Jay P. Greene's Blog

(Guest Post by Patrick J. Wolf)

Charter schools remain controversial in the world of education.  Charters are public schools granted independence from some of the regulations that constrain traditional, district-run, public schools.  In exchange, charters promise to meet specific performance goals or close.  About 2.5 million students were enrolled in public charter schools in 2014-15, representing over 5 percent of the school-age population.

Charter schools not only are operated differently from district-run public schools, they also are funded differently.  They tend to receive little or no local property tax dollars, and often have to finance their buildings through operating funds, while district-run schools have access to capital funds to fund their facilities.

Today my crack school finance research team released Buckets of Water into the Ocean: Non-Public Revenue in Public Charter and Traditional Public Schools. Across the 15 states with both a sizable charter school population and detailed 2011-12 data about school funding from both public and non-public sources, we were able to determine that traditional public schools (TPS) received $13,628 per pupil in public revenue while charters received only $10,922, a funding gap of $2,706 per student or 20 percent.

Some commentators claim that the gap in public funding of charters and TPS is not a concern because public charter schools receive large donations from philanthropies to make up the gap.  (I’m talking about you Gary Miron and Bruce Baker.)  We set off to determine if that was true by conducting the first detailed study of the non-public revenues received by district-run public schools and charters across multiple states.

In fact, of the $504 million in charitable funds received by the schools in our study, $331 million or two-thirds was given to district-run schools while only $171 million or one-third was given to charter schools.  On a per-pupil basis, district-run schools received $18 from philanthropy while charters received $264, because district schools serve far more students than charters.

Importantly, the average amount of charitable funding per-pupil in the charter school sector masks the reality that a small number of charters receive the bulk of philanthropy.  Over 95% of all charter school philanthropy was directed at schools that enrolled just one-third of the charter students in our 15-state sample.  Over one-third of the charter schools received no philanthropic dollars at all.

Philanthropy isn’t the only source of non-public funds to schools, however.  Schools also receive non-public revenue from food service and investment income, among other things.  The district-run public schools in our study received $112 per pupil in food service revenue while the public charter schools received just $30.  The district-run schools also earned more investment income than the charters, $46 per pupil compared to $32.

When we add it all up, district-run public schools received $6.4 billion in non-public revenue in 2011-12 compared to $379 million for public charter schools.  On a per-pupil basis, district-run schools netted $353 in non-public funds compared to $579 for charters.  All sources of non-public funds together only shrunk the yawning per-pupil district-charter funding gap by $226, from $2,932 to $2,706 less in revenues for each charter school student.

Philanthropy simply can’t be expected to eliminate the gap in the public funding of district-run and charter schools.  The gap is too large and philanthropy too small for that to happen, a point already made with great eloquence by the namesake of this blog.  Almost 98 percent of all funding for district-run and charter schools comes from public sources.  Only 0.2 percent specifically comes from philanthropy.  If students in public charter schools are to receive funding on a par with students in traditional, district-run, public schools, it will have to come from more equitable public school funding laws.  Saying that charitable donations can make up the funding gap between district-run and charter schools is like saying that throwing buckets of water into the ocean will change the tide.

Under the direction of my colleague, Gema Zamarro, the Department of Education Reform is launching the Character Assessment Initiative, or Charassein.  Charassein (χαράσσειν) means to engrave, scratch, or etch and is the Greek root for the word character.  The idea of the initiative is to define, develop, and validate measures of what have often been called non-cognitive skills, but we think are more accurately described as character traits.  Once we have improved our understanding and ability to measure these traits, we will also be interested in evaluating potential interventions for shaping or altering them.  You should check out the Charassein web site to learn more about what we have and will be doing.

In some previous posts I’ve mentioned the incredibly innovative paper by Collin Hitt, Julie Trivitt, and Albert Cheng that looks at student non-response on surveys (leaving answers blank or saying “don’t know”) as a proxy for conscientiousness or effort.  They find that non-response is predictive in six different national longitudinal data sets of later life outcomes for students, including attainment, employment, and earnings, even after controlling for other relevant factors including cognitive ability.  That paper is part of our Character Assessment Initiative and I am pleased to report that it is moving closer to publication.  It received a positive R&R from a leading journal and, after the necessary revisions, is back under review.  You can find the updated paper here.

