July 2, 2019

Long Weekend Edu-Reads

Mike Antonucci stays with the Janus story and recommends we take the long view.

Ashley LiBetti has three things Head Start programs can do to get better.

The Center for American Progress has a new agenda for education policy.

A new working paper finds that, at least among survey participants, “the provision of growth data causes participants to choose less white and less wealthy districts.”

Here’s an interesting new study by Jason Grissom and Brendan Bartanen on turnover among Tennessee principals. The key graph is below: For the most part, principals who leave their positions are less effective, although the very highest-performers are also slightly more likely to leave.

The stories coming out of Providence and Oakland right now are sobering. If nothing else, they are a reminder that there’s plenty more we could be doing rather than getting bogged down in the existential debate over ending poverty versus fixing schools. We can (and should!) do a better job of delivering educational services to low-income students than what these districts are providing today.

–Guest post by Chad Aldeman 

Posted on Jul 2, 2019 @ 10:26pm

July 1, 2019

American Schools Are Not Resegregating

Let’s start with a math problem. Say you have ten marbles. Eight are red and two are blue. If you blindly draw two marbles, what are the chances you’ll draw the two blue ones?

Now let’s say you have ten marbles, but this time five are red and five are blue. If you blindly draw two marbles again, what are the chances you’ll draw two blue ones this time?

The answer to the second question is ten times higher than the first (2/9 versus 1/45). By changing the composition of the problem by a little bit, the odds have changed dramatically.

This is roughly what’s happening in American schools. In 1968, eight out of ten students were white, and two out of ten were non-white. By 2012, the share of white students had fallen to about five in ten.

While students are not marbles, we’re facing the same basic math problem when we talk about school segregation. On the surface, American schools today appear to be more segregated than ever. The chance that a black or Hispanic student will attend a school with students of their same race has increased significantly, but that’s due to the composition of our society, not sorting. Once you take into account changes in student demographics, American schools are actually less segregated than in the past.

There’s a body of academic research documenting this phenomenon. But perhaps it helps to see it visually. The graph at right comes from a piece by Steven Rivkin for Education Next, documenting the decline in what’s called the “dissimilarity index.” As Rivkin explains it, the dissimilarity index measures, “the percentage of blacks who would need to change schools if blacks and whites were to attend each school in the same percentage as their percentage of public school enrollment.” Across the country, the dissimilarity index has not improved all that much at the district level, but at the school level it’s fallen from 81 percent in 1968 to 66 percent in 2012.

Most of these gains occurred in the 1970s, and the gains have been smaller since then. But remember, these gains are on top of America’s growing diversity. American society has become more diverse, and our schools have integrated even faster.

I’d like to see even more integration, for lots of reasons. There are real benefits to integration efforts, but we should be careful to diagnose our problems accurately, avoid distorting the data, and be wary of universal solutions. As our country gets more diverse over time, it’s going to look like segregation is getting worse. But we have to go deeper to understand what’s really happening.

–Guest post by Chad Aldeman 

June 28, 2019

Weekend Edu-Reads

Jason Weeby on school boards, charter schools, and democratic control of schools. Ashley Jochim’s response is also worth checking out.

Josh Mitchell and Michelle Hackman take a look at the Kalamazoo Promise program for the Wall Street Journal. The entire piece is worth your time, but this graph really tells the story:

Ashley LiBetti interviews Kelly Riling, the manager of the AppleTree Early Learning Teacher Residency program in Washington, D.C. Unlike other residency programs, they make it work…  and offer residents a salary with benefits!

Matt Kraft, John Papay, and Olivia Chi look at teacher development through the lens of teacher performance ratings from principals. Like with value-added, teachers tend to improve over time, but the most promising early-career teachers make even faster improvements.

Conor Williams neatly summed up this week’s Democratic debates. With respect to education, “Precisely zero of the current Democratic candidates for that party’s presidential nomination believe that public education is the primary cause of American inequality.”

