Where and How K-12 Spending Went Down and Scores Went Up

by Martin Harris

It’s four-score years ago (a little Gettysburg lingo there) that Progressive Supreme Court Justice Louis Brandeis penned his now-famous “States as laboratories for governance experiments” quote, which has since enjoyed admiration across the ideological spectrum. Here’s the first sentence, verbatim: “It is one of the happy incidents of the Federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.” Since then, that depths-of-the-Great-Depression observation has been recited some three dozen times, mostly by Governors eager to use taxpayer money to stimulus-fund a favored sector (think Vermont’s ‘60’s ambition to become “the education State” or Wyoming’s ‘90’s ambition to become “the broadband State”) and occasionally by theoreticians eager to get State Capital help for their pet economic or governance notion (think ethanol-from-corn, an experiment which went Federal before being accused of –unforgiveable—raising consumer food prices, and since then pretty much a non-topic for “green” theorists) and even more occasionally by seeming accident. In this latter category falls the now-four-year-long Arizona K-12 experiment with per-pupil spending reduction. In contrast with the far-more-widespread forty-year-long “experiment” with spending increase, it has already produced some remarkable stats. Vermont edu-crats should take notice.

During the present downturn, State reductions in per-pupil spending have been more than just occasional : data from the Center on Budget and Policy Priorities show 35 States which have reduced State component of per-pupil cost, averaging 44% of the total spending. Thirteen States increased the Capital’s share of their per-pupil total. Arizona’s experiment is unique because a. it’s the largest, percentage wise (down 21.8%); b. it is primarily aimed at reducing instructional cost (on average, near 60% of total spending) by increasing class size; and c. student achievement scores have gone up, not down, during the experiment. In its on-the-Web report, the CBPP waxes gloomy about these cuts –“…serious consequences for the Nation…” and “…scale-back for educational services…” and so on, but no mention of the ensuing actual Math and Reading achievement results, either coincidental or consequential. You can find them on the Arizona SED Web page. They’ve gone up.

Like the absent-minded professor’s experimental “flubber” enabling wheeled vehicles to fly, this wasn’t supposed to happen; ever since the edu-theorist-origin class-size reduction campaign starting in the late 50’s, the edu-dogma has always been that as class sizes shrink, student achievement grows, and vice versa. Even 40 years of Federal achievement tests, pretty consistently showing stagnant numbers in the low-to-mid-200’s out of a possible 500, have failed to silence the dogmatists. Arizona’s scores, like those from Vermont and every other State, fall in the same dismal fewer-than-50%-of-students-Proficient category: what’s noteworthy here is that, as class sizes were increased, a subject not discussed on the SED Web site, achievement scores also increased, and are now Web-discussed in detail. Here they are: “Math: AZ Grade 4 students scored significantly higher in 2011 (235) than in 2009 (230)” and “Reading: AZ Grade 4 students scored higher in 2011 (212) than in 2005 (207)”. We’re not given current class size numbers, but a 23 Oct 12 Wall Street Journal newspiece on the subject carried a photo showing a Grade 2 class with 26. In earlier years, the State average was in the 21 range. Maybe the old Headmaster’s admonition to your Humble Scribe, years ago, was accurate: “students learn from watching their peers succeed (or not) at the chalkboard before their own turn: larger classes mean more learning.” Humble Scribe guess: AZ edu-theorists knew their Brandeis-style class-size experiment would do no achievement harm and might indeed succeed modestly, even if they dared not say so out loud. Small-class large-cost edu-advocates in Vermont, take note.

* * * * *

A four-year experiment carries only a tenth the weight of a 40-year trial, but the AZ findings indirectly confirm the nationwide NAEP achievement-score results: class-size reduction (by nearly 50%) which was expertly promised to produce higher student Proficiency percentages, glaringly didn’t; while class-size-increase (by 5 added to 21, or about 25%) modestly did. Correlation or causation? You decide. While deciding, look at the top three States for ed-spending increase during the same four years that AZ (and 34 others) reduced theirs: North Dakota, up 28.2% (new fracking revenues, probably); Iowa, up 10.6%; and Maryland, up 7.4%. None used the extra “investment” to hire more teachers and reduce class size: all three increased it slightly: ND from 12.3 to 12.7, IA from 13.6 to 13.8, and MD from 14.1 to 15.8, changes which, in standard edu-crat theory, should have damaged test scores slightly. Didn’t happen. In ND, NAEP Grade 4 reading scores went from 226 in 2007 to 226 in 2011; in IA, they went from 243 to 243; and in MD, they went from 225 to 231 (the State which increased class size most showed a Reading gain) while in Vermont, which during the downturn reduced spending 1.3% and increased class size from 10.0 to 11.3, Reading scores went from 228 to 227; Proficiency went from 41% to 41%. Correlation or causation? You decide.

Like Sherlock’s Holmes “curious incident” of the dog that didn’t bark, the US public edu-system is silent about its own Proficiency history: up through the ‘50’s, it produced near-100% P-rates, even amongst slum-poverty and immigrant non-English-speaking groups, and was the global model of accomplishment, with large (in the 30’s) classes, rigorous teacher-classroom control, remarkably low annual per-pupil costs, and, critically, no-social-promotion policies. But today even the best P-State, Massachusetts, gets fewer than half its students to P-status, even though teachers are more credentialed than ever before. Edu-authors rarely write about the 1880-1950 decades of K-12 success, but a brave few are addressing a new and probably-major variable: student peer-pressure for disengagement. Two such are David Steinberg’s “Beyond the Classroom” and Judith Hanna’s “Disruptive School Behavior”. Both marshal stats to show that almost all students could make P-status if they wanted to; most don’t. Although disdain for learning is mostly low Socio-Economic-Status-correlated, both address the one minority, East Asians, which makes the highest P-scores whether the families are poor or not. Why other student peer groups, even at the higher SES levels, now disdain classroom achievement, they don’t address.