Using data from Michigan, the authors identify the effects of charter schools on special education students by comparing special education students who enroll in charter schools early with those who enroll in charter schools late. This research design addresses a common concern in charter school research: students who choose to enroll in charter schools may differ from those who remain in traditional public schools in unobservable ways. Simple comparisons between charter and traditional public school students can therefore be misleading.

To overcome this challenge, Imberman and Johnson compare early charter entrants to late entrants. Because both groups eventually attend charter schools, they are more comparable to one another than to students who never enroll. The effect of charter school attendance is identified by examining differences in outcomes before the late entrants make the switch.

In my view, the study’s two most important findings are:

• Charter schools use special education programs and service assignments that are less intensive and expen­sive than in traditional public schools.
• Charter schools improve special education students’ academic achievement and attendance.

The authors also conduct a parallel analysis of general education students. They show that the positive effects of charter schools on special education students are similar to the positive effects on general education students.

This study complements recent work by Elizabeth Setren, who examines special education students in Boston who randomly win or lose lotteries to attend charter schools. Because lottery outcomes are random, this design provides especially strong causal evidence that factors other than charter school attendance are highly unlikely to drive the results. Setren likewise finds that charter schools improve test scores for special education students.

Special education students are an important subpopulation. They account for nearly 15 percent of K-12 enrollment in the United States and receive disproportionate funding. Both of these studies find charter schools serve special education students more effectively, and contribute to the large and growing body of evidence showing that charter schools outperform traditional public schools.

The post Charter Schools Do Special Education Better appeared first on Show-Me Institute.

Eric basically said, “Why do it all at once? Why not experiment at one school?” This suggestion may not make sense in a small, rural school district with one elementary school, but it makes perfect sense for a larger school district. Take the Independence School District, for example. The district has over 14,000 students and 20 elementary schools. The district decided to move to a four-day school week.

Think about what district leadership could have done if they had approached this like a researcher.

The Independence School District could have selected two elementary schools to pilot a four-day school week. If it is as appealing as the district says it is, then many teachers and students would likely want to move to that school. The district could have held a lottery to randomly accept teachers and students into the school.

Then we could have had a random assignment evaluation of the school district that moved to the four-day school week. We would have had a group of students in a five-day school and a group in a four-day school and the only difference between the two groups would have been random chance. This is the gold standard of social science research.

I understand the impulse of superintendents and school boards to consider the four-day school week, but they do not have to make the move all or nothing. Pilot it. Evaluate it rigorously. A move this significant deserves that kind of consideration.

The post Considering the Four-Day School Week? Pilot It and Evaluate It appeared first on Show-Me Institute.

James V. Shuls is the director of research and distinguished fellow of education policy at the Show-Me Institute.

Listen on Apple Podcasts 

Listen on SoundCloud

Produced by Show-Me Opportunity

The post Character Education with James V. Shuls appeared first on Show-Me Institute.

Dan Goldhaber is an AIR vice president and director of Analysis of Longitudinal Data in Education Research at AIR. He is also an affiliate professor in the School of Social Work at the University of Washington, the director of the Center for Education Data & Research, and the co-editor of Education Finance and Policy.

Listen on Apple Podcasts 

Listen on Stitcher 

Listen on SoundCloud

The post Is There Actually a Teacher Shortage? appeared first on Show-Me Institute.

Macke is the founder and director of the Center for Research on Education Outcomes (CREDO) at Stanford University.

Listen on Apple Podcasts

Listen on SoundCloud

The post Podcast: Let’s Redesign High School with Stanford’s Macke Raymond appeared first on Show-Me Institute.

A policeman sees a man looking for something under a streetlight and asks what he has lost. He says he lost his keys and the policeman decides to lend a hand looking for them. After a few minutes, the policeman asks if he is sure he lost them there, and the man replies, no, he lost them on the other side of the street. The policeman asks why he is searching here, and the man replies, “well, this is where the light is.”

When education researchers want to measure the impact of a policy or program, they are forced to look where the light is. That usually means looking at student test scores, graduation rates, and a set of relatively limited short-term indicators.

