Question about discussion of T tests in Chapter 6

[kburchfiel]

Hi there,

I have found this book to be a very helpful reference in conducting various statistical tests with survey data. I did have a question about one statement in Chapter 6, though. You write within the Comparison of Proportions and Means section that:

"We use t-tests to compare two proportions or means. T-tests allow us to determine if one proportion or mean is statistically different from another. They are commonly used to determine if a single estimate differs from a known value (e.g., 0 or 50%) or to compare two group means (e.g., North versus South)."

However, I have heard that a chi squared test might be a better fit than a t test for comparing two proportions. Would you say that this is the case, or am I misunderstanding something?

For example, let's say that we wanted to conduct post-hoc tests following the calculation shown in 'Example 3: Test of homogeneity' within this chapter. We see from the chi squared test result that different age groups appeared to vary in their level of support for Biden, Trump, and 3rd-party candidates. However, it would then be helpful to conduct post-hoc analyses in order to determine whether a given variation was statistically significant. (For instance, did the proportion of 40- to 49-year-old voters who chose Biden differ significantly from the proportion who chose Trump?) I think a pairwise chi squared test could be a good fit here, but would a T test be a valid option here as well?