## The Educational Achievement Gap, Race, Correlation, and Causation

I recently read a news source that was talking about education in the United States. In particular, it showed a graph that illustrated the mathematics achievement gap between white and black students. The article stated that there is an approximately 10% gap between races, and the difference has been relatively constant for decades. Then the article went on to make all kinds of conclusions related to school funding.

Because I’m deeply interested in education, and because I’m always skeptical of information found on the Internet, I went first to Wikipedia. The Wikipedia entry on “achievement gap” confirmed the disparity between races. But weirdly, the most recent data on Wikipedia entry was for the year 2004 — 13 years ago.

I wondered why the Wikipedia article didn’t use up-to-date data. So I went to the actual source of the data, the National Assessment of Educational Progress test administered by the U. S. National Center for Education Statistics. See https://nces.ed.gov/nationsreportcard/.

After a bit of navigating through their Web site, I found the raw source data for the years after 2004 (through 2015). I dropped that data into Excel and made a graph.

Well, the roughly 10% gap in achievement between races has in fact persisted with almost no change. Furthermore, the gap occurs in math, English, and science, and at all grade levels.

So, my next thought was wondering why the gaps exist. At a high level, such a gap can be explained by innate factors (genetic) or external factors (socio-economic), or a mixture of the two. The research, and common sense, pretty clearly indicate that the difference in achievement is a combination of innate and external factors. But there’s disagreement about what percentage of the difference is due to innate factors. The most convincing research I’ve read suggests that roughly 50% of the difference in achievement is due to innate factors.

Now the next question is, “What does this data mean?” Well, because this is just correlation data between race and educational achievement, you can’t draw any causal conclusions. All the data really says for sure is that there is an achievement gap, and that it has persisted for decades with little change.