Which factors correlate with per pupil spending rates?

Photo by moren hsu on Unsplash

At a recent family gathering, I joined a conversation about schools, centering around one question: why do some school districts spend so much per student? There were easy culprits to blame, depending on your ideology. Bureaucrats or politicians; incompetence or greed. But what do the numbers say?

I decided to look for relationships between per pupil spending (PPS) and a few different factors:

• Enrollment
• Median Income
• Poverty Rate
• Cost of Living
• Median Teacher Salary

The github repo for this project can be found here.

Some of these metrics are tracked at the county level, some at the state level. County-level education spending data were found here, while state education spending was found here.

County-Level Measures

I first looked at enrollment and PPS for a relationship:

Most districts are clustered at the low-end of enrollment.

It turns out that enrollment (the size of student body) doesn’t vary much, so that didn’t provide much information.

Next, I looked at household median income and PPS:

While median income varies more, still no clear relationship.

The final county-level metric I looked at was poverty rate:

Again the poverty rate varies considerably. But this alone can’t tell us much about spending per student.

State-Level Measures

I couldn’t find cost of living indices for county level data, but it was available for states. Here is that measure related to PPS:

A diffuse relationship is visible.

Finally, I looked at teacher salaries:

A linear relationship is visible.

And with the line of regression:

A clear linear relationship is visible (Pearson’s R = 0.79)

This is the clearest relationship so far. (It should be noted that these salaries aren’t adjusted for standard or cost of living; analyses which incorporate these adjusted still show big differences in compensation.)

Conclusion

It’s not a big surprise that teacher salaries are related to per student costs; other analyses on this topic have explored potential reasons for salary differences, whereas others looked into other factors altogether, like the amount of special-needs students, or the number of support staff.

Many of these analyses come from a place of skepticism (slight “get off my lawn” vibes). It’s important to be critical of how tax dollars are spent, but that needs to be balanced with the knowledge that teachers in general are overworked, underpaid, and under-appreciated in the United States. Let’s be critical, but with an eye towards improvement and enhancement, not reduction for its own sake.