Research Paper On Simpson Paradox

When you hear the name “Simpson,” who comes to mind? Is it Homer and Marge, or is it O.J.? Most people hear “Simpson” and either think of the popular American sitcom or the NFL running back turned alleged killer. I am unique, because when I hear “Simpson,” I think of Edward H. Simpson, the man who popularized Simpson’s Paradox. Simpson’s Paradox is a statistical phenomenon that occurs when a trend appears in groups of data, but when the groups are combined, the trend appears to be reversed. The most famous example of this occurred in 1973, at the University of California-Berkeley. The school was sued for sex-discrimination when it appeared that they admitted 44% of male applicants, but only 35% of female applicants. People were outraged, until

statisticians examined the data more closely. It was discovered that the acceptance rates for women were higher than they were for men, in four of the six departments. This was not represented in the overall acceptance rate because the more-selective departments received far more female applicants than they did male. I was first introduced to this example during my junior year, in AP Statistics. I have maintained a strong interest in this topic because I am unsettled by the fact that the true results of a study could be so different than what is presented to the public. People put so much trust into the statistics they see, and to know that they could be so wrong is very disappointing to me. After learning about Simpson’s Paradox, I realized that I wanted to one day have a career involving statistics. I want to be able to work with data to insure the public is properly informed. Simpson’s Paradox has both proved my uniqueness and inspired my intended field of study.