Bacon's Rebellion

Sorry, Smaller Class Sizes Don’t Help Disadvantaged Students Either

—

by

James A. Bacon
in
Source: Cranky’s Blog

A few days ago I addressed the issue of class size in K-12 schools, citing a Campbell Collaboration study claiming that the teacher-student ratio is an insignificant variable in influencing academic outcomes. In support of that finding, I displayed data, compiled by our friend John Butcher at Cranky’s blog, confirming that there is essentially zero relationship between average class in a school district and the average Standards of Learning pass rates.

But there was a loophole in the Campbell Collaboration report that could justify spending more money to shrink classroom sizes in some cases. As I summed up that finding, “While the overall effect is negligible, smaller class sizes have been shown to be more beneficial for students from socioeconomically disadvantaged backgrounds.”

So, Cranky ran another analysis. This time he plotted teachers per 1,000 students against average SOL scores for disadvantaged students and non-disadvantaged students. His conclusion, as can be gleaned from the graph above: “These data reach the same result as the overall pass rate data: Divisions with more teachers per student do not, on average, have better pass rates.” Even for disadvantaged students, the correlation is essentially zero.

If we want to improve K-12 educational outcomes, we need to find a better solution than hiring more teachers and shrinking class sizes.

Share this article

ADVERTISEMENT

(comments below)

Comments

Comments

9 responses to “Sorry, Smaller Class Sizes Don’t Help Disadvantaged Students Either”

  1. djrippert Avatar
    djrippert

    If more teacher time per student doesn’t improve results than what does that say about teacher time?

    Fairfax County certainly seems to be misguided on this. They bend over backwards to have low student to teacher ratios in economically disadvantaged schools offset by high student to teach rations in areas of relative wealth.

  2. Lwood Avatar
    Lwood

    About 1972 one of my education professors stated that the best predictor of student success was family background. I taught for 35 years and this was pretty much true.

    1. James A. Bacon Avatar
      James A. Bacon

      John Butcher’s correlation analyses at Cranky’s Blog suggest that about half the variability in SOL performance can be attributed to the socioeconomic composition of the student body. That’s a lot. But it leaves half to be explained by other factors that school systems can control. Teacher quality is a significant factor. I would argue that the ability to maintain classroom discipline in school — partially a function of the teacher’s experience and skill but also of school district policies — is another.

      1. Reed Fawell 3rd Avatar
        Reed Fawell 3rd

        DJ says, “If more teacher time per student doesn’t improve results than what does that say about teacher time?”

        And what else might work?

        Jim – as mentioned, I have been working hard on this, and hope to have something new and fresh on subject by Friday, posted here if its still up.

        PS – hard for an old guy to keep up with all these fine articles the three amigos keep posting up on the board these days. And DJ, congratulations on winning the yellow.

  3. Steve Haner Avatar
    Steve Haner

    But I bet it made a big difference in your life when you had 20 sets of papers to grade rather than 30! Having my wife’s life revolve around the amount of work she has done outside the classroom, outside the contract hours (and thus for free), has created some sympathy for the small-class argument. But I agree that overall the success does not depend on class size – unless hands-on instruction is the goal (which it often is with career training.)

  4. LarrytheG Avatar
    LarrytheG

    It’s an interesting chart but I’d still like to see it done for all the schools within a given school district…. for specific grades in part because if we believe this data – then things like Head Start and Title 1 have no effect either.

    what this chart seems to say is that no matter the expertise of the staff – whether it’s a 30 year veteran teacher or one right out of college – it does not matter – and on that – I’m wary and wondering about the data itself.

    And too bad we can’t get an apples to apples comparison with non-public schools … which might also be informative.

