Simpson paradox (Simpson &; #39; Paradox) has also been translated into Simpson's paradox, which was put forward by the British statistician E.H.Simpson in 195 1, that is, two sets of data under certain conditions will meet certain properties when discussed separately, but once considered together, they may reach opposite conclusions. When people try to explore whether the enrollment rate of freshmen is related to gender, and whether the compensation is related to gender, they will study in groups respectively. Simpson paradox is a phenomenon that sometimes appears in this kind of research under certain premise. That is, the dominant party in group comparison will be the one who loses power in the overall evaluation. This phenomenon was discussed in the early 20th century, but it was not officially described and explained until E.H. Simpson published a paper in 195 1. Later, this paradox was named after him. In order to avoid Simpson paradox, it is necessary to consider the weight of each group and multiply it by a certain coefficient to eliminate the influence caused by the difference of group data cardinality. At the same time, we should clearly understand the situation and comprehensively consider whether there are potential factors.