Correlation does not imply causation

One of my college professors taught us in Basic Statistical Analysis that correlations can lead to mistaken inferences about causation.  His analogy was classic.  He told us in the mid-60's that he knew what was causing Polio.  He showed us two charters; one showing the distribution of incidents of Polio throughout the year and the other showing the consumption of Bubble Up (a popular soda drink) during the year.  The two graphs were identical.  "Therefore", says the professor, "Bubble Up causes Polio."


The topic came up in a Nov 2013 article in Nature magazine.  The authors of this article put it this way:  "It is tempting to assume that one pattern causes another. However, the correlation might be coincidental, or it might be a result of both patterns being caused by a third factor — a 'confounding' or 'lurking' variable. For example, ecologists at one time believed that poisonous algae were killing fish in estuaries; it turned out that the algae grew where fish died. The algae did not cause the deaths2." (Borsuk, M. E., Stow, C. A. & Reckhow, K. H. J. Water Res. Plan. Manage. 129, 271–282 (2003)


So, when we set out to interpret educational data, we need to beware of making "confounding" conclusions or inferences from correlation data.  Whether it's SAT or ACT score trends or distribution in the locations of poorly performing schools, we need to be sure we are not "extrapolating beyond the data".  


http://www.nature.com/news/policy-twenty-tips-for-interpreting-scientific-claims-1.14183