can you understand this scatterplot?

The Atlantic Monthly published this damning scatterplot of school district funding in Pennsylvania. The dots are color-coded by the racial composition of the districts.


What this means is that, at any given income level, whiter schools are better-funded. It’s not just that black people happen to be poor, which is also true. It’s that, for a given level of poverty, the distribution of funding is still racist. The point of “colorblindness” as a political agenda is to prevent the collection of the data in the scatterplot. Since we have the data, it should become a part of consensus reality that there’s systematic racism against children in the state of Pennsylvania.

That’s unlikely to happen, because about half the country doesn’t have the education to understand scatterplots. Just recently, Pew Research Center asked American adults 12 questions about science, including this one:


63% got it right. Notice that the plot of school funding and race is more complicated because there are two independent variables, one of them on the X axis and the other one color-coded. From experience, I know that psychology undergraduates taking a research methods class have difficulty interpreting the results of experiments with 2 independent variables. “Do a 20-minute in-class experiment, do an ANOVA with post-hoc tests in SPSS, write an APA format paper about it, and then do a group PowerPoint presentation” takes an entire quarter, and the students had to pass statistics the previous quarter to be there. They’d already completed a lot of their general ed requirements, since they needed the class to declare themselves psychology majors (i.e., had already been in college for at least a year). Psychology was the most popular major on campus.

What that tells me is that scatterplots are difficult and possibly meaningless for the average person, given the state of math education in this country. It troubles me that I can’t think of a clearer way of displaying the data. Those are the racial compositions of the schools, and the amount of poor children attending thoes schools, and the funding levels of those schools. The relations are depicted in the scatterplot. It’s not like a table of numbers would make the point better.

Stated plainly: half the public isn’t capable of understanding the current state of empirical reality. They have intuitions and ideological preferences. There’s no sense that there even is an objective, nonpartisan reality in need of explanation and fixing. “You can make up charts that say anything. I’m a proud skeptic!” Actually, that person is very gullible n00b.

It’s almost like the left should drop everything it’s doing until we’ve done something about education and people can understand scatterplots.