Fortunately, school-choice researchers are starting to look at outcomes beyond just test scores, casting light into areas that were previously shrouded in darkness.

In fact, not only does new research show that school choice can boost test scores and increase the likelihood that low-income students finish college, but studies also suggest that students in school choice programs are less likely than their traditional school peers to commit crimes.

Researchers studying the high-performing Promise Academy in the Harlem Children’s Zone found promising results for students at that school, which uses a lottery system to place students in the limited number of available spots. Four percent of lottery “losers” were incarcerated compared to none of the lottery “winners.” In addition, charter school students were 17 percentage points more likely to enroll in college immediately after high school, and female lottery winners were 10.1 percentage points less likely to report having been pregnant as a teenager than lottery losers.

The Charlotte-Mecklenburg district in North Carolina used a lottery system to place students into schools that have a limited number of available seats. A study by David J. Deming at Harvard University shows that high school students who “won” the lottery and were placed into their first-choice school were arrested 70 percent less for drug charges and 45 percent less for other felony charges compared to students who entered the lottery but did not secure a spot in their first-choice school.

Schools that use lottery systems for admission are especially helpful for comparison studies because they allow researchers to compare two groups of students who both showed a desire to attend a school of their choosing. But even in situations where there is no lottery, researchers can use other techniques to help ensure the validity of their findings.

For example, researchers from the University of Arkansas found the following reduction in crime rates for male students, relative to incidence rates for their age among the general population: 79 percent for felony crimes, 93 for drug related crimes, and 87 percent for thefts. Because private school enrollment through a voucher program is not capped in Milwaukee, researchers couldn’t sort students into lottery-winner and lottery-loser groups. Instead they “used comparison groups constructed through an algorithm that matched [voucher] students with Milwaukee Public School (MPS) students based on grade, neighborhood, race, gender, English language learner (ELL) status and math and reading test scores” (see page 6 of the study).

These studies are encouraging, and suggest that school choice can not only enrich the lives of students, but also help make our cities and communities safer. No wonder school choice is becoming more popular.

The post Studies Show Benefits of School Choice Extend Beyond the Classroom appeared first on Show-Me Institute.

The post Mike McShane on Pre-K Effectiveness appeared first on Show-Me Institute.

Or, better yet, they may consider how little research there is on the matter. According to a recent article by Haley Glatter in the Atlantic, there are very few studies that thoroughly examine state-run and state-funded pre-K programs. Moreover, these studies are inconclusive on whether or not pre-K offers lasting benefits.

In Georgia and Oklahoma—states with universal pre-K programs—there is evidence that pre-K has reduced achievement gaps. The jury is still out in New York, which established universal pre-K only two years ago. Tennessee, on the other hand, implemented targeted pre-K for low-income children. Positive results were evident when these children entered kindergarten, but the benefits began to fade by first grade. By third grade, these students were performing worse than other students on statewide assessments.

Here are a few things we can take away from these states’ experiences:

• The quality of the pre-K programs offered matters immensely—not all pre-K is created equal.
• Good pre-K programs will not make up for the deficits of a failing K-12 system.
• There are only a handful of programs operating at a state-wide scale which makes it difficult to draw firm conclusions on pre-K.                                                                                                 

One of the lingering questions about Amendment 3 is what exactly the pre-K program it creates will look like. In his recent analysis of Amendment 3, the Show-Me Institute’s Michael McShane writes:

To date, it simply isn’t clear what the regulations for participating schools will be. In one sense, this uncertainty means we should probably reserve judgment, but in another it makes it hard to support a program when we don’t know what that program will ultimately look like.

On the one hand, it makes sense for us to gather as much information as we can about the effectiveness of pre-K programs before Missouri designs a program of its own. But then, does it also make sense to wait until we have a well-researched plan in place before amending the constitution and investing such a significant amount in early childhood education?

The post Is Expanding Pre-K a Good Idea? The Jury Is Still Out. appeared first on Show-Me Institute.

I wanted to follow up on that post with a quick addendum on a related issue that charter critics often raise in discussions about the discipline practices of charter schools. They often claim that the draconian discipline systems within charter schools are used to push out students who are lower performing or are too disruptive to handle. Like the myth of charter school suspensions, this one isn’t true either.

