With more than 40 million unique visitors each year, the school ratings site GreatSchools.org is an undeniable hit with users.
The site’s 1-10 ratings of schools is often among the first results that come up when searching online for a school by name. GreatSchools is also licensed by real estate listing sites such as Redfin and Zillow, which allow potential homebuyers to easily see the school zones for houses for sale, or permit users to search for houses for sale that are zoned to a particular school. In addition to the ratings, the site also collects reviews from parents, who can offer their own star ratings.
But what do such ratings mean to the schools themselves?
Nationwide, the population of public school students is becoming more diverse, and a growing percentage of schools are predominantly black and Hispanic, with high concentrations of students eligible for free or reduced price lunch. Such schools tend to have fewer experienced and effective teachers, a more challenging school climate, and less rigorous academic offerings than schools with a wealthier student body.
GreatSchools, some argue, has become a player in the ongoing debate about school equity.
One research paper receiving renewed attention suggests an earlier version of the GreatSchools scoring system may have accelerated already existing trends of racial and socioeconomic segregation.
To be clear, that segregation existed well before GreatSchools came on the scene, said Sharique Hasan and Anuj Kumar, who examined demographics and housing trends between 2006 and 2015, when GreatSchools was expanding its reach.
But, they argue, more-affluent families have a greater ability to leverage the ratings provided by GreatSchools when they’re making a decision on where to live, they say.
“Knowledge is indeed power, but only for the powerful,” said Kumar, a professor of information systems at the University of Florida, in a recent lecture. He studies how information technology affects the behavior of organizations and individuals.
“Unless appropriate housing and educational policies are designed to specifically address the differentiated use of online school ratings, I’m afraid it will create further divides in our society in times to come,” Kumar said.
Hasan and Kumar looked at demographic changes and housing prices during the time that GreatSchools was growing to incorporate ratings of schools throughout the country. At that time, GreatSchools ratings related almost entirely to test scores. The researchers were able to compare demographic information in ZIP codes that had high-performing schools and had been rated by GreatSchools, to similarly high-performing ZIP codes that had not yet been evaluated by the site.
ZIP codes with schools performing above academic averages based on standardized test scores were already more affluent than ZIP codes where the schools were average performers. Those above-average ZIP codes also have a higher percentage of white and college-educated residents.
But when GreatSchools rated those communities, the neighborhoods became even more concentrated with white and Asian, and college-educated residents and families earning more than $100,000 annually. In addition, the Hispanic population of those communities went down, but the black population remained statistically unchanged before and after GreatSchools rated those highly performing schools.
In an interview, Kumar said that the analysis is not meant to be a criticism of GreatSchools itself.
“They’re just putting out information for people. They’re not policymakers,” he said. “Providing school information, is it a bad thing? Absolutely not, it is a good thing. It is just an unintended consequence,” he said.
Other research on different school rating systems have shown similar effects on housing values. A 2004 study looked at home prices in Gainesville, Fla., after the state moved to an A-F grading system for schools. Homes that were zoned for schools with an ‘A’ grade were priced about $9,000 more than similar houses zoned for ‘B’ grade schools, even though the actual academic differences between schools with those grades were minor, that study found.
School districts are heavily dependent on property taxes, which are driven by home values. Schools in areas with high-priced homes generally have more money to spend on teachers and other educational resources.
A Principal’s View of GreatSchools
Paul Kelly, the principal of Elk Grove High School in Elk Grove Village, Ill., a Chicago suburb, says he’s seen how ratings affect the perception of schools in his community. His high school is rated a 9—nearly at the top of GreatSchools’ scales. The school of 1,900 students is 44 percent Hispanic, 41 percent white, and 10 percent Asian. About 30 percent of the students qualify for free or reduced-price lunches.
However, the elementary and middle schools that feed into Elk Grove High have much more modest ratings, from a low of 2 to as high as 6. Potential home buyers are told to avoid those schools, but his own school’s high rating is a testament to the quality of the education those elementary and junior high schools are offering, Kelly said.
“I have absolutely heard that the high school is great—how is it those sender schools are so bad?’ ” Kelly said, recounting conversations by others.
