Tuesday, March 06, 2012

77% of all statistics are either misleading or made up

I heard a news story on the radio this morning that made me stop and think. The story reported that a new study showed that African-American students in the Chicago Public Schools system are more likely to receive harsher discipline than other students. The reporter went on to quote some statistics from the study. It was at this point that I paused. Did the statistics actually report what the study say they reported? Here are the statistics taken from the Chicago Tribune version of the story:

"African-American students (approximately 45 percent of students) comprised three-quarters (approximately 75 percent) of school suspensions in 2009-10, the year data was collected for the national civil rights survey. Latino students made up 42 percent of CPS' enrollment but 20 percent of the suspensions. White students, who represented less than 10 percent of district enrollment, made up 3 percent of suspensions."

Statistics showing that 45 percent of the students getting 75 percent of the punishment seems to validate the studies findings. But is that what those statistics actually tell us? Hearing a quick story on the radio or even reading a brief summary from the newspaper don't always give us all the facts. So my next step was to look at the original study to see what other factors were looked at. But unfortunately the source website is down for maintenance which means that we can only deal with the statistics at hand. So do the reported statistics actually tell us that African-American students in the CPS are more likely to receive harsher discipline than other students? Are there limitations to what statistics can tell us?

Yes there are limitations and no–as reported–the statistics do not tell us what the news reports are saying that they are telling us. Let me take a moment and be very clear. I am not commenting on whether or not African-American students are more or less likely to receive harsher discipline than their fellow students of other ethnic/racial backgrounds. My only comment is on whether or not the statistics–as reported–determine either a correlation or a causation.

In other words these statistics cannot show that African-American students are being subjected to harsher punishments because they are African-American students. The only thing that they actually show is that about 75 percent of the students that are being suspended are African-American which is a different percentage than the total population makeup of the schools. This means that African-American students are more likely to be suspended. In order to make the judgement that African-American students are being subjected to harsher punishments we would need to look at the punishments being given out for various misbehaviors.

This is one of the great problems with simply quoting statistics. We always need to look at statistics with a critical eye and ask if they are actually saying what we are being told that they are saying. Then we need to do the hard work of asking why these statistics are the way that they are. It is only then that we can get at the underlying problems.

There is no doubt in my mind that the educational system in Chicago is not functioning as best as it can. It seems to me that African-American students being more likely to be suspended is no less of an issue than African-American students being subjected to harsher discipline. The real question is why are they being suspended at a greater rate?


No comments:

Post a Comment