Educational Redlining: How Zillow’s School Ratings Help Segregate Communities

Share this:
LinkedInGoogle+EmailShare

Heights Community Congress has just released Educational Redlining: How Zillow and GreatSchools Profit from Suspect School Ratings and Harm Communities,  a report on the practice by Zillow, the real estate website, and GreatSchools to guide home buyers to choose communities according to color-coded school ratings posted online. Heights Community Congress (HCC), founded in 1972, is Greater Cleveland, Ohio’s oldest fair housing enforcement organization.  For over four decades HCC has been conducting audits of the real estate industry to expose and discourage racial steering and disparate treatment of African American and white home seekers.  This blog covered HCC’s preliminary work on educational redlining here.

Ralph Day, author of HCC’s report, explains that GreatSchools, launched by the venture capital group, NewSchools Venture Fund, “arbitrarily divides schools into three categories—green, yellow, and red.”  In most states GreatSchools’ ratings merely represent the aggregate standardized test scores of each school’s students.  Day continues: “The act of coloring red whole communities of schools is an alarming reminder of mortgage redlining of recent past, which was declared illegal with the passage of the Fair Housing Act in 1968.  Anecdotal evidence suggests that many home seekers automatically exclude whole communities with ‘red’ labeled schools from their search.  And Zillow in its book (Zillow Talk) exacerbates this trend by cheerfully steering home buyers into the highest rated (highest income) GreatSchools areas they can afford.”

Day summarizes GreatSchools’ methodology:  “In most states, GreatSchools rates schools on their standardized test scores alone.  Schools of the same grade levels are ordered within their state from high scores to low.  Then a school’s rank is converted to a number 1 through 10, where 10 means the school is in the top 10 percent, 1 the lowest 10 percent, 5 and 6 about in the middle, and so on.”  Day continues: “In the past, mortgage redlining was practiced to systematically withhold financing from certain communities that lenders chose for disinvestment… Educational redlining recycles this pernicious idea by taking aim at schools, and discouraging home buyers from purchasing homes in moderate income communities.  Because school test scores have repeatedly been shown to highly correlate with a community’s socio-economic status, and GreatSchools ratings draw heavily on test scores, the effect of coloring certain schools red is to disinvest in moderate income communities by discouraging home buyers… This ignores the obvious, that many dimensions contribute to school quality, and… metrics can only begin to describe it.”  And, according to Day, “GreatSchools makes money by licensing its ratings to third parties like Zillow. It also apparently makes money by linking its viewers to Zillow.”

What do the ratings miss?  “GreatSchools’ ratings are simplistic.  They are almost useless for determining a school’s overall quality or its suitability for a particular child.  They do not begin to capture a school’s activities and specializations that matter to children and parents.  Furthermore, by focusing only on the average score, GreatSchools obscures the fact that a group of students may be doing very well, another group may be greatly improving (high value added), or that students in a specialized subject area are quite strong.”

Zillow’s invitation to home seekers that they look to buy in a school district with a “green” rating and avoid schools rated “red” surely exacerbates growing residential segregation by income across America’s metropolitan areas, a trend documented by Stanford University sociologist Sean Reardon.  Reardon has shown that the proportion of families in major metropolitan areas living in either very poor or very affluent neighborhoods increased from 15 percent in 1970 to 33 percent by 2009, and the proportion of families living in middle income neighborhoods declined from 65 percent in 1970 to 42 percent in 2009.

Heights Community Congress declares: “HCC realizes that the best defense against misinformation is an educated public… HCC believes that GreatSchools, its supporting foundations, and Zillow have a civic responsibility to refrain from actions that harm communities… GreatSchools must abandon the use of red, yellow, green colors… Zillow must clearly disclose how the ratings are calculated and what they mean… Zillow must state that there is no such thing as a single measure of school quality… Zillow must stop making judgments about school systems and drawing comparisons among school districts, actions that are inappropriate for a real estate business with no experience or expertise in education.”

In a speech at the Cleveland City Club last February, Richard Rothstein of the Economic Policy Institute explained how—because test scores reflect primarily the aggregate economic level of the families in a community—rating and ranking schools drives segregation by income.  Rothstein speaks of the “A” through “F” letter grades that are frequently these days assigned to schools as they are rated and compared.  The “A” through “F” school “grades” function similarly to the red, yellow and green rankings used by GreatSchools and Zillow: “These rating systems really just describe the social class of the students in the schools. And the high ratings don’t necessarily mean they’re better schools. Many of these schools that are rated ‘A’ because they happen to have a lot of middle class children with highly educated parents may add less value to their students than schools rated ‘F,’ where parents may be working the kind of contingent schedules I described earlier. Those ‘F’ schools may actually be better schools in terms of what they add to students than the ‘A’ schools, but most people don’t understand that. And so if you label schools with ‘A’-‘F’ ratings, people who attend a ‘C’ school, which may be integrated, are going to want to move their children to an ‘A’ school.  This will increase the segregation of schools by convincing people that these ‘A’-‘F’ ratings accurately reflect the quality of the school.”

I urge you to read and consider HCC’s new report on educational redlining. Think about the report in the context of demographic trends you can observe across a metropolitan area with which you are familiar.

Comments

  1. Brian Wagner says

    “The act of coloring red whole communities of schools is an alarming reminder of mortgage redlining of recent past,”

    Yes, because giving consumers information they desire in making their own free choices is EXACTLY the same as denying them needed loans based on closely held information which they are not provided.

    “HCC believes that GreatSchools, its supporting foundations, and Zillow have a civic responsibility to refrain from actions that harm communities”

    Harm by whose definition? The same crowd who also define words as violence?

    “GreatSchools must abandon the use of red, yellow, green colors”
    “Zillow must state that there is no such thing as a single measure of school quality”
    “Zillow must stop making judgments about school systems and drawing comparisons among school districts”

    So much for liberals valuing free speech, eh?

    “Zillow must clearly disclose how the ratings are calculated and what they mean”

    Obviously they already do, since you have that information, or did you abduct and waterboard some Zillow employees to get it?

    “explained how—because test scores reflect primarily the aggregate economic level of the families in a community—rating and ranking schools drives segregation by income”

    drives, or facilitates? Correlation is not causation. Income segregation is driven by people’s innate desires, and those with means will find the information they need one way or another.

    Zillow is giving people what they want.

    “These rating systems really just describe the social class of the students in the schools.”

    And did it ever occur to you that their users are actually interested in the social class of the students?

    “And the high ratings don’t necessarily mean they’re better schools. Many of these schools that are rated ‘A’ because they happen to have a lot of middle class children with highly educated parents may add less value to their students
    than schools rated ‘F,’”

    You’re conveniently ignoring the value of peer influence – one’s fellow students are a large component of the value added. One of the reasons given in support of court-ordered desegregation has been that low performing students will
    benefit from exposure to high performing students. Well, then the best possible school is the one that has “a lot of middle class children with highly educated parents.”

    In summation, you’re demanding the suspension of the First Amendment to prevent the distribution of information so

    that you can trick middle class educated parents into immersing their kids in a pool of bad influences.

  2. says

    Thank you for writing this Jan. It points to the reality that algorithms allow all sorts of fascinating things to be done without a human touch, and in so doing, they hide all of the assumptions and nuances that go into making them work.

    • Brian Wagner says

      Funny Bob, but the last person I heard say something like that was someone you’d probably call a climate denier.

      I think the algorithm does exactly what it aims to do and measures exactly what its intended users desire. The problem is that Jan, and presumably you, don’t approve of the users’ goal, which is to place their kids in a school with other middle class kids with educated parents.

Leave a Reply