UCB Admissions Issue

In 1973 at UCB, there was what appeared to be an admissions problem. This problem was that there was a significantly higher percentage of males than females admitted to the graduate school (Grigg, 2018). This problem occurred because people were only looking at the percentage of each gender admitted and not the other underlying factors. This idea is represented in the graph below. This graph makes it clear that there is a large difference between males and females admitted to the graduate school.

The Actual Data

When statisticians took a deeper look at the data, they found that the admissions rate favored females. The reasoning behind the initial misinterpretation of data is that females were applying for departments that had lower acceptance rates while the men were applying to ones with higher acceptance rates. Due to this variable not being mentioned in the original data, it created the appearance that there was gender bias at UCB (Grigg, 2018). The graph below shows data for six of the departments that students can apply for. This graph shows that in four of the six departments a larger percentage of females were admitted than males while in the other two departments there is very little difference.

Simpson’s Paradox

Simpson’s paradox occurs in statistics and can often create misinterpretations of data. One way it is defined is, “A trend or result that is present when data is put into groups that reverses or disappears when the data is combined” (Grigg, 2018). Simpson’s paradox occurs due to lurking variables. Lurking variables are also known as hidden variables and can be defined as, “A variable that is not included in a statistical analysis yet impacts the relationship between two variables within the analysis” (Bobbit, 2019). The UCB admissions situation is an example of Simpon’s Paradox because it has a lurking variable. This variable is the departments that applicants applied for. Since this lurking variable exists, it creates a sense that UCB had gender bias in their admissions process when that was not the case. By having someone truly examine the data, they were able to determine there was never a gender bias and there were no issues with the UCB admissions process.

Works cited

Grigg, T. (2018, December 9). Simpson’s paradox and Interpreting Data. Medium. https://towardsdatascience.com/simpsons-paradox-and-interpreting-data-6a04435167

Bobbit, Z. (2019, May 3). Lurking variables: Definition & examples. Statology. https://www.statology.org/lurking-variables/