Now it is true that, on average, there are differences between charter schools and traditional public schools in terms of the types of students that they enroll. But careful research has yet to find evidence that charter schools actively push out low-performing students at rates higher than those of traditional public schools. Ron Zimmer of Vanderbilt and Cassandra Guarino at Indiana University, for example, analyzed data from an anonymous large urban school and found no evidence of pushing out low-achieving students. Marcus Winters similarly found no evidence that charter schools in New York disproportionately pushed out low-achieving students. It does appear that in Chicago, charter schools expel students at a higher rate than traditional public schools do, but in the nearly all-charter district in New Orleans the expulsion rate is lower than in the rest of the state, even though the students in New Orleans are more disadvantaged. These seemingly contradictory results are why we should value research with the appropriate statistical controls.

There is also little evidence that charter schools “skim” the best students from the public school system. In fact, a team of researchers also led by Ron Zimmer found no evidence in the seven locations they examined.

At best, proponents of these theories offer isolated anecdotes or decry the actions of a particular school or school network without asking if that school or network is representative of the system as a whole.

All of this research aside, I think there is an important conversation to be had about the value of discipline in and of itself. I would argue that we have to be open to the idea that that suspending more students actually makes for a better learning environment. We know that disruptive students have a huge (and I mean huge) negative effect on their peers. In some cases, suspending students might the only way around that. I hope that isn’t the case—but it very well could be. The autonomy of charter schools gives them the latitude to experiment with different discipline practices. Hopefully we can learn from their efforts and continue to improve student discipline practices.

The post No, Charter Schools Don’t Push Out Kids Who Are Too Hard To Teach appeared first on Show-Me Institute.

For years now, critics (and even some supporters) have known for sure that charter schools had harsher discipline policies than traditional public schools and suspended or expelled students at much higher rates. According to new research from Nat Malkus of the American Enterprise Institute, that just ain’t so.

Malkus used data on school suspension rates collected by the federal government to compare charter schools to the traditional public schools that surround them. As Malkin's graphic (above) shows, in most cases, there is no substantial difference in the rate of suspensions between traditional public schools and charter schools. In fact, while 17 percent of charter schools do see rates higher than neighboring public schools, 29 percent see rates that are significantly lower.

It’s time we put to bed the idea that charter schools are draconian institutions where the joy of learning is taken from students and where harsh punishment is meted out for the most minor of infractions. The data do not back that story up.

The post The Charter School Discipline Problem that Isn’t appeared first on Show-Me Institute.

All I could think was, here we go again.

If you’ve followed education policy for any length of time this routine looks familiar. Researchers or policymakers visit some other country that performs better than we do on international assessments and then come back with the secret sauce that makes them do so well. The recommendations are so anodyne that that anyone with a passing knowledge of the education system will probably agree with them. The policy flavor of the month (national education standards, universal pre-K, etc.) is usually highlighted. Rinse and repeat.

For a while we were told to emulate Shanghai—until we learned that Shanghai systematically excludes disadvantaged students from its testing pool to juice its scores. Then it was Finland, at least until Finland’s scores dropped precipitously on the very test that ushered in its rise to prominence.

Here is the problem with this approach, and why it never actually yields the information we’re looking for: it violates the basic precepts of research design. If you want to know if a certain policy affects an outcome, you develop a hypothesis and test it. You follow children that are subject to the policy and children that aren’t—doing your best to make sure all other aspects of the two groups’ educational environments are identical—and you see what happens. If the children who were subject to the policy do better than those who weren’t, then you have reason to believe that the policy caused the improvement.

“Do what the best schools do” research does the exact opposite of that. It sees a result it likes and then tries to work backwards to the cause without isolating other variables that might also explain the outcome. That is not how science works. Are there places, for example, that do the very things that Finland or Shanghai or any of these other countries do that don’t meet with success?  Are there countries meeting with success that don’t do these things? As I’ve written before, the Netherlands is a very high-performing country that is marked by an incredible degree of school choice. So why does a universal voucher system always seem to be missing from the list of recommendations? In short, we have no idea whether the policies these folks advocate are really behind these countries’ successes.

