White House, with a good economic idea, misleads public on education

A report released Saturday by the White House (PDF) says that about 300,000 educator jobs have been lost since the recession ended in June 2009. The report is entitled “Investing in Our Future: Returning Teachers to the Classroom,” and that title highlights the president’s political message, as he said in a speech on June 9:

So it should concern everyone that right now — all across America — tens of thousands of teachers are getting laid off. … Think about what that means for our country. When there are fewer teachers in our schools, class sizes start climbing up. Our students start falling behind. And our economy takes a hit.

I’ll let you read the report for yourself if you’d like, because it’s largely a political, partisan comparison between the budget the president wants for the country and the one recently approved in the US House of Representatives but not in the Senate, put forth by Republicans. However, the report asserts some facts, some in the text, some in graphics, which we can easily check:

The national student-teacher ratio increased by 4.6 percent from 2008 to 2010, rolling back all the gains made since 2000. Further layoffs in 2011 and 2012 mean that the student-teacher ratio will continue to increase as we enter the 2012-13 school year. From Florida to Ohio to California, districts have faced teacher shortages, have cut preschool and kindergarten programs, and have shortened the school week and school year.

Student-teacher ratio is rising

Citing data from the National Center for Educational Statistics, also known as the “Common Core Data”—not to be confused with the “Common Core State Standards”—the president’s report uses a line graph to make its point:


The graph may look odd to people who know a little about math or about the presentation of data. We have three values, for 2001, 2002, and 2003, that are all reported as 15.9, yet the line doesn’t go straight across.

Note to the president’s staff: Don’t show data with the same value in the labels on the graph and a different point on the graph itself. This is just a rounding issue (the actual values were 15.891, 15.881, and 15.920). The way you represent the data should not cause people to pause in order to try to figure out why points that should be plotted at the same level on the y axis are not. We just checked the source data, and we can confirm that the graph is correct, and the labels on the points are correct as well, at least in terms of rounding to the nearest tenth of a student per teacher. However, the numbers on the graph are misleading in that they reflect no change when a change did, in fact, occur.

OK. When we saw how the data were presented and figured that this might be a sign that the president is trying to hide something, we had to keep digging. As for as the claim that it rolls back “all the gains made since 2000,” we need to go back a ways. In 2000, the student:teacher ratio in public elementary and secondary schools was 16.05, while the ratio in 2010 was 15.97. Close enough, we say.

Educators losing their jobs

The data don’t tell us how many teachers were “fired” vs the number that retired or were reduced to part-time. The student:teacher ratio uses what is known as “full-time equivalent” units instead of actually counting the number of human beings doing the teaching. Statistics are usually impersonal, so this shouldn’t surprise you.

The numbers from NCES show that in 2008, the highest number of full-time equivalent teachers, there were a little less than 3.22 million. By 2010, that number had dropped to less than 3.1 million. That’s a loss of about 125,000 teacher jobs. But the claim in the report is not specifically about “teachers.” Rather, it says over 300,000 “education jobs” were lost:

Since the end of the recession in June 2009, the economy lost over 300,000 local education jobs. The loss of education jobs stands in stark contrast to every other recovery in recent years, under Republican and Democratic Administrations.

That includes secretaries at the district office, janitors in the middle school, lunch ladies, and other support staff. Here, we consider this a little “slight of hand” with words. The report starts out talking about the student:teacher ratio and slips right into “education jobs.” Technically irrefutable, especially since the claim is partially backed by a survey conducted recently of school districts showing many of them plan to reduce their workforce in the 2012-13 school year, but a little tricky to read.

Why focus on student:teacher ratio?

Benefits in an actual classroom may not be that great, but one thing a lower student:teacher ratio does is provide more jobs for teachers. Jobs are a good thing, especially right before an election, if you want to win.

There are definitely fewer teachers working in our schools, and the student:teacher ratio is definitely headed back up, wiping out gains it experienced when President George W Bush was in the White House.

Now, here’s the paradox: The student:teacher ratio went down during the Bush presidency, but many people acknowledge no improvement in education during that same period. That begs the question: Is education quality tied to student:teacher ratio?

Nobody can deny that we want to provide jobs in this economy. I would like to see nothing more than everybody working. But there are counterindications that a lower class size will improve education. If the president is trying to convince people to get on board so we can create jobs for people in education, that’s fine. But as with the slight misrepresentation of data in the line graph, the president should say that’s what he’s trying to accomplish, rather than disguising his motives under a veil of preventing our students from “falling behind.” Yes, our economy “takes a hit” when jobs go away, but linking it to students “falling behind” is a bit of a leap.

The report says “a detailed look at the evidence — based on well-designed randomized experiments — confirms that larger class sizes have lasting negative effects: lowering high-school graduation rates, reducing the chance that students take college entrance exams like the ACT or SAT, and lowering the chance of college enrollment and completion.”

If someone can please show me a “well-designed randomized experiment” involving kids who had big class sizes in elementary school and then enrolled in or completed college at a reduced rate, where all other variables except class size have been eliminated, I’m nominating that person for the Nobel Prize in economics. President Obama already has a Nobel Prize, so he doesn’t get to apply for another.

The point is, magic with numbers and statistics after an experiment is performed—or in the case of the literature explicitly cited by the president, after other people have performed an experiment that you are then analyzing—is not the same thing as having a well-designed experiment. As we teach students in middle school, a good experiment varies one parameter at a time, so it can be known that the effect was caused by that parameter. There are literally hundreds of independent variables in any experiment involving students and their educational progress. Let’s not even try to look for experiments that would support the claims made by the president. They don’t exist.

Plus, several experts in the field of education think reducing class size is not as important in producing high-quality education as some other people might tend to believe.

Politicians, like President Obama’s opponent in the upcoming election, Mitt Romney, can point to studies that show, as our national example from the George W Bush presidency seems to suggest, that class size is not an important factor in determining the success of students and their educational outcomes.

Early childhood programs being cut

The report claims that, “From Florida to Ohio to California, districts have … cut preschool and kindergarten programs …” That is true, and we’ve provided sources in each of the state’s explicitly listed in the president’s report. And those aren’t the only places contemplating eliminating or reducing kindergarten and early childhood programs. The report does an excellent job of enumerating districts in several states, though none in Maryland or Illinois, where early childhood programs are being cut or are in jeopardy of being cut or reduced.

Of course, it’s easy to find examples of places in America where kindergarten programs are thriving. I always like to get a plug in there for early childhood programs in my home state.

So, while this is a problem for many Americans and cutting kindergarten is not a good option, the president’s phrasing is a little dramatic. It’s almost as if he’s saying, “From one end of civilization to the other, early childhood programs are in a crisis.” There are some, but it’s not easy to gauge how widespread the problem is. Kids tend to be pretty resilient, though, meaning it’s also possible to look at this as an inconvenience to working parents, who may have to reduce their hours at work in order to take care of their pre-K children at home.

We’ll keep our eye on this problem, however, along with the concept that school districts are reducing the number of hours or days high school students spend in school. These are important, and if what the president says is true, both trends are moving in the wrong direction.

About the Author

Paul Katula
Paul Katula is the executive editor of the Voxitatis Research Foundation, which publishes this blog. For more information, see the About page.