Teacher quality trumps class size

Would you prefer your child to be in small class taught by a mediocre teacher or a slightly larger class taught by an excellent teacher? Small class size is overrated, writes Larry Sand, president of the California Teachers Empowerment Network, in City Journal.

. . . since the mid-1950s, the U.S. student population has increased by 60 percent, while the number of public education workers, including teachers, administrators, and other non-certificated staff, has exploded by 300 percent.

. . . What’s more, according to the National Center for Education Statistics, teacher-pupil ratios across the nation have diminished steadily since 1955, when the ratio of public school teachers to students was 26.9 to one. By 1970, the ratio was 22.3 to one. And by 2007, the last year for which federal government statistics are available, the ratio came down to 15.5 to one.

Tennessee’s STAR experiment found lasting benefits, especially for black students, for classes of 14 to 17 students in kindergarten through third grade. Most of the gains appear to have occurred in kindergarten and first grade. 

Other studies have found no achievement gains in smaller classes, Sand writes.

In 1998, economist Eric Hanushek analyzed 277 class-size studies: 15 percent found achievement improved, 72 percent found no effect and 13 percent found reducing class size reduced achievement.

When California paid schools to cut K-3 classes to 20 students, suburban districts were able to hire good teachers to teach the additional classes. Inner-city schools made do with anyone they could find.  As a result, a RAND analysis found class-size reduction had no benefit for urban students.

If districts fired the lowest-performing 5 percent of teachers without hiring replacements, class sizes would rise only slightly, Sand writes. The savings could be used for “increased salaries, books, computers, or whatever the individual school district chooses.”

About Joanne


  1. SuperSub says:

    A couple of thoughts…

    Of course, bad teachers make all students suffer no matter the class size.

    Teachers favor smaller classes because it reduces paperwork and makes classroom management easier.

    Teacher to student ratios are horribly misleading because they often (at least in my experience) look at the whole school ratio as opposed to averaging mainstream classrooms, so all the SpEd and Alternative teachers pull down the results.

  2. Another non-educator with all the answers, from a publication owned by a right-wing think tank.

    This article is a who’s-who of right wing nut-job research including Caroline Hoxby, who has been ridiculed in the past for shoddy work, and the “Wal-Mart schoar” Jay Mathews.

  3. I stand corrected, Larry Sand is a retired teacher.

  4. CarolineSF says:

    I think you mean Jay Greene, Mike… Jay Mathews is the (just-retiring) Washington Post/Newsweek ed columnist.

    Regarding small class size — every high-end private school I know of touts its small classes as a primary asset and marketing feature. That includes Bill Gates’ alma mater — even though Gates is one of the non-educator, unqualified voices disdaining small classes.

    When it comes to their own kids, I have a lot of confidence in the financial decisions made by high-net-worth families. It’s very expensive to maintain small classes and contributes to much higher tuition at high-end private schools. If wealthy families believe it’s worth spending considerably more money for smaller classes, I think we should be listening to them.

  5. I grew up in a small district where there were typically 15-18 kids per class. Some electives were smaller (the smallest was the combined 3rd/4th year Latin class that had only 6). Some teachers who had an excellent reputation tended to have larger class sizes (the largest was my AP Bio class with 27) because everyone wanted to enroll in their class. I didn’t find class size mattered in that district.

    That said, I think it is absolutely *NUTS* that the district in which we live now crams 31 students into primary grade classes, including special ed and English Language Learners.

  6. Stuart Buck says:

    If wealthy families believe it’s worth spending considerably more money for smaller classes, I think we should be listening to them.

    If wealthy families think it’s worth buying their 16-year-olds a Mercedes, we should spend public dollars doing the same for everyone.

    Hmm, well, maybe not. Maybe wealthy people buy lots of things for their kids that are just luxury items, that the rest of us can’t afford and don’t really need anyway.

    So you need a better argument than “rich people do it.”

    (You also need to show some ability to think one step ahead — even if smaller classes are the best thing ever when a few rich people do it and are able to snap up some wonderful teachers, think about whether there is a limitless supply of wonderful teachers such that we can decrease class size everywhere without having to hire a bunch of people who really shouldn’t be teaching at all.)