But as much as he’d like to take the credit, Elk Grove High School alone is not transforming those kids into academic superstars, he said. The feeder schools “get judged on where [the students] used to be,” said Kelly, the 2018 Illinois principal of the year. “We get judged on where they ended up.”
Said Kelly: “By the time they leave us in high school, we have had a lot of time to build on the strengths that they developed before their scores got so great.”
The conversations Kelly describes, steering homebuyers to one school zone or another, were vividly outlined in a recent investigation by Newsday, which sent potential homebuyers of different races and ethnicities to real estate agents, then secretly recorded what the agents told those buyers.
Many of the agents used school district ratings as a selling point, or a marker for areas potential buyers should avoid. A white potential homebuyer was steered towards homes in Massapequa, N.Y., that are served by the Massapequa Union Free district, which is more than 90 percent white, with high test scores and graduation rates. That same potential buyer was told to avoid homes for sale in the portion of Massapequa that is served by Amityville schools, where more than 90 percent of the students are black or Hispanic, and test scores and graduation rates are lower.
The agent did not offer any assessment of the school districts to the black potential homebuyer.
In an interview, Kumar suggested two potential policy changes that could lessen the segregration trends found in his study: changing school boundaries to equalize access to high-performing schools, as well as incorporating more measures in a school rating than just test scores, which are highly correlated with race and family wealth.
School boundary changes are often political land mines for school districts. But incorporating more measures into school ratings is just what GreatSchools is doing, said Jon Deane, who has been chief executive officer of the site since February.
In 2017, the site started building in a variety of data points other than test scores, including student academic growth when available, graduation rates, and Advanced Placement offerings and enrollment.
The ratings also incorporate “equity scores” that are weighted based on gaps in academic performance for students who have disabilities, are minorities, or are from low-income families. The scores, based on state data, appear on the information page for a given school. The equity scores are also incorporated into a school’s final 1-10 rating.
A recent investigation by the news organization Chalkbeat suggests that the equity score has had some effect in reducing the correlation between GreatSchools ratings and school demographics, but the highest-rated schools still tend to have more white and Asian students, and fewer students eligible for free or reduced-price lunches.
Deane said the parent users have said they’re looking for a richer analysis of school data, and the site’s web traffic proves that users are looking for that information.
“Our mission has always been to support parents with the best information possible so we can help them support their kids,” Deane said. “We recommend that parents take that information in context with all the other things they do to support their child—visit the school, start up a conversation around what’s working and what isn’t in their community.”
Deane also says that it’s not just higher-income families who are able to make choices based on GreatSchools ratings. The site has partnered with a researcher at Columbia University, Peter Bergman, who has found that families with housing vouchers who are given GreatSchools information were more likely to choose neighborhoods with higher-rated schools.
Jose Arenas, the vice president of organizating for Innovate Public Schools in California, works directly with low and moderate-income families in Los Angeles and the Bay Area. GreatSchools is one of several data sources that the organization uses with parents in order to help them as advocates, as well as to make better informed decisions for their own children.
“People should have access to information. As an organization, we would agree that a number doesn’t say everything—but it says a lot,” Arenas said. And these are families just as capable of making nuanced decisions as any other family, he added.
“To suggest that only educated progressives can do that is inherently racist. I don’t think anybody looks at data that bluntly. Our parents understand school systems are complex,” Arenas said.
GreatSchools argues that it needs better data from states in order to make its own ratings more nuanced.
“We see information as the best way to drive equity, and hold up the lens on what’s happening across the country,” Deane said. “We just want to make sure we don’t lose sight of this here.”
Kelly, the high school principal in suburban Chicago, says that the ratings themselves only tell a narrow part of a school’s story. He has spoken out in support of his educator colleagues in the elementary and junior high schools that feed into his high school.
“I look at the good colleagues and good people who work every bit as hard I do and care every bit as much. They get lit up, they get criticized by folks because of one aggregate number,” Kelly said. “They do not tell you about the quality of the program, or the place.”
A version of this article appeared in the January 15, 2020 edition of Education Week as Is GreatSchools.org Making Segregation Worse?