The online sitehas released a massive expansion of its school-ratings system, intended to provide more nuance about school quality by highlighting student test growth, achievement gaps, school climate, and course offerings.
The expansion comes as GreatSchools and other school-ratings sites ride skyrocketing popularity, and as states work to incorporate more multifaceted measures into their own school accountability systems. But the changes also respond to ongoing concern by education and fair housing advocates that test-focused school-ratings sites can push users to overly simplistic or even stereotypical judgments of schools.
“[Ratings-site data] often gets used at the individual level—I’m going to talk to my friends and pick a school that’s right for my children—but these data need to be about how we can improve a system ... not individual winners and losers,” said Rebecca Jacobsen, an associate professor for education politics and policy at Michigan State University who studies how parents interpret and use education data.
GreatSchools previously provided ratings based primarily on test scores and parent feedback for more than 130,000 public and private pre-K-12 schools nationwide. As simple as they were, those ratings were used widely by parents and real estate-related companies, some of which linked to the ratings on their websites.
Beginning Thursday, the site will also rate schools based on student growth from year to year, and, for secondary schools, on graduation rates, Advanced Placement offerings and enrollment, and students’ average performance on college placement tests. It also includes an “equity” rating, based on whether there are achievement gaps for students from racial minority groups, those in poverty, and those with special needs, as well as on how students in those groups fare compared to similar students statewide. The revamped site also uses federal civil rights data to flag schools that show high rates of student absenteeism or disproportionate discipline for different groups of students.
“What we really want is for people to see the top-line data and dive in and explore it more,” said Adam Gorski, GreatSchools’ lead data scientist. “We hope it will more accurately reflect what’s going on in a school besides just its demographics.”
The new ratings, which were piloted in California this summer, will be launched in 42 more states and the District of Columbia this week. The remaining states—Alaska, Idaho, North Dakota, South Dakota, Maine, New Hampshire, and Vermont—will be updated once the site has additional school data.
“We’re looking at how can we turn the data into information that’s really designed to help parents to make sense of it—and what’s the insight on what you can do about it,” said Matthew Nelson, the president of GreatSchools.
The San Francisco-based nonprofit GreatSchools mirrors the rise of other social media review services like Yelp. Its growth has been spurred by unprecedented access to school accountability data online, including its partnerships with real estate sites and the Housing and Urban Development Administration, which shares the school ratings with families who use housing vouchers. GreatSchools has expanded from 44 million users in 2013 to more than 50 million users today.
John Luedemann is one of those users. Two years ago, he and his family faced a move from Houston, Texas, to Houston County, Ala., a few months before his daughter Morgan was scheduled to start kindergarten.
The online school-ratings site GreatSchools.org has redesigned its 10-point evaluation system, to incorporate more measures of student growth, academic access, and efforts to reduce disparities among different student groups. Here is a breakdown of what goes into the new rating:
- Academic achievement: Includes the website’s pre-existing evaluation of state test scores in math, reading, and science for tested grades, as well as a new measure of student growth, based on either state-reported growth measures or an in-house measure of cohorts of students from year to year.
- College Readiness: Includes state and federal data on a school’s four-year high school graduation rate, average ACT/SAT college placement exam rates, and the percentage of high school graduates who meet entrance requirements for local state university systems.
- Advanced Coursework: Includes the average number of advanced courses each student takes in core academic subject categories—English; science, technology, engineering, and math (STEM); social studies; and foreign languages—as compared to the state average.
- Equity: Includes how low-income students and those from traditionally disadvantaged racial or ethnic groups in that state perform on state tests as compared to students from more traditionally advantaged groups in the school, as well as the same student groups statewide. This section also includes data on achievement for students with disabilities, but they are not included in the overall rating.
- Environment: Includes the demographics of the school, average teacher-student and counselor-student ratios, average teacher experience, and parent reviews. This section also flags schools with high or disproportionate rates of student absenteeism or school discipline.
“We had no family there, no connection there, no previous knowledge of the schools. You kind of have to start from scratch with nothing to go on,” Luedemann said.
So Luedemann narrowed his search based on the GreatSchools ratings available through his home searches on Zillow and other real estate sites. “There was a pretty big discrepancy; a handful that were 7 and above, and then the rest were like 4; there was an obvious fall-off” on GreatSchools’ 10-point scale. “I tried to drill down more ... but I don’t remember there being a whole lot more.”
Critics point to divides like the one that Luedemann found for Dolthan, Ala., as a downside of the site’s previous, primarily test-score-based school ratings, which they argue can be a better proxy for a school’s race or poverty demographics than for school quality. Lisa Rice, the executive vice president of the National Fair Housing Association, noted that some of her group’shad found real estate agents describing the same schools using different metrics to white and Latino home buyers, recommending a school to Latino parents that they had panned to white parents.
“The information they are providing gives only a small snapshot of each school, and no one should make a decision to send their child to a school based only on the data in these websites,” Rice said.
Luedemann agreed—he and his wife visited both neighborhood and magnet schools in the area before finally settling on a list of homes—but he said the site can help provide a baseline for parents who are new to a school system, either because they have young children or are entering an area.
Jacobsen said she has found thatabout schools can help parents make better decisions about their children’s education. In a study of parents reviewing school data, Jacobsen found participants used different—and often more detailed—data to analyze a school they were familiar with than they did for one they had never heard of. “When good additional data gets added, we’ve found people begin to match the experiences of parents with children who actually attend the school,” Jacobsen said.
She cautioned, however, that when multiple, detailed ratings are presented—particularly number ratings—on a ratings site. Both Jacobsen and Rice argued for more practical information on review sites that would allow parents to have conversations about school improvement.
GreatSchools, at least, seems to be trying to move in that direction. Nelson noted that GreatSchools’ expanded reporting also will include more tips for parents on how they can talk to school administrators about flagged problems, such as high chronic absenteeism for a particular student group. Eva Orbuch, the lead community organizer for the San Francisco Bay-area advocacy group Innovate Public Schools, said she has started using the site’s expanded data in workshops with mostly low-income and monolingual Spanish parents. “The equity rating has been very helpful,” she said. “The low-income number is often a big difference from the overall number, and that’s often a big wake-up call for people.”
While California’s state education site also provides student achievement data, Orbuch said the private data site has been easier for parents to use on their phones while meeting with administrators.
“I work in schools where there is sometimes an extreme achievement gap between white and nonwhite students” she said. “Now that the parents are regularly using data, I can see how empowered they are to say no, the school really is not making growth here. ... It does complicate our work, but it has definitely provided a lot more questions for parents to ask and feel like they’ve done their homework in talking to schools.”
A version of this article appeared in the November 15, 2017 edition of Education Week as GreatSchools Expands Its Ratings on Schools