(This story is continued from the first article here)
Okay, so I’ve explored briefly a dataset on public schools in Massachusetts1. Specifically, I detailed some of the racial distributions of students on maps. How then, do these racial distribution (as well as other factors) affect play a role in shaping the students’ education? What are some trends between race/resources/family condition and academic performance (in this case, SAT scores)? Since the dataset included SAT performance data per school, I decided to look at that first. Recall that I had once analyzed SAT performance in CT compared to wealth indicators (here). These trends only corroborate my previous findings. Here are a few that were surprising, unexpected, and worrying to me.
Here’s % of students economically disadvantaged2 versus their SAT performance (light and dark bubbles indicate % attending college):
(r-squared = 0.731239, a very high correlation)
What about race? Here’s comparing the percentage of white students in a given school to their performance:
And here is the same chart, but comparing against % hispanic students in each school:
And if that doesn’t convince you that standardized testing is racially biased (to a certain extent), then maybe this will. Say this bias is truly an economical reason, and schools aren’t able to provide an adequate education to the students living in less fortunate districts, then we’d see a strong correlation between how much each individual school is able to provide for the children against how well they score on tests like the SAT. However, this does not seem to be all that accurate. When we compare the average expidentures per student versus how well they perform, there is a very weak correlation3.
This signifies that it is not actually much of the school’s economic influence on the students that ultimately affects the student’s performance, but is the student’s family background or upbringing. Think about that: non-school factors affecting performance on a standardized exam meant to test *only* for academic/scholastic aptitude.
- Massachusetts Public Schools Data
- According to MA DoE: “Calculated based on a student’s participation in one or more of the following state-administered programs: the Supplemental Nutrition Assistance Program (SNAP); the Transitional Assistance for Families with Dependent Children (TAFDC); the Department of Children and Families’ (DCF) foster care program; and MassHealth (Medicaid).”
- An r-squared value of approx. 0.07