Tag: public school spending

Survey Says: Public Wants to Know the Total Per Pupil Cost of Public Schooling

A new public opinion survey commssioned in Rhode Island by the Friedman Foundation reveals that people want to know the honest-to-goodness total per-pupil cost of public schooling.

Unfortunately, the full cost is regularly omitted from state education department websites, as revealed in a recent Cato study by Jason Bedrick. What’s more, the full figure is seldom reported by the media. Instead, newspapers and local TV news outfits usually report just a portion of the cost that excludes things like construction spending, interest on debt, and pensions. Education officials obviously have an incentive to make their operations look as frugal as possible, so it’s no surprise that they would offer reporters these partial spending figures (known as “operating” or “current” spending).

Public School Spending. There’s a Chart for That!

What better time than back-to-school season to revisit the trends in U.S. student achievement and public school spending? With that thought in mind, I present a newly updated version of my chart showing the total amount spent over the course of a single student’s k-12 career, along with student achievement trends for 17-year-olds. The achievement data come from the Department of Education’s own National Assessment of Educational Progress “Long Term Trends” series, which regularly tests nationally representative samples of U.S. students, drawing from the same pool of questions in use since the tests were first administered around 1970. These are the best data we have on what our kids know by the end of high school and how much it has cost to get them there.

In the past, some readers have wondered if the use of two separate scales ($ on the left and % on the right) might skew the way we perceive these numbers, making the public school productivity collapse look worse than it really is. To allay that concern, I present an alternate version of the chart that places all the data on the same percentage scale. Alas, the second picture is no less bleak than the first.

If music players had suffered the same cost/performance trends we’d all still be lugging around cassette boom boxes, but they’d now cost almost $1,800…. Aren’t you glad we didn’t give tax-funded state monopolies to 19th century Victrola manufacturers?

Census Bureau Confirms: DC Spends $29,409 / pupil

Four years ago, I wrote an op-ed for the Washington Post revealing that DC spent nearly $25,000 per pupil during the 2007-08 school year. I calculated this figure from the public budget documents of the District of Columbia, which I subsequently summarized and linked on this blog.

No education reporter followed up on my findings, and much lower per pupil figures continue to be reported to this day. My $25,000 figure was even greeted with skepticism by analysts at free market think tanks. One state education policy analyst wrote to say that my figure was “out of line with credible information,” and that I gave my critics “too much ammunition with this clearly questionable set of statistics.”

Indeed, the Census Bureau figures for DC’s total K-12 expenditures were substantially lower than mine. I made a note to track down the discrepancy, but other projects intervened. When I updated my calculation to use DC budget estimates for the 2008-09 school year, I found that District spending had risen to over $28,000 / pupil. The comparable number for that year reported by the Bureau of the Census was just $18,181 (which you get by dividing the total expenditure figure in Table 1 by the enrollment figure in Table 15).

So you can see why most folks were skeptical. Skeptical, but wrong.

Back in March of this year I asked my then research intern to contact the Census Bureau and ask where they got their total spending data. It turns out, they got them from a DCPS official. We presented evidence to the Bureau that that DCPS official had missed a few line items when completing the Census Bureau’s forms—to the tune of about $400 million. The Census Bureau agreed and is in the process of obtaining corrected data for the 2008-09 year. In the meantime, they made sure to ask DC officials to include all relevant items when filling out their forms for the 2009-10 school year. The result: Census Bureau data now show DC spent a total of $29,409 per pupil (obtained by dividing total expenditures in Table 1 by enrollment in Table 15). This is just a bit higher than my calculation for the preceding year.

Kudos to the Census Bureau for taking the initiative and getting DC to accurately report its public school expenditures. Now that education reporters can simply open a Census Bureau .pdf file and divide one number by another, I wonder if any will report what DC really spends per pupil? I suspect that they still will not, continuing to mislead the general public, but I would be delighted to be proven wrong.

Oh, and, BTW, this spending figure is about triple what the DC voucher program spends per pupil—and the voucher students have a much higher graduation rate and perform as well or better academically.


This One Is of the Charts

Education professor Sherman Dorn imagines foul play and education policy maven Matthew Ladner is withholding judgment for the time being. Ladner recently made use of some of my charts of the public school productivity collapse, and Dorn has taken issue with one of them, depicted below [from my February 2011 testimony to the House Education and the Workforce Committee].

Actually, the earlier version of the chart Ladner used really did have some incorrect data in the first decade of the spending series [yes, even people who worked at Microsoft sometimes mess up cut and paste], but the corrected February 2011 version also shows the roughly tripling in cost to which Dorn objected, so he would presumably still hold to those objections. Here they are:

First, once I looked at Table 182 from the 2009 Digest of Educational Statistics, it became clear that the cost figure increases (supposedly the total cost of a K-12 education taken by multiplying per-pupil costs by 13) are false. If you look at the columns in the linked data (Table 182), the per-pupil costs when adjusted for inflation approximately double rather than triple as asserted in this figure. Second, there is no possible source for the approximate “0%” line from NAEP long-term trends data, unless there is an additional calculation unexplained by Coulson.

As described in its legend and title, this chart presents the “running 13-yr  (K-12) total spending per pupil” to arrive at the “cost of a k-12 public education” in constant, inflation-adjusted 2010 dollars. For those unfamiliar with the concept of a running total, here’s Wikipedia’s explanation. So for a student graduating in 2009, the running total cost of k-12 education is the sum of average per-pupil spending in 2009 and the preceding 12 years. It is, put another way, the average cost of having sent a child through the public school system, from k through 12. Dorn’s notion that a running total can be calculated by simply multiplying a number by a constant is mistaken, and that seems to be the source of his confusion.

For the class of 2009, the running total adds up to a little over $151,000, which is the final data point making up the blue spending line above. The rest of that line is made up of the corresponding running totals for the preceding years—each one the sum of spending for that year and its preceding 12 years (interpolating missing year data, as noted in the legend).

As for the academic achievement data series, the chart indicates that they represent the “percent change in the performance of 17-year-olds” on the “NAEP Long Term Trends” tests. I’m not sure what difficulty Dorn has with this, since calculating the percent change from an old value to a new one is straightforward. For example, the Long Term Trends NAEP reading score for 17-year-olds in 2008 was 286, and the corresponding score in the first year tested was 285. So the percent change to year 2008 = (286 - 285) / 285 = 0.0035 = 0.35 percent. That is the last data point in the green series in the chart above.

If he’d bothered to ask, I would have been just as happy to explain this to Dorn privately as I am to do so publicly.