I don’t want to say there’s nothing we can learn from other states or other countries about how to improve education here at home. However, if we are going to make claims that one policy or another causes a particular outcome, we need to back those claims up with research done the proper way. 

The post Not This Again appeared first on Show-Me Institute.

Recently, the Center for Public Education, an arm of the National School Boards Association, released a report on the merits of school choice. The paper claims to summarize “what the research says.” Interestingly, the report fails to include almost every analysis that has found benefits to private school choice programs.

When Anna Egalite, an assistant professor of educational leadership, policy, and human development at North Carolina State University, conducted a systematic review of the competitive effects of private school choice programs, she found 21 studies. She concluded that the results “unanimously find positive impacts on student achievement. Such overwhelming evidence supports the development of market-based schooling policies as a means to increase student achievement in traditional public schools.” Interestingly, the Center for Public Education did not cite any of these studies.

Similarly, there have been 12 random-assignment studies of voucher programs. These are considered the “gold-standard” in social science research because they are the best at determining causality. Eleven of the 12 studies have found positive effects from voucher programs. The Center for Public Education review only cites one of these studies.

The report cites plenty of useful statistics from the National Center for Education Statistics and other sources, but does not even attempt to cite the plethora of useful research on school choice programs.

Nevertheless, the report does get at least one thing right—private school choice tends to boost graduation rates.  This was highlighted in the evaluation of the Washington D.C. Opportunity Scholarship program, which showed a 21 percentage point increase in the graduation rate for voucher users.

Not surprisingly, given that they neglect to cite any of the ample evidence showing that school choice succeeds, the Center's conclusion is that “In general, we find that school choices work for some students sometimes, are worse for some students sometimes, and are usually no better or worse than traditional public schools.”

In addition to simply being inaccurate, this conclusion does a great disservice to the students most in need of another educational option; particularly the students in the six Missouri school districts with the worst graduation rates (see above).

School choice may not be a cure-all, but the evidence is quite clear: students benefit from school choice. Indeed, students at Sumner and the other schools listed above may benefit the most. 

The post Denying Evidence, Denying Opportunity appeared first on Show-Me Institute.

The countries that have left the United States behind in math and science education have one thing in common: They offer the same high education standards — often the same curriculum — from one end of the nation to the other. The United States relies on a generally mediocre patchwork of standards that vary, not just from state to state, but often from district to district. A child’s education depends primarily on ZIP code.

That could eventually change if the states adopt the new rigorous standards proposed last week by the National Governors Association and a group representing state school superintendents.

The editorial’s optimism is unwarranted. It’s true that the countries that score above us on international tests have national education standards. But we can’t conclude that standards contributed to their success, because almost all of the countries that score below us have national standards, too. Maybe standards would make us more like those high-scoring countries, or it could be that they would bring us down to the level of the lower-scoring countries. Data from international tests can’t tell us what effect national standards would have on U.S. education.

Uniformity is a hindrance when you’re trying to research the effects of national standards. It’s also a problem for researchers who study curriculum and teaching methods. All the research behind the proposed national standards was possible because lots of schools did different things. If every school took the same course of action — for example, by following national standards — less research would be possible. The standards’ authors defend their proposal as “research-based,” all the while advocating for a policy that would impede research if states agreed to it.

The post Uniform Standards Would Make Education Research More Difficult appeared first on Show-Me Institute.

Field will never get the proof it wants. To show that single-sex schooling improves student achievement, you would have to randomly assign some students to single-sex classes and others to coed classes. Otherwise, you can’t tell whether the students in the single-sex track are comparable to their peers in the coed track. Perhaps students who struggle are more likely to look for a change and to opt in to the single-sex classes. Or, it could be that the most involved parents seek out single-sex education, and that parental involvement gave those students an advantage. It’s impossible to sort out these factors as long as the program remains voluntary.

The post Columbia School Reluctant to Open More Single-Sex Classes appeared first on Show-Me Institute.