirable result.

Posted by: someguy100 | September 13, 2010 8:50 PM | Report abuse

I teach biology for MCPS- a course with an end of the year HSA. This year class sizes increased and I have 33 students in a late afternoon biology class. It's chaos. The scores are lower, the students get less attention, and the behavior problems are worse. On the other hand, my first class of the day has 25 or so students and is the exact opposite.

Perhaps we don't need smaller class sizes across the board but it makes a noticeable difference when dealing with students who have come to believe that they aren't good at school!

Posted by: MCPSteacher | September 13, 2010 8:51 PM | Report abuse

'I presented the data I found in my own classroom that supports class size as a worthwhile investment in education.'

Against what possible metric?

As a rough order of magnitude, lowering average class size from 25 to 15 represents somewhere on the order of a 2x operating cost and likely a significant capital cost increase.

To put it another way, which is the better educational investment - hiring top notch teachers and paying them double the current rates but giving them 25 students per class, or hiring average teachers and paying them the current rate but allowing them to have classes of 15? Both cost about the same - are you really willing to bet the better investment is in lower class sizes?

'And you're only looking at the numbers.'

No, I'm asking what educational investment makes the most sense. We don't have unlimited dollars (even here in Howard County, MD).

If we decide that class size is the key driver, that necessarily means we need to hire more teachers per pupil. If we need to hire more teachers per pupil, we either need to pay them less or we need to find other cuts in the budget. I have not seen any empirical evidence that suggests that reducing class size is worth the cost of reducing teacher experience / quality or reducing investments in other areas (e.g., technology).

There's no free lunch. If you increase the number of teachers (to reduce class sizes) the money has to come from somewhere - either teacher salaries or other educational initiatives. If you look at the economics, increasing the teacher per pupil ratio and therefore accepting a lower teacher productivity level will over time result in a reduction in the pay and then the quality of the teacher workforce.

Posted by: FYIColumbiaMD | September 13, 2010 9:08 PM | Report abuse


You are making two assumptions in your argument:

1=Higher salaries attract top-notch teachers.

No one who gets into teaching does it for the money. The hours, workload, and stress don't pay out. GOOD teachers do it for the intrinsic rewards. Higher salaries just attract greedy people. And before you argue that higher salaries are what draw educated/experienced teachers, that's not a perfect correlation. Some of the highest paid and most experienced teachers can also be some of the laziest teachers you'll meet.

2=That the budget is fixed, and something has to be given up in order for something to be gained.

I understand the concept of economics and a balanced budget. In order to make so drastic a change as dropping class sizes by 10 students, school systems may have to find additional funding outside the state and county level. Whether we find the money from the federal government or private industry, to make a change like this would certainly require additional funds. That's obvious. Charter schools, which are currently viewed as competition for the current education system, have looked into making drastic changes with the help of private investors. If we are going to compete with a global economy where already our students are falling behind those in China, India, and Japan, then our country needs to find more money and resources to pour into education, children, and our nation's future.

Posted by: lhughe1 | September 13, 2010 9:59 PM | Report abuse

'1=Higher salaries attract top-notch teachers. '

On average, higher salaries attract a more qualified talent pool.

When average starting teacher salaries were close to the average starting salary for other college degree majors, there was a higher percentage of teachers with high SAT scores and a higher percentage of teachers from competitive admission colleges. There is considerable empirical evidence to support that overall teacher quality has decreased as teacher salaries have become less competitive.

'2=That the budget is fixed, and something has to be given up in order for something to be gained.'

Overall educational spending as a percentage of GDP increased considerably over the past 50 years and has stabilized since the 80's. Given that on a per pupil basis it still exceeds the rest of the world, it is unlikely that we are going to see substantial increases beyond current levels (especially as compared to GDP).

Over time, a decision to reduce (or even maintain) productivity in one field while others witness productivity increases yields the cost disease that Baumol described. That is what we are seeing with education now - and what we will continue to see absent productivity increase.

Posted by: FYIColumbiaMD | September 13, 2010 10:20 PM | Report abuse

The fact of the matter is that the Star study makes no such claims. Their class size study is limited to grades 4 and below and to a certain segment of minority students only. Multiple studies since maintain that in most cases (and there are exceptions) class size is not itself a factor in improved learning.

Frank Krasicki

Posted by: krasicki | September 14, 2010 12:08 AM | Report abuse

I have taught high school English for 22 years. I teach both a low ability freshman English class and a college prep senior English class. In my opinion, class size does matter no matter what the ability level. If I have more than 20 students in my college prep English class, I cannot give each student the time needed to coach young writers. Even within tracked classes, a range of abilities exist. Whether I am teaching reading or writing skills, each student has his or her own strengths and weaknesses. Rather than teach to the middle, I must differentiate instruction so each student can reach his or her full potential. Business leaders should be able to accept this comparison - students are like clients; would a business person tell a client, "I don't have time for you. I have to help another client who is having more problems?" Common sense should dictate that students will perform better if given time and attention from their teachers. When class size is too large, there is simply not enough time to help each student.

Posted by: cwalker79 | September 14, 2010 5:42 PM | Report abuse


You made so many excellent points. :)

Posted by: lhughe1 | September 14, 2010 11:47 PM | Report abuse

If the expectation is that teachers merely regurgitate information to students, then class size is irrelevant. That said, if we are going to view students as individuals and differentiate instruction according to their strengths and weaknesses, class size is absolutely essential. As an elementary teacher for more than two decades, I know first hand that the difference made by having a few students absent is substantial making class size reduction worth every penny spent.
But there is a bigger issue that is largely being ignored: administrative bloat in educational spending. Recently I read an article about how spending in California has risen steadily while the actual per pupil has fallen. While trying to find that article again, I came across dozens (in the first three pages of results) of articles about inflated bureaucratic spending. How is it possible to debate spending that directly impacts children while ignoring the bigger and more costly issue of government/educational waste?

Posted by: KinderTeacher | September 18, 2010 3:14 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company