Education Research

Causality Rules Everything Around Me

In a Slate article published last October, Daniel Engber bemoans the frequently shallow use of the classic warning that “correlation does not imply causation." Mr. Engber argues that the correlation/causation distinction has become so overused in online comments sections and other public fora as to hinder real debate. He also posits that correlation does not mean causation, but “it sure as hell provides a hint," and can “set us down the path toward thinking through the workings of reality."

Correlations are extremely useful, in fact essential, for guiding all kinds of inquiry. And Engber is no doubt correct that the argument is overused in public debates, often in lieu of more substantive comments. But let’s also be clear about something – careless causal inferences likely do more damage to the quality and substance of policy debates on any given day than the misuse of the correlation/causation argument does over the course of months or even years.

We see this in education constantly. For example, mayors and superintendents often claim credit for marginal increases in testing results that coincide with their holding office. The causal leaps here are pretty stunning.

Revisiting The "Best Evidence" Theory Of Charter School Performance

Among the more persistent arguments one hears in the debate over charter schools is that the “best evidence” shows charters are more effective. I have discussed this issue before (as have others), but it seems to come up from time to time, even in mainstream media coverage.

The basic point is that we should essentially dismiss – or at least regard with extreme skepticism - the two dozen or so high-quality “non-experimental” studies, which, on the whole, show modest or no differences in test-based effectiveness between charters and comparable regular public schools. In contrast, “randomized controlled trials” (RCTs), which exploit the random assignment of admission lotteries to control for differences between students, tend to yield positive results. Since, so the story goes, the “gold standard” research shows that charters are superior, we should go with that conclusion.

RCTs, though not without their own limitations, are without question powerful, and there is plenty of subpar charter research out there. That said, however, the “best evidence” argument is not particularly compelling (and it's also a distraction from the positive shift away from obsessing about whether charters do or don't work toward an examination of why). A full discussion of the methodological issues in the charter school literature would be long and burdensome, but it might be helpful to lay out three very basic points to bear in mind when you hear this argument.

Why Nobody Wins In The Education "Research Wars"

** Reprinted here in the Washington Post

In a recent post, Kevin Drum of Mother Jones discusses his growing skepticism about the research behind market-based education reform, and about the claims that supporters of these policies make. He cites a recent Los Angeles Times article, which discusses how, in 2000, the San Jose Unified School District in California instituted a so-called “high expectations” policy requiring all students to pass the courses necessary to attend state universities. The reported percentage of students passing these courses increased quickly, causing the district and many others to declare the policy a success. In 2005, Los Angeles Unified, the nation's second largest district, adopted similar requirements.

For its part, the Times performed its own analysis, and found that the San Jose pass rate was actually no higher in 2011 compared with 2000 (actually, slightly lower for some subgroups), and that the district had overstated its early results by classifying students in a misleading manner. Mr. Drum, reviewing these results, concludes: “It turns out it was all a crock."

In one sense, that's true – the district seems to have reported misleading data. On the other hand, neither San Jose Unified's original evidence (with or without the misclassification) nor the Times analysis is anywhere near sufficient for drawing conclusions - "crock"-based or otherwise - about the effects of this policy. This illustrates the deeper problem here, which is less about one “side” or the other misleading with research, but rather something much more difficult to address: Common misconceptions that impede deciphering good evidence from bad.

Are Charter Schools Better Able To Fire Low-Performing Teachers?

Charter schools, though they comprise a remarkably diverse sector, are quite often subject to broad generalizations. Opponents, for example, promote the characterization of charters as test prep factories, though this is a sweeping claim without empirical support. Another common stereotype is that charter schools exclude students with special needs. It is often (but not always) true that charters serve disproportionately fewer students with disabilities, but the reasons for this are complicated and vary a great deal, and there is certainly no evidence for asserting a widespread campaign of exclusion.

Of course, these types of characterizations, which are also leveled frequently at regular public schools, don't always take the form of criticism. For instance, it is an article of faith among many charter supporters that these schools, thanks to the fact that relatively few are unionized, are better able to aggressively identify and fire low-performing teachers (and, perhaps, retain high performers). Unlike many of the generalizations from both "sides," this one is a bit more amenable to empirical testing.

A recent paper by Joshua Cowen and Marcus Winters, published in the journal Education Finance and Policy, is among the first to take a look, and some of the results might be surprising.

A Few Points About The Instability Of Value-Added Estimates

One of the most frequent criticisms of value-added and other growth models is that they are "unstable" (or, more accurately, modestly stable). For instance, a teacher who is rated highly in one year might very well score toward the middle of the distribution – or even lower – in the next year (see here, here and here, or this accessible review).

Some of this year-to-year variation is “real." A teacher might get better over the course of a year, or might have a personal problem that impedes their job performance. In addition, there could be changes in educational circumstances that are not captured by the models – e.g., a change in school leadership, new instructional policies, etc. However, a great deal of the the recorded variation is actually due to sampling error, or idiosyncrasies in student testing performance. In other words, there is a lot of “purely statistical” imprecision in any given year, and so the scores don’t always “match up” so well between years. As a result, value-added critics, including many teachers, argue that it’s not only unfair to use such error-prone measures for any decisions, but that it’s also bad policy, since we might reward or punish teachers based on estimates that could be completely different the next year.