    One

  5. LarrytheG Avatar
    LarrytheG

    I’m looking at where Cranky got his data which seems to be:

    Table 17a of the Superintendent’s Annual Report for Virginia
    Ratio of Pupil to Classroom Teacher Positions – Regular Day School

    School Division Elementary Teaching Positions 2 End-of-Year Average Daily Membership K-7 3 Pupil/Teacher Ratio K-7 Secondary Teaching Positions 4 End-of-Year Average Daily Membership 8-12 5 Pupil/Teacher Ratio 8-12

    sorta gobbly-gook in the limited format that BR supports but what it is – is that they do categorize according to elementary and other levels.

    While I’d not be that surprised for something like history in the 10th grade, I’d be very surprised at reading in the 2nd grade.

    I just don’t think you can glom the data together like this and really get real insight.

    As stated earlier – the class size issue has been questioned and debated in the education community itself for quite a while and we do know that many freshmen college classes can have a hundred of more students in a class as the role of the teacher as the grade level goes up – but at the very early grades k-4, the class size has to be important and if I saw data generated just for those class grades especially on a per school basis that showed the split between non-ED and ED – it would then be more convinced. If I further saw similar data from non-public schools confirming this or actually showing better results with the same or bigger class sizes then I would want to know more ….

    But to this point – it’s a conundrum and the Federal govt actually runs what is called “Title 1” programs for kids who have reading disabilities in the early grades and those teachers have to be specialists in reading and have Masters degrees So if you just read this chart – one might be tempted to conclude that the Title 1 program does not work… and I’m not convinced of that at all and the data shown is not at a granular enough level to really see such differences that I’m pretty sure do exist. One of the problems with Title 1 by the way is that there is not enough funding for enough Title 1 teachers and many school systems don’t come up with the additional resources – in part because they tend to treat all teachers hired as “equal” and not specialists who are paid more.

  6. Lovettsville Lady Avatar
    Lovettsville Lady

    Having studied this issue for decades, there is no study that shows smaller class sizes help to educate children unless the class size is less than 16. Even then there are few studies that show that.

    How to make a difference for kids on the bottom? Teach real math, real English and writing, and real history/Civics. For many decades educrats have believed if they can just lower standards enough, expect less and less of students, dumb teaching down enough, students on the bottom won’t look so bad! Everyone will be equally uneducated! If they don’t teach anyone much of anything there will be no socioeconomic or racial gaps in learning.

    The reality is the middle class (and above) students hire private tutors or go to one the tutoring services on every corner to learn real math, reading/writing, in preparation for standardized testing and SAT. More and more parents are using private schools and homeschooling so their children can get a good education. Students on the bottom, who have parents without resources, stay on the bottom while the other students learn outside of school. Before educrats decided to lower all standards for everyone it was not necessary to be educated outside of school. All students were education IN school. That doesn’t happen today.

    As I often say about liberals, educrats had the best of intentions with the worst of outcomes. Until that abates, until educrats decide that education is the solution for ALL students, the gap will remain and poor children will continue to suffer the most.

  7. Reed Fawell 3rd Avatar
    Reed Fawell 3rd

    Lovettsville Lady:

    You sure do know what you are talking about, going right to heart of the matter. Things like:

    “How to make a difference for kids on the bottom? Teach real math, real English and writing, and real history/Civics. For many decades educrats have believed if they can just lower standards enough, expect less and less of students, dumb teaching down enough, students on the bottom won’t look so bad! Everyone will be equally uneducated! If they don’t teach anyone much of anything there will be no socioeconomic or racial gaps in learning.

    The reality is the middle class (and above) students hire private tutors or go to one the tutoring services on every corner to learn real math, reading/writing, in preparation for standardized testing and SAT. More and more parents are using private schools and homeschooling so their children can get a good education. Students on the bottom, who have parents without resources, stay on the bottom while the other students learn outside of school. Before educrats decided to lower all standards for everyone it was not necessary to be educated outside of school. All students were educated IN school. That doesn’t happen today.”

    One of your keywords here is the word “EDUCRATS.” They are the central culprits, not the teachers. The teachers are abused victims like the kids.

Leave a Reply

ADVERTISEMENT