–Guest post by Chad Aldeman 

Posted on Jun 28, 2019 @ 2:52pm

Friday Fish Pics — Son Catches, Dad Releases

Just because Andy is gone fishin’ doesn’t mean we pause Friday Fish Pics. This week we have Bellwether’s Jason Weeby and his 6-year-old son. The younger Weeby caught a beautiful 12″ rainbow trout on Sardine Lake in Northern California:

The fish nearly escaped Jason’s grasp, yielding this gem of a photo:

–Guest post by Chad Aldeman 

June 25, 2019

Universal Student Loan Forgiveness Solves the Wrong Problems

Bernie Sanders is proposing to forgive all student loan debt, an outstanding balance of about $1.6 trillion. Lots of people have pointed out flaws with this idea, but Matt Bruenig has one of the simplest explanations for what’s wrong with the Sanders proposal:

Thus it appears that the universe of people selected for this program is both over-inclusive and under-inclusive. It is over-inclusive because people who graduate from Harvard’s MBA program in 2021 will receive hundreds of thousands of dollars of debt relief (though, strangely, people who enroll in that program in 2022 will receive nothing). It is under-inclusive because people who drained $50,000 off their net worth to pay for an undergraduate degree at a public college — but who do not currently carry student debt — will receive no reimbursement for their ongoing financial suffering.

$1.6 trillion is a lot of money. If we’re sticking solely in the education world, that would buy a lot of Pell grants for poor students or slots for Head Start or universal pre-k. Heck, the federal government could wipe out all state pension debt for that kind of money, which would translate into boosting teacher salaries by an average of about 12 percent.

Maybe Senator Sanders is so attached to the idea of universal programs that he just doesn’t care about how well they target benefits. Maybe I’m missing the point, but if we’re going to invest that kind of money, I’d prefer we target it to individuals who need it the most.

–Guest post by Chad Aldeman 

June 23, 2019

What To Make of the Research on Teacher-Student Race-Match?

Constance Lindsay has a nice run-down of the research on the importance of teacher-student race match. To put it bluntly, black students do better on a host of educational outcomes when they have black educators in their lives. What can we do about this problem?

The simple and most obvious answer is to diversify our teacher workforce. There are a lot of reasons to pursue a more diverse teaching corps, but, given our current trajectory, that could take a very long time.

Another possible lever would be to train or re-train our existing teachers. My Bellwether colleagues Max Marchitello and Justin Trinidad wrote a report on the merits of that approach.

However, training efforts aren’t exactly straightforward. We can’t even train teachers how to teach reading or math effectively, so we may want to be cautious that we’ll be able to train teachers on anti-racism at scale.

And then there’s the sheer size of the teacher workforce. Is it reasonable to expect all 4 million public school teachers to change their attitudes and dispositions? That’s especially true if the race-match effects are being driven by more subtle differences about expectations rather than overt racism.

It would be easy to see these obstacles and throw up our hands, but the race-match finding is too important to simply ignore. Yet instead of addressing the race-match issue through individual teachers, we should be thinking more systemically. Reforms like double-blind disciplinary reviews, universal testing and screening policies, and other policy reforms could take the human element out of the equation and more immediately address the consequences of the teacher-student race-match problem.

–Guest post by Chad Aldeman 

Posted on Jun 23, 2019 @ 11:29pm

June 21, 2019

Weekend Edu-Reads

Matt Kraft has a thread about a new OECD study looking at how American teachers compare with teachers in other developed countries. Or read the full study here.

In the context of the charter school debate, David Osborne makes the point that we’ve “privatized” all sorts of public services, from trash and recycling to Medicare and Medicaid. Even in the education sector, we spend billions of public dollars every year on privately provided early childhood and higher education services.

Join me and Education Week‘s Daarel Burnette II to talk about what education reporters need to know about teacher pensions.

