Race and the NFL: Teaching the Chi-Square Goodness of Fit Test

Abstract

The lesson focuses on the chi-square goodness of fit test, a statistical test that examines whether observed frequencies deviate from what might be expected based on a proposed hypothesis. This statistical test is used in all manner of research areas. For example, researchers might examine whether consumers deviate from expectations when they select particular products. Researchers could similarly examine whether the presence of personal attributes in a subpopulation (yes or no) differs from the observed frequencies in the broader population. I typically teach this lesson after students have learned levels of measurement, descriptive statistics, fundamentals of inferential statistics (e.g., the nature of p-values), and other inferential statistics tests (e.g., z tests, t tests, and so forth).This lesson has an additional purpose. I want to show students that they can use statistics to gain insight into contemporary social issues. To that end, I decided to have my undergraduate students compare the observed frequencies of National Football League (NFL) head coaches to expected frequencies based on race. Conceptions of race are socially constructed, but that does not negate their impact. 1 The context of race and NFL head coaches provides a statistically driven method for considering the nature of prejudice in contemporary United States society. This type of examination is especially timely given the recent prominence of the Black Lives Matter movement and the political climate around inclusion more generally (pewrsr.ch/3kEIKzH).