school choice or peer choice?

A school choice quandary: parents care more about who attends a school than about its quality, in NYC study

PHOTO: Cassandra Giraldo

A basic tenet of school choice is that families will choose higher-quality schools when they can, spurring schools to improve in order to compete for students. Bad schools will fail the grueling test of the market, while good ones will thrive.

Now a new study raises questions about this basic premise.

The analysis examines high school choice in New York City, where students in district schools have a bevy of options and can attend schools outside their neighborhood. But families aren’t flocking to the most effective schools — they are looking for schools with higher-achieving students.

“Among schools with similar student populations, parents do not rank more effective schools more favorably,” write researchers Atila Abdulkadiroglu, Parag Pathak, Jonathan Schellenberg, and Christopher Walters. “Our findings imply that parents’ choices tend to penalize schools that enroll low achievers rather than schools that offer poor instruction.”

The result: school choice programs may incentivize schools to do more to attract students more likely to perform well, not help students learn more.

It’s a strong indictment of the theory behind school choice, though the research — like any single study — is hardly definitive. Prior studies on vouchers and New York City charters have shown that district schools generally see (small) increases in test scores when parents and students have more choices about what school to attend. Charter schools in several states have improved over time, which may be evidence of choice and and competition working.

But the study highlights some of the often-unspoken factors that drive school choice and how schools, in turn, are likely to respond.

Peers trump school quality in the eyes of families

The paper, which has not been formally peer-reviewed and was released through the National Bureau of Economic Research, examines how families of eighth-graders chose public high schools in New York City between fall 2003 and spring 2007.

Because the city allows students to rank many district high schools, and then assigns them one, the researchers have a treasure trove of data to draw from. (The latest analysis does not examine charter or private schools.) The study then connects how students ranked schools to metrics like test scores, high school graduation, and college attendance.

It is true that better schools — defined as schools improving those specific outcomes — are ranked higher, but that seems solely due to the fact that those schools also have higher-achieving students. Comparing schools with similar students, better schools don’t get a boost in parent demand.

“Our findings imply that parents’ choices tend to penalize schools that enroll low achievers rather than schools that offer poor instruction,” the authors write.

Perhaps surprisingly, there is not much evidence that schools that seem to do better with certain groups of kids are more likely to attract those students. In fact, schools that are particularly effective with low-achieving students tend to be especially popular with high-scoring kids.

It’s not clear which interpretation of the results is correct

There are a number of ways to interpret these results.

One, is that families value characteristics — like safety or after-school programs — besides the metrics of school quality used in this study. That said, the study includes measures like high-school graduation and college attendance, that parents and students are likely to care about.

Another hypothesis is that families and students simply don’t have good data on which schools are good.

“Without direct information about school effectiveness … parents may use peer characteristics as a proxy for school quality,” the study suggests. Indeed, there is evidence that families respond to information about school performance, but it’s unclear to what extent they would prioritize sophisticated measures of school quality, even if given that additional data.

Perhaps families are simply more concerned about peers than schools. Families may consider the types of students at a school as a proxy for school success — something that might be deeply ingrained and difficult to overcome. It may also be due to biases, including racism.

This, the authors suggests, has troubling implications for policy.

“If parents respond to peer quality but not causal effectiveness, a school’s easiest path to boosting its popularity is to improve the average ability of its student population,” the paper says. “Since peer quality is a fixed resource, this creates the potential for … costly zero-sum competition as schools invest in mechanisms to attract the best students.”

Want to learn more about NYC high schools? Come to Chalkbeat’s event this Thursday on how to make the high school admissions process more fair. Also be sure to sign up for Chalkbeat’s national and New York newsletters

Measuring schools

State education officials prepare 0 to 100 index to measure schools, slam push for A-F grades

PHOTO: Denver Post file

State education officials are preparing to roll out a new tool for parents to quickly learn which schools are succeeding and which ones are struggling. They’re also lashing out at another school measurement approach that’s been proposed in the legislature.

The dueling options are part of a national debate about the best way to measure schools.

Michigan’s elected board of education last year scrapped plans to assign letter grades to every school in favor of providing parents with a dashboard of information about test scores, graduation rates, and other measures of success such as attendance rates and student discipline.

