2
$\begingroup$

I was recently explaining the Welsh–Satterthwaite equation's application to two sample t-tests of differing variances, but while going through an example using Excel's built in t-test function (type = 3 for unequal variance) I for the life of me couldn't get my calculation to match up exactly with Excel's built-in t-test function. It's pretty close and I wouldn't want anyone to lose any sleep over this but I've looked around and haven't found any info on the actual code of this function.

You can find the sample workbook here, hope someone out there has a quick answer to this!

Improve this question
$\endgroup$
3
  • 1
    $\begingroup$ The difference could be in exactly how the degrees of freedom is calculated as it is fractional. $\endgroup$
    – Michael R. Chernick
    Commented Jul 20, 2017 at 5:19
  • 1
    $\begingroup$ I can't find anywhere that Excel actually says it uses Welch-Satterthwaite for the unequal variance case $\endgroup$
    – Glen_b
    Commented Jul 20, 2017 at 7:27
  • $\begingroup$ Thanks for the help here guys, I guess I just can't think of what other df calculation Excel would be using especially considering I'm so close in using the Welch-Satterthwaite df approach. $\endgroup$
    – Ben Saunders
    Commented Jul 21, 2017 at 23:18

1 Answer 1

Reset to default
1
$\begingroup$

P-val (raw) is calculated using a df of 11. That's right, the function rounds DOWN. P-val (round) is the same value as the data analysis toolpak, it also rounds UP to 12 as you did. The function T.TEST uses the decimal value of 11.7, so it gives the most correct value.

I used this website to calculate the p values given any given df value and t stat.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.