If you like that paper, you’ll love a new paper by Albert Cheng in which he looks at how teachers may affect student conscientiousness and later life outcomes.  Albert examines how teacher non-response on surveys influences student non-response.  Albert uses the Longitudinal Study of American Youth and confirms that student non-response on surveys is predictive of later life outcomes.  In fact, he shows that student non-response on surveys in grades 7-9 is more strongly predictive of graduating high school and completing a bachelors degree than math and science standardized test results. Albert then goes on to show that non-response on surveys administered to teachers is predictive of the non-response of their students.  Using a student fixed-effects model, he shows that student non-response tends to go up when they have a teacher who is more non-responsive on his/her surveys and tends to go down when students have teachers who are more responsive on surveys.  If we understand non-response as a proxy for conscientiousness or effort, then Albert has found that students become more conscientious when they have teachers who are more conscientious and less conscientious when they have teachers who are less conscientious.

This is an amazing breakthrough in character (or non-cog) research.  Albert demonstrates that teacher character influences student character and that student character is predictive of later life outcomes.  Check out his new paper and the updated version of the Hitt, Trivitt, and Cheng paper.  And for folks who think this is as cool as I do, keep in mind that both Collin and Albert will be hitting the job market in a year or two.

I recently recommended the Netflix series, Daredevil, to a friend.  He replied: “Aren’t you tired of comic book-based movies?  They just tell the same story over and over.”  I suggested that the familiarity of comic book characters was precisely what allowed them to be so good.

Comic book stories are just the modern version of Greeks myths or medieval saint stories.  The audience is familiar with the characters and basic plot, but the art is in the particular telling of those familiar stories.  And because the characters and basic plot are mostly known by the audience, much less energy needs to be devoted to explication.  Only the ways in which the characters and plot deviate from the familiar pattern require greater development.

Originality in story-telling is often over-rated.  Shakespeare mostly relied on familiar characters and well-worn plots.  His contribution was in how he used what was already well-known.  Conversely, I love original Chris Noland films, like Inception and Memento, but it’s amazing how much time in those movies is consumed with explaining how dreaming or memory work.  There is an efficiency in using the familiar.

This all being said, good comic book stories still require some development of plot and character, especially to highlight the ways in which the current telling deviates from previous ones.  I thought the Avengers movies fell short in devoting far too little to character and plot development and way too much to mind-numbing action.

But the new Netflix series, Daredevil, is a wonderful example of how comic book stories can be excellent if they are told well.  I remember the boring Ben Affleck movie-version of Daredevil, so it is striking how different the same story can be if it is just done better.  The new Netflix series version benefits from a serialized format to develop its version of the plot and characters more slowly.  It also uses Foggy for great comic relief.

And best of all, it crafts a sympathetic Wilson Fisk — so sympathetic that at times I wasn’t entirely sure whether he or Matt Murdock were actually the villain.  Fisk clearly uses evil methods but appears genuine in his regret about those methods and sincere in his goal for a better city.  Murdock/Daredevil, on the other hand, admits that he derives pleasure from his violence and only differs from Fisk in his methods in his unwillingness to kill his enemies.  Deriving pleasure from torturing one’s opponents hardly seems better than killing them with remorse.

Like all good comic book-based stories, Greek myths, and saint stories, Daredevil’s characters are archetypes worth exploring.  They capture some essential element of humanity, with its mix of tragic flaws and potential for greatness.  Check out Daredevil and let me know if you don’t think it was better than the Avengers and worthy of attention, like stories about Hercules or St. Francis.

(Guest Post by Matthew Ladner)

Interesting article in the WSJ today on the use of random assignment studies in anti-poverty efforts. Known as “Randomistas” these economists show a lot of promising results. At one point in the article Columbia’s Jeffery Sachs attempts to pour cold water on the process:

Prof. Sachs says that “many, almost surely most, of the cutting-edge breakthroughs in actual development in recent years did not result from [randomized controlled trials].” He believes that tackling problems at the level of communities or entire societies, rather than just households, is likely to be more effective—though, he adds, randomized controlled trials should be “a part of a diverse arsenal of analytical and policy tools.”

Retorts Prof. Duflo of MIT: “The big difference between Jeffrey Sachs and us is that he knows what needs to be done, and we don’t. We’re trying to learn it.”