  7. Thanks for the correction CarolineSF, I always seemed to get them confused, especially early in the morning before I’ve had coffee:)

  8. Cranberry says:

    It’s not so simple. The choice isn’t between small classes with ineffective teachers and slightly larger classes with superior teachers. When a school district cuts staff, many students will end up in much larger classes with inferior teachers.

    Our school system counts special educators as teachers. The official, district-wide teacher-to-student ratio does not translate directly into class size ratios.

    No high-end private school I know of has a unionized workforce. Pat Bassett, the head of NAIS, has been trying to encourage private schools to consider increasing class size.

  9. I find that my class size doesn’t matter nearly as much as the composition of the class. 2-3 students who have no interest in school, no family support, and a history of disruptive behavior will bring down any class. However, it is much easier to manage and mitigate the effects of those 2-3 in class of 20, then 3-5 in class of 30.

  10. palisadesk says:

    Hmmm. I see that Larry Sand (probably deliberately) equates “class size” with “student-teacher ratio.” These are not the same thing, and I’m sure he knows that. It appears to me that his argument is disingenuous at best, dishonest at worst.

    Class size, per se, is not the important variable. Class size for whom? For what instruction? Class size for multiply-handicapped children must be smaller (for safety’s sake, not merely instructional efficacy) than for a secondary honors history class. Effective intervention for struggling readers may require a grouping of 5 or fewer, while “average” students will do fine in a class of 20.

    Full inclusion also changes the dynamic. While an “effective teacher” can do better than an “ineffective teacher” with a full inclusion classroom of 30 5th graders (say), there is no way that the lowest-performing (or highest-performing) students will be adequately served with this model.

    When discussing class size, it never is a stand-alone variable. The who and the what also figure into the equation. Who are the students and what are they supposed to be learning?

    Mr. Sand fails to address those issues. More nuanced treatment of the topic, please.

  11. The high-end private schools tend not to hire graduates of the local state college ed schools, either. They also have the ability to kick out disruptive students, require parents to either volunteer so many hours per month or buy their way out of that commitment, typically require the wearing of a uniform (with no parental opt-out), and so on and so forth. All of these policies would likely improve government-run schools if we permitted their implementation, but we lack the political will to do so.

  12. I’ve had class sizes that varied. My Honors Physics in 1st semester had 31 kids – it actually worked, as they had relatively good behavior and self-control, as well as high personal motivation to do well.

    I’ve had classes of less than 10 that were a nightmare – bad behavior, little response by admins to referrals, no parent contacts (if they didn’t respond, the admins didn’t kick their kid out – MAJOR incentive to ignore contact with school).

    All things being equal, the lowest-level kids should have the smallest classes. But, too often, the Honors, AP, and G & T kids get that.

  13. Richard Aubrey says:

    Ref upscale families: My wife was talking to the caterer who ran a small shower for our niece. The lady said she’d gotten a call from a mom who wanted a cake for about one hundred kids. How much would that be? About $300. “Oh,” said the lady, “I have a lot of money and I was thinking of $1000. I have to keep up with my sister.”
    While many private schools give good educations, they also provide a brand. Look at where we are nationally and figure if the Ivies give us good people for government or just an obsolete brand.
    Sometimes the brand is the issue, and worth–some would think–the money. Especially if it’s better than your brand. Or at least costs more.

  14. I have a friend who lives in the school district that includes New Trier High School in Winnetka, IL. Best high school in the state, according to test scores; possibly the best in the Midwest. BUT, she says that much of the money spent there (and it is a massive amount) is wasted; the kids would do just as well without all the bells and whistles.

  15. I will be teaching at a private Catholic school this fall. My science classes will have 25-27 students each. I’m not sure if this fits Caroline’s definition of private school class size. While certainly not the most expensive private school in this metro area, at $7,100 per child it’s not cheap. People send their kids here because it has a good reputation and they want Catholic education, not because of small class sizes.