On that front, Mike Antonucci has a quick dive into California pension figures.  This is right-on:

We tend to frame the pension issue as a thin line between school districts and the state on one side, and public employees on the other. But as pension obligations grow, it also becomes a tug-of-war between the 300,000 CalSTRS retirees and the 650,000 educators who are enrolled in CalSTRS but still working.

Mike Petrilli with a happy reminder that educational outcomes are trending upward, even if gaps remain and we’ve made more progress for younger than older students.

–Guest post by Chad Aldeman 

June 19, 2019

Is Teach For America Our Largest Provider of New Teachers of Color?

Yesterday I noted that Teach For America is likely our largest provider of new teachers. Although it’s impossible to prove definitively given current data limitations, TFA is also likely our largest provider of new teachers of color.

First, let’s go with TFA’s estimate that half of their corps members identify as people of color. Since they’re currently placing about 3,500 teachers a year overall, that means TFA is responsible for about 1,750 new teachers of color entering America’s public schools every year.

There are very few teacher preparation programs that prepare that many teachers overall. Even if you count all bachelor’s and Master’s degrees in education, only seven institutions train as many total teachers as TFA does teachers of color. And for those other institutions, that’s the total number of prepared teachers; it includes people of all races, people who earn education degrees but never teach, and people who are earning in-service Master’s degrees and aren’t strictly new to the profession.

Once you take those caveats in mind, it’s almost guaranteed that TFA is the largest single provider of new teachers of color.

Maybe this comparison isn’t completely fair. After all, TFA is one national program, but it’s made up of 51 regions across the country. In that way, you could argue TFA is more like a system than one standalone entity.

But even viewed more expansively, TFA is still likely at or near the top of the list. On a cumulative basis, the 101 HBCUs with education programs operating in 2016-17, the most recent year for which we have data, granted a total of about 3,000 bachelor’s and Master’s degrees in education.* Again, not all these people will teach, and not all of them are new teachers.

As I mentioned yesterday, the California State University system is probably the closest parallel. It’s impossible to know given current data systems, but it’s likely that TFA’s placement of 1,750 new teachers of color probably outpaces even Cal State.

I went through the evidence on TFA’s impact in more detail yesterday, but it’s worth remembering that TFA is placing its corps members into disadvantaged schools with high concentrations of poor and minority students. On average, TFA corps members are at least as effective with those students as other teachers in the same schools, but there’s also suggestive evidence that TFA corps members have an especially strong positive effect on black and Hispanic students.

We don’t know exactly why TFA gets the results it does, whether it’s the “grit” of their corps members, or the expectations they set for their students, or whether there’s something about the teacher-student race match. But regardless, TFA is helping to diversify our schools, and those efforts are paying off for the students it serves.

*Note that TFA also recruits from and partners with colleges and universities. In the event someone earned an education degree and participated in TFA, they would be double-counted in these numbers. But if they attended the college, graduated with a non-education degree, and then joined TFA, TFA would get the credit in these totals.

Disclosure: Bellwether has worked with TFA in the past. 

–Guest post by Chad Aldeman 

June 18, 2019

Teach For America Is Our Largest Provider of New Teachers… And Likely Our Best!

There’s lots of talk today about this ProPublica story on Teach For America. To put it bluntly, I don’t think it’s a very fair or nuanced piece of journalism, and it’s missing two major pieces of context.

One, Teach For America is probably the largest provider of new teachers in the U.S.

This is hard to prove definitively given our fragmented data systems, but I’m quite confident it’s true. Consider our largest teacher preparation programs, in terms of total graduates. I pulled the latest data on all 2016-17 graduates with a bachelor’s or Master’s degree in education. The list of the largest teacher preparation programs may surprise some people, but here’s the top ten:


Institution Total Education Graduates (2016-17)
Grand Canyon University 5037
Western Governors University 4009
Concordia University-Portland 2320
Walden University 2202
Liberty University 2081
Ashford University 2066
National University 1821
University of Phoenix-Arizona 1580
University of Central Florida 1522
Ball State University 1272

These are big numbers, but they reflect graduates, not teachers, let alone new teachers. These are important distinctions. We know based on state and national numbers that many graduates with education degrees do not actually become teachers. Depending on the year and region, 25-40 percent of people who earn degrees in education never use that degree as a teacher.

