... years of experience is correlated with the response, there’s a possibility that the level of experience confounds the interpretation of the two-sample t-interval. Perhaps men in this sample have more experience than women, and it’s this difference rather than gender that explains the difference in average salary. A variable such as experience is a confounding variable if it is correlated with the response and the two groups differ with respect to this variable. Figure 25.2 shows that the number of years of experience is correlated with salary, so it meets the first requirement of a confounding variable. To meet the second requirement, male managers ...

Get Statistics for Business: Decision Making and Analysis, 3rd Edition now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.