by Martin Harris
No long-running experiment in modern American society can match, in clarity of result and (because that unwelcome result has been rejected by almost all modern American education experts) the systematic non-acceptance of findings in evidence, the 40-plus-year-long campaign by the leadership in public education to reduce class size. From the outset, the edu-crats pledged that student achievement would rise directly as class sizes shrank, and so the added instructional expense (“investment”) would be, just like a well-advertised cosmetic product, “worth it”. The Framingham health study and the nationwide smoking campaign have been longer-lasting; they’ve also been validated in broad statistical-epidemiological studies and have therefore enjoyed wide public acceptance; in sharp contrast, since the adoption of the Federal annual student achievement tests since 1969, student scores nationwide have stagnated in the low 200’s-out-of-500 range, with only a few points of gain in selected demographic categories while average class sizes were shrunk, by administration-local-board action, from near-30 in the ‘50’s to half that number today. Of course, per-pupil costs rose dramatically, but the “investment” has yet to show any reasonable return. If it had, the edu-crats’ claims-of-success would have been broadcast everywhere: it hasn’t and therefore they haven’t. For example: in the 2010 National Digest of Educational Statistics, Table 124, you can read that soon-to-graduate 17-year-olds scored 285 in Reading in 1971, 286 in 2008. Most can’t read well enough to make “Proficient.”
The converse is more plausible: since the 50% cut in class size showed no achievement gains –an IBM Chairman wrote in 2008 that “…test scores… are flat or down over the past two decades…”— a much smaller increase in class size, say from 15 to 17 in Tennessee schools should do no harm (a little Hippocratic-Oath lingo, there) and might well do noticeable good in two categories. Nationwide, pupil-teacher ratio (roughly the same as class size) class size was 17 in 1992, and student achievement then was indistinguishable from 1972 and probably from 2012.
The first category is educational: a private-school headmaster explained to your Humble Scribe, decades ago, that students learn more from watching other students’ efforts, less from more-teacher-time-per-student, and so the more chalkboard struggles by others before your turn comes, the better you’ll do. The second is financial: as researcher Eric Hanushek, also decades ago, wrote, reducing class size won’t raise scores of students but will raise costs of instruction. His predictions have proven correct. In terms of productivity results, you might say that money now allocated within the Direct Instruction budget category to enable smaller class size is under-employed, and could be better deployed elsewhere in the State budget. High-profile example: TN Guv Haslam’s expressed wish to end the in-State Hall tax on passive income (to discourage capital flight from the State) and his regret that he can’t “fill” the resulting $172 million revenue hole in the budget. But the $172MM resides, unproductively, as in-State achievement-test scores have shown, in the Direct Instruction part of the TN school budget, where the D.I. cost-per-pupil (using Fall ’08 data) is $4935, and the total D.I. cost is 971,950 enrollment times $4935, or $4.86 Billion. Capturing $172MM or 3.5% would require raising class size by 3.3%, from 15 all the way to –gasp- 15.5. Would raising average class size by half a student harm student achievement? Would ending the Hall Tax help the State economy? Would that be a better job for the presently-under-employed 172 million dollars? We report, you decide.
* * * * *
If you’ve followed the numbers so far, you’ve noted that the first suggestion, a two-student average-class-size increase from 15 to 17, generates far more in Direct Instruction cost savings than the Guv needs to replace the $172 MM, equal to the D.I costs for 34,853, lost through a Hall Tax repeal; a two-student (13%) increase in class size distributed across 64,796 class sections of 15 each cuts D.I. cost of $4935 per student times 129, 592 students to save $635MM in Direct Instruction costs. And, if you’ve followed this discussion-thread in earlier columns, you already know that a few States (Tennessee and Vermont, maybe more) have already adopted class-size guidelines which, if actually put in place, would raise average class size to 20 or more. They haven’t been, and the reason is the political clout of the teacher unions, which recognize that smaller classes haven’t accomplished anything for the interest of students (see the dismissive AFT Prez Al Shanker quote on this subject) but have accomplished quite a lot for the unions in terms of increased teacher membership, higher union dues collected, and, of course, enlarged political clout. But that clout may be shrinking. Consider, for example, the early December comments by NYC Mayor Michael Bloomberg in a speech at the Massachusetts Institute of Technology; in his prescription for public education he advocated doubling class size (back to where it was in his student years) and doubling teacher pay, the increase in productivity of the latter enabling an increase in their rewards without raising costs and taxes. It’s a theme which has similarly been raised in these Humble Scribe columns. Was it merely a tongue-in-cheek rhetorical throw-away? You decide.
If you conclude that Hizzoner was serious, you find yourself in a politically-incorrect minority, dismissed not only by those who argue “the common sense of small-class superiority” (see the website of Left-leaning “SchoolsMatter”, for example, where you can also savor the labeling of charter-schools with low disruptive-student-tolerance as “chain-gang education”) but even more by those who have a dollars-and-power personal interest in ever-smaller classes; the education lobby and their responsive politicians. On your side will be only such serious analysts of the actual results as academic-economists Richard Vedder and Eric Hanushek. There’s room here for only a summary quote from Hanushek’s “The Evidence on Class Size” which has been presented in venues from Washington DC to local school boards, all to no avail. “Existing evidence indicates that achievement for the typical student will be unaffected by instituting the type of class size reductions that have been recently proposed or undertaken. The most noticeable feature of policies to reduce overall class sizes will be a dramatic increase in the costs of schooling, an increase unaccompanied by achievement gains.” That was in 1998. Every year, the next set of NAEP data on class size and test scores proves it again. And that includes the supposed Hanushek-disprover: the Tennessee Star Study.