How Ethnically Representative is your School?

Los Angeles is a diverse city. LAUSD mirrors that diversity, as you can see by the enrollment percentages below:

However, Los Angeles is also a prime example of what sociologists call “de facto” segregation. Schools are a mirror of that segregation. Many schools pull enrollment from their local areas, which are segregated and they cannot hope to representative the broader citywide diversity. It is not a school’s fault if they are segregated or unbalanced. Still it, is pretty interesting to marvel at the segregation – to see the patterns that exist and consider its affects.

I think I have found a way to measure how much a school represents LAUSD. Let’s start with this assumption: The optimally representative school would have the same ethnic balance as the district as a whole. In other words, it would be close to 72% Latino, 9.6% White, 8.6% African American, 5.6% Asian and so on.

Of course, the only way to do this would be bussing (ooo…I said a bad word). But yeah let’s go there –  considering all the hype surrounding Magnet Schools, which are an opt-in bussing program, the idea isn’t too far away. Let’s imagine that we are trying to get all schools to be optimally representative of Los Angeles. 

Using this idea of optimal representation, we can measure how far away a school is from the district’s diversity. I am going to call it the Optimal Representation Deviation Percentile, because the method I used to create the measure is similar to a standard deviation.

I will explain the boring methodology for the end of this post, but basically, I ranked all the schools based on how close they are to the ethnic representation of LAUSD. A 100% would mean a school is similar to the district as a whole, while a 0% would mean they are far away from the percentages of the district.

Behold: Optimal Representation Deviation Percentiles for all LAUSD schools….(click school name and it will alphabetize)

There are a lot of interesting patterns that pop up in the data. I will discuss those patterns tomorrow.

Methodology of the Diversity Deviation Percentile (For All You Nerds)

Statisticians often measure deviation from a number by finding the difference and squaring it. What I have done is found the enrollment percentage from each ethnicity at each school, found the difference from the district’s overall diversity, and squared it. I then add them together, and square rooted the number. The result was a number that measured how far off the school deviated. Then I ranked them using a percentile ranker.

For example, according to this metric, the most ethnically diverse school is Lemay Street Elementary – whose enrollment by ethnicity looks like this: Screen Shot 2016-06-21 at 3.05.25 PM.png

As you can see, this school’s percentages are very close to the LAUSD overall percentages. Compare that to some of the lower ranked schools:

Screen Shot 2016-06-21 at 3.10.27 PM.png

These schools are predominantly White or African American, and are therefore not representative of the district.