The underlying principle of the Trump administration’s approach to K-12 education is the need to give parents more ability to choose their child’s school.
In defending the President’s proposed 2018 budget before a House Appropriations subcommittee Wednesday, Secretary of Education Betsy DeVos repeatedly cited the need for parent choice: “We cannot allow any parent to feel their child is trapped in a school that isn’t meeting his or her unique needs.”
Yet new research suggests that parents pick schools for widely diverse reasons, and low-income parents in particular may need support to understand what different schools have to offer. For example, low- and high-income parents both select schools based on school quality—but they use different measures of quality, according to a new study of school choice in the journal Educational Evaluation and Policy Analysis.
Mathematica researchers Steven Glazerman and Dallas Dotter analyzed how more than 22,000 applicants to the District of Columbia’s citywide lottery ranked their preferences among more than 200 regular and charter public schools in the district. They looked at school demographics, how close schools were to students’ homes, as well as several measures of schools’ academic quality: academic proficiency and growth rates, the district’s three-tiered charter quality ratings, and the districtwide five-tier accountability ratings for all schools.
“The assumption was income might be a proxy for different types of social capital: access to information and the ability to use information to make decisions for your kids,” Glazerman said.
They found that in middle schools, for example, low-income parents ranked schools higher if they had higher academic proficiency rates—information that was easily available on the MySchoolDC website—but high-income parents tended to rank schools based on their accountability ratings, information that tended to be harder to find.
Similarly, parents of incoming kindergartners were also more likely to rank schools based on academic proficiency rates, while parents of high school students—who likely were more familiar with the school system—more often ranked their school choices based on the accountability ratings.
“It’s somewhat of a hard pattern to explain,” Glazerman said. “It’s likely people consume information in very different ways, and we need to understand how we present that information.”
Commute time played a role, too. Parents rated schools higher if they were within a mile of their homes—presumably walking or biking distance—but the commute was less important if the distance was longer and parents would likely need to drive. The researchers were not able to triangulate schools to parents’ work locations, so they were not able to tell how families’ broader transportation needs affected their school choices.
Reasons for School Choice
The findings back up a 2014 study of school choice in Canada, which also found that in addition to academic quality, parents gave weight to a school’s concentration of students in poverty or special education, and to commute time to get their children there.
“There’s lots of things we need to explore in more depth,” Glazerman said. “One of the things that was most surprising is we expected neighborhood characteristics would influence people’s decisions. You would expect if a neighborhood had higher violent crime, you would be less likely to send your children there.”
But neighborhoods with lower crime or higher-income families weren’t neccessarily more likely to draw students if other schools had attractive academic ratings and demographics, they found. “There are a lot of charter schools whose location might be based on the availability of real estate,” Glazerman noted, and successful charter schools might drive parents’ interest in locations they wouldn’t otherwise consider.
And again, the study noted that crime statistics for the area around a particular school would not be easy for parents to find; the researchers had to request data from the Federal Bureau of Investigation’s uniform crime statistics and map it to each school’s location.
“I don’t think there’s an easy answer, but there are some useful clues in our research about how choice can sort kids in different ways than if you just had neighborhood schools,” Glazerman said.
The researchers are working on a follow-up study based on 2015, 2016, and 2017 lotteries to understand more about how students’ own achievement levels affect their parents’ school choices, and how they do at their new schools.
They are also exploring how different ways of presenting information about the schools could affect parents’ choices. They created a mock website based on the MySchoolDC platform, with 72 different variations in how school information was presented. Some 3,500 parents of school-age children in all 50 states were randomly assigned to use the website and rank a selection of schools they would want their child to attend.
“We wanted to really understand how these really subtle factors affect how people engage with information,” Glazerman said. “We want to understand, could you get parents to make different choices by sorting differently, or using graphs versus numbers?”
The next studies are expected to be out in mid-fall.