PISA Test Data Reveals a Predicable Correlation

The recent release of international test score data is causing a predictable outburst of commentaries.  The US remained pretty much in the same place it has always been on this test.  But, there’s an interesting correlation in the comparison of  test scores to levels of poverty.

Tracking the PISA Math Literacy scores and plotting them as individual points on a graph and adding the data points for the percentage of Children in Families below 50% of Median Income produces a predictable correlation.  As the percentage of families in the “below 50% median income” increases, the PISA Math Literacy score for that nation tdeclines.  If you draw a line through the separate data points, establishing the mean score trend, the PISA score line descends from about 520 to about 450.  If you combine the individual data points with the trend line, you get a picture that shows most countries fairly close to the trend line, with some outliers both above and below the line. 

•  All countries on the graph have a lower percentage than the US of children below 50%, except Turkey, Chile, and Mexico.  That alone is a stunning commentary of the US. 

•  All three of those countries are far below the PISA scores trend line, while the US.is well above the trend line.

•  Denmark has the lowest average of children below 50%, but it’s PISA score is below the trend line.

•  Finland is the second lowest average of families below  50% and is dead on the trend line.

•  Korea and Japan are outliers on the positive side of the graph; their PISA scores were first and second, with about 10% (Korea) to 14% (Japan) of their children below the 50% line.

•  Greece is an outlier on the negative side of the graph.  With about 13% of their children below the 50% line, their PISA score in the fourth lowest.  Only Turkey (23%), Chile (24%), and Mexico 26%) have lower scores.