A New School Ranking System for Chicagoland that Accounts for Parent Income

For those with short attention spans, here is the short version of the story: 1. School ranking websites utilize test scores to rank schools. 2. Test scores are mostly influenced by parent income, not the quality of the school. 3. As a result, people are selecting schools in privileged areas with high real estate prices, long commutes, and very little diversity. 4. We have developed a school ranking system that accounts for parent income and adds points to schools that have diversity. 5. We hope our system will uncover hidden gems that others overlook. Happy Hunting!

For all others, read below!

School Rankings are Flawed

Most ranking websites focus on the average test scores at a school in their ranking systems. The problem with this is that parent income is BY FAR the biggest driver of test scores, and as a result, people are choosing privileged communities (often in the suburbs with long commutes), with expensive real estate and little diversity.

For sure, some websites out there take into account other factors, but there is always a large component of the score that is derived entirely from the average test score at a school.

Many Parents Have the Same View

As a realtor, I have noticed that parents tend to have the same view when it comes to the “school score” published by greatschools: they want to see a 7 or better.  If a school is a 6, then that feels like a D and is probably not a good school.

Parent income is not part of the conversation when people are thinking about school rankings.

Consider a school located somewhere with a score of 7. Is that a good score? Without the context of parent income at the school, a 7 is actually meaningless. If the school is located in Kenilworth (one of the more expensive Chicago suburbs), then I’m not impressed with a 7…it should be higher. But if the school is located in the Pilsen neighborhood in Chicago…then WOW, what is this school doing right?

Parent income drives test scores, but most parents aren’t thinking this way, and as a result, they are overlooking schools that have low scores, not realizing sometimes the low scores are a result of socio-economic diversity.

Comparison to State Averages: Not Interesting

Greatschools.org compares a schools test score to the average test score for the entire state.  In Illinois, the average for the latest scoring data available through the US Dept of Education is 37%.  That’s 37% of kids are meeting or exceeding state standards on the annual standardized tests, on average. Greatschools compares the state average score to the score of particular schools and declares a school “good” or “worrisome” based on where the school falls in comparison with the 37% average.

We don’t like this method.

Why?

Because it ignores the single biggest influencer of test scores at a school: parent income.

Let’s start with the state average test score: 37%. The median household income in Illinois is $63,000 according to this link.  If your family is making twice the average household income, do you think 37% is an appropriate bar by which to measure your child’s performance on standardized tests?  Probably not. 37% is too low a bar, and as a result, many schools are being hailed as “great” when in reality, there should be some commentary about what the scores should be given the parent income at the school.

Conversely, if you make less than average household income, is 37% an appropriate bar by which to measure your child’s performance on standardized tests? Maybe not, it might be too high a bar and as a result, some schools are hailed as “bad”, not because the school quality is low, rather there is a higher percentage of children coming from lower income families.

A New Ranking System

Our view on school rankings is that parent income has be a factor, and using data science, we came up with a way to rank schools by normalizing for parent income.

For every public school with a neighborhood boundary throughout Chicagoland, we plotted test scores against parent income (in the form of % of students that qualify for free lunch), and the results were pretty amazing.

Click this link to see the actual table, you can hover over each dot to see the school name and other info.

Screen Shot 2019-12-04 at 2.51.38 PM

We segmented the data by grade, so for all schools that have a 3rd grade, this graph shows the test score trend as the % of parents that qualify for free lunch increases. No surprise, as parent income falls, so do test scores.

The trend lines have an equation.  The equation has something called the “R-squared”. I recall from a stats class that an “R-squared” of 0.2 means two things are probably correlated.  In this graph the R-squared between % free lunch and test scores is 0.60! MOST of the reason test scores are where they are are because of parent income.

Reversing a Myth

We segmented the data between City of Chicago Public Schools and Suburban Schools. Notice how the orange schools (suburban) are clustered in the upper left quadrant of the graph: high income and high test scores, while the City of Chicago schools are clustered in the lower right quadrant: low income and low test scores. This is why you hear people say “the good schools are in the suburbs”, because ON AVERAGE the suburban test scores are higher.

But when adjusted for parent-income, the City significantly outperforms the suburbs!

Notice the trend lines for the City of Chicago vs the Suburbs.  At all income levels, the City of Chicago outperform suburban schools. In fact, this trend is consistent in all grade levels through highschool.  You can see for yourself at this link.

Our Ranking System

Our ranking system calculates the difference between the actual test scores and the expected test scores and ranks schools by this difference. In this way, we are leveling the playing field, effectively lowering the bar for schools with more disadvantaged students, while raising the bar for schools in higher income communities.

We also add points to schools that have diversity. Our diversity calculation adds anywhere from 5 to 40 points based on the number of races represented at the school. For example, to get 40 points, a school must have at least 3 races that EACH represent 20% of the student population of the school

You can see the results of our ranking system for all Chicagoland Schools at this link.

When it comes to school rankings, there is no perfect system, and there is no substitute for visiting the schools in person. But hopefully our new ranking system will uncover schools that you might have otherwise overlooked.

 

 

 

 

 

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s