The concerns underlying these arguments are well-founded (and, often, casually dismissed by supporters and policymakers). At the same time, however, there are a few points about the stability of value-added (or lack thereof) that are frequently ignored or downplayed in our public discourse. All of them are pretty basic and have been noted many times elsewhere, but it might be useful to discuss them very briefly. Three in particular stand out.

When You Hear Claims That Policies Are Working, Read The Fine Print

When I point out that raw changes in state proficiency rates or NAEP scores are not valid evidence that a policy or set of policies is “working," I often get the following response: “Oh Matt, we can’t have a randomized trial or peer-reviewed article for everything. We have to make decisions and conclusions based on imperfect information sometimes."

This statement is obviously true. In this case, however, it's also a straw man. There’s a huge middle ground between the highest-quality research and the kind of speculation that often drives our education debate. I’m not saying we always need experiments or highly complex analyses to guide policy decisions (though, in general, these are always preferred and sometimes required). The point, rather, is that we shouldn’t draw conclusions based on evidence that doesn't support those conclusions.

This, unfortunately, happens all the time. In fact, many of the more prominent advocates in education today make their cases based largely on raw changes in outcomes immediately after (or sometimes even before) their preferred policies were implemented (also see here, here, here, here, here, and here). In order to illustrate the monumental assumptions upon which these and similar claims ride, I thought it might be fun to break them down quickly, in a highly simplified fashion. So, here are the four “requirements” that must be met in order to attribute raw test score changes to a specific policy (note that most of this can be applied not only to claims that policies are working, but also to claims that they're not working because scores or rates are flat):

The Impact Of Race To The Top Is An Open Question (But At Least It's Being Asked)

You don’t have to look very far to find very strong opinions about Race to the Top (RTTT), the U.S. Department of Education’s (USED) stimulus-funded state-level grant program (which has recently been joined by a district-level spinoff). There are those who think it is a smashing success, while others assert that it is a dismal failure. The truth, of course, is that these claims, particularly the extreme views on either side, are little more than speculation.*

To win the grants, states were strongly encouraged to make several different types of changes, such as adoption of new standards, the lifting/raising of charter school caps, the installation of new data systems and the implementation of brand new teacher evaluations. This means that any real evaluation of the program’s impact will take some years and will have to be multifaceted – that is, it is certain that the implementation/effects will vary not only by each of these components, but also between states.

In other words, the success or failure of RTTT is an empirical question, one that is still almost entirely open. But there is a silver lining here: USED is at least asking that question, in the form of a five-year, $19 million evaluation program, administered through the National Center for Education Evaluation and Regional Assistance, designed to assess the impact and implementation of various RTTT-fueled policy changes, as well as those of the controversial School Improvement Grants (SIGs).

Do Top Teachers Produce "A Year And A Half Of Learning?"

One claim that gets tossed around a lot in education circles is that “the most effective teachers produce a year and a half of learning per year, while the least effective produce a half of a year of learning."

This talking point is used all the time in advocacy materials and news articles. Its implications are pretty clear: Effective teachers can make all the difference, while ineffective teachers can do permanent damage.

As with most prepackaged talking points circulated in education debates, the “year and a half of learning” argument, when used without qualification, is both somewhat valid and somewhat misleading. So, seeing as it comes up so often, let’s very quickly identify its origins and what it means.

Examining Principal Turnover

Our guest author today is Ed Fuller, Associate Professor in the Education Leadership Department at Penn State University. He is also the Director of the Center for Evaluation and Education Policy Analysis as well as the Associate Director for Policy of the University Council for Educational Administration.

“No one knows who I am," exclaimed a senior in a high-poverty, predominantly minority and low-performing high school in the Austin area. She explained, “I have been at this school four years and had four principals and six algebra I teachers."

Elsewhere in Texas, the first school to be closed by the state for low performance was Johnston High School, which was led by 13 principals in the 11 years preceding closure. The school also had a teacher turnover rate greater than 25 percent for almost all of the years and greater than 30 percent for 7 of the years.

While the above examples are rather extreme cases, they do underscore two interconnected issues – teacher and principal turnover - that often plague low-performing schools and, in the case of principal turnover, afflict a wide range of schools regardless of performance or school demographics.

Low-Income Students In The CREDO Charter School Study

A recent Economist article on charter schools, though slightly more nuanced than most mainstream media treatments of the charter evidence, contains a very common, somewhat misleading argument that I’d like to address quickly. It’s about the findings of the so-called "CREDO study," the important (albeit over-cited) 2009 national comparison of student achievement in charter and regular public schools in 16 states.

Specifically, the article asserts that the CREDO analysis, which finds a statistically discernible but very small negative impact of charters overall (with wide underlying variation), also finds a significant positive effect among low-income students. This leads the Economist to conclude that the entire CREDO study “has been misinterpreted," because it’s real value is in showing that “the children who most need charters have been served well."

Whether or not an intervention affects outcomes among subgroups of students is obviously important (though one has hardly "misinterpreted" a study by focusing on its overall results). And CREDO does indeed find a statistically significant, positive test-based impact of charters on low-income students, vis-à-vis their counterparts in regular public schools. However, as discussed here (and in countless textbooks and methods courses), statistical significance only means we can be confident that the difference is non-zero (it cannot be chalked up to random fluctuation). Significant differences are often not large enough to be practically meaningful.

And this is certainly the case with CREDO and low-income students.