Welch's t-test and Satterthwaite's DF

[DrLynTaylor]

CAMIS/R/ttest_2Sample.qmd

Line 81 in 5e19d3d

The following code from the **procs** package was used to perform a two sample t-test. Note that the `proc_ttest()` function performs both the Student's t-test and Welch's (Satterthwaite) t-test in the same call. The results are displayed on separate rows. This output is similar to SAS.

Could we tidy up the language here, as welch t-test is not the same as Satterthwaite's calculation of the degrees of freedom... This site may help us to explain better https://www.statology.org/satterthwaite-approximation/

NOTE: there are pooled method, and satterthwaite method.. Please check, but I think if variances for the 2 treatments are 'similar' assumed the same in the population then OK to pool, but if not similar / not same in population, then should use Satterthwaite.. Need to add this detail. Check all SAS/R/ Python / Comp to make this clearer &

Add another example.

• clarify welch t-test vs pooled / satterthwaite in R page
• clarify welch t-test vs pooled / satterthwaite in SAS page
• clarify welch t-test vs pooled / satterthwaite in Python page
• clarify welch t-test vs pooled / satterthwaite in Comp page
• Add example where variances are different in the 2 treatment groups showing difference betweeen Pooled & Satterhwaite methods.. ideally do in R, SAS, python, COMP pages