Moreover, these numbers include bachelor’s and Master’s degrees. it’s impossible to know from the data, but it’s likely that the totals are padded by active teachers pursuing a Master’s degree to boost their salaries.

Let’s use Grand Canyon University as an example. It granted the most bachelor’s and the most Master’s degrees in education, but only 1,564 of those were bachelor’s degrees. If we assume that only those attaining a bachelor’s were truly new, and those graduates went on to teach at the national averages, Grand Canyon might have been responsible for somewhere around 1,000 new teachers that year.

The Cal State system provides another example. While no single institution made the list above, the Cal State system as a whole produced about 2,100 bachelor’s degrees and about 3,100 Master’s degrees in education that year. But in order for those teachers to qualify for a full-time license, California requires teachers to go through their preparation program after completing their bachelor’s degree. So again, these totals are deceptively high in terms of the total new teachers placed by the Cal State system.

In contrast, Teach For America currently has about 7,000 corps members, and they regularly churn out cohorts of 3,000 to 4,000 teachers.* Granted, TFA corps members are spread out all across the country, so it may not be the largest provider in any one state or region, but, collectively, TFA is larger than any other provider of new teachers.

Two, Teach For America may be the best teacher preparation program in the country.

Again, this is a hard statement to prove definitively, but TFA is certainly the most studied, and it reliably produces results that are at least as good if not better than its peers. TFA corps members outperform other incoming teachers in states like North Carolina and Tennessee and they even perform as well as other veteran teachers working in their same schools. TFA teachers do have higher turnover rates than other new teachers, but, on balance, students are still better off. This research mainly focuses on student growth in terms of achievement scores, but TFA teachers may also help boost student attendance

However, there’s a much larger difference across TFA corps members than between TFA and other preparation programs. (If you want to see what this looks like visually, Figures 2-3 here illustrate that point nicely.) That finding alone should make us stop and pause.

In fact, this variance issue plagues much of the journalism about TFA. Given the enormous size of TFA and the wide variety of outcomes of its teachers, it’s easy to find corps members who fit whatever narrative you might want to apply to it.

So I get why TFA is such a lightning rod. It’s huge, and it provides a stark contrast to the traditional teacher preparation programs. But I don’t find the politics around TFA all that interesting or illuminating. I’m much more interested in how this enormous, diffuse, short-term training program manages to produce new teachers who are still roughly on par with their peers.

*TFA often partners with higher education institutions, and when their corps members complete a degree, those completions would be counted toward the host college or university, not TFA. 

Disclosure: Bellwether has worked with TFA in the past. 

–Guest post by Chad Aldeman 

Latest Edu-Reads

Go broad or go deep? I’m on the record suggesting the education field is tilting too far in the direction of specialization. If you’re interested in this question, I strongly recommend this podcast conversation between Patrick O’Shaughnessy and David Epstein. Epstein’s new book Range also looks promising.

Make sure to read Lauren Camera’s deep dive on the Census, and how including a citizenship question on the 2020 Census would affect schools.

Over at The Line, Andy Rotherham writes about America’s “crisis of purpose,” and what we can reasonably ask schools to do to resolve some of society’s thorniest problems.

Max Marchitello on a wonky but important story out of West Virginia. The state’s pension system is massively under-funded, biting into education budgets and teacher paychecks, and yet the state’s efforts to reform its pension plan didn’t go well. Max dives into what happened and what we can learn from it.

Smarter than the average ____? This is an interesting article on the intelligence of bears.

–Guest post by Chad Aldeman