That “parent dashboard” was unveiled last month. As soon as next week, the state is planning to beef up the dashboard with a new score, from 0 to 100, that is intended to summarize the quality of every school in the state.

The new index will give each school a single number based on seven factors, including test scores and graduation rates, the availability of classes like art and music, and proficiency rates for English learners. The index was part of the state’s plan to comply with the new federal school accountability law. 

Several factors will go into the index, though most points will be determined by test scores: 34 percent will be based on the percent of students who pass state exams. while 29 percent will be determined by whether test scores show students are improving. The rest of the score will be driven by school quality factors such as availability of arts and music (14 percent), graduation rates (10 percent), and progress by students learning English (10 percent). The last 3 percent will measure the percentage of students who take the state exam — a factor designed to discourage schools from giving the exam only to their highest-performing students.

Venessa Keesler, deputy superintendent at the Michigan Department of Education, said the index is not a ranking system, so multiple schools could end up with the same index score.

That’s a switch from the school ranking system Michigan has been using in recent years in which every school was placed against all other state schools, primarily on test scores. The schools in the bottom 5 percent of state rankings faced intervention, including the threat of closure.  

But GOP lawmakers say the parent dashboard and the index are too complicated, and they want to see an A-F letter grade system.

Lawmakers introduced legislation last week that would give every school a report card with six A-F grades measuring their performance in different categories. Bill sponsor Tim Kelly called it a “middle of the road” option that isn’t as simplistic as giving schools a single letter grade.

That plan came in for significant criticism Tuesday from the state board of education.

“This really isn’t OK,” said Nikki Snyder, a Republican board member. “If we want parents, students and teachers to be empowered, this is not the kind of chaos and confusion we should inject into our system. I absolutely do not support it.”

Another school board member, Casandra Ulbrich, the board’s Democratic co-president, raised concerns over how the scores would be decided.

“Someone has to create a complicated algorithm to determine the difference between A to B to C,” she said. “I have some real concerns about that.”

“I generally agree with Rep. Kelly,” said Richard Zeile, the Republican board co-president, “but school letter grades would be more misleading than helpful.”

A-F school ranking systems, which were used in 18 states as of last spring, have been divisive across the country, with some hailing them is a tool to increase transparency and others viewing them as too simplified and too easy for parents to misunderstand.

next steps

How to tackle New York’s severe school segregation? State policymakers spitball ideas

PHOTO: Monica Disare
Regent Collins and Regent Johnson engage in a discussion after a Board of Regents meeting in 2016.

A New York conference on the extensive research on the benefits of school integration. A convening of the state’s civil rights groups. A commission on equity and integration.

Those are some of the ideas being considered by a group of state policymakers tasked with addressing school integration in New York, which has some of the country’s most severe racial segregation. The group was established by Board of Regents Chancellor Betty Rosa in 2016 to research topics that the board would have to weigh in on; over time, it has come to focus on school integration and racial equity.

At its meeting Tuesday during the Regents’ monthly gathering, the group also floated ways to desegregate schools. One idea was to create incentives for schools that take steps to enroll students from different racial and socioeconomic backgrounds.

The group’s ideas build on other state efforts to combat school segregation. In 2014, New York’s education department launched a series of grants designed to improve schools by integrating them; the latest rounds of grants will expand the program to more schools and is more focused on training district leaders to combat school segregation. And as part of a plan they were required to submit last year under the federal Every Student Succeeds Act, officials discussed the idea of developing a new measure of school and district integration.

Those efforts come four years after a widely cited study by researchers at the University of California, Los Angeles found New York’s schools to be the most segregated in the country.

The group’s plans are still in their infancy: They will likely still be submitted to the full board, which would then have the chance to vet them before voting on whether to approve them.

Meanwhile, the group is still debating its own mission and objectives. During Tuesday’s discussion, one member suggested having the incentive program focus on “equity” rather than desegregation because some schools are unlikely to ever enroll many students of different races.

Regent Judith Johnson, who co-chairs the group, said Tuesday that she has struggled to figure out exactly what it should focus on — and how much to push integration in parts of the state where doing so could prove deeply unpopular. In New York City, many parents have resisted changes that would reroute their children to different schools in order to promote integration; in less diverse cities and towns, integration would likely require moving students across district lines.

“Not every district wants to address this issue,” Johnson said. “And so the question becomes: What is the role … of the Board of Regents?”