I'm looking to know what difference I am able to detect with the number of participants I have now and the number of participants I would have in 4 months if recruitment stays at the same rate. The difference I'm interested in is survival time between three groups. I'm finding it hard to make sense of what I've found so far. Thanks!
I hope I understood your question right.
Assume you have 2 survival time datasets: the current one and the later (4mos) one. You can conduct two tests:
(1) t-test to test the null hypothesis of nonchanging
means i.e.; the mean survival time now is the same as after 4 mos
(2) Fisher’s F test to test the null hypothesis of non-changing variances between the 2 samples
Thanks for answering so quickly Faye. I'm not sure I've explained it that well. I have 3 groups. I want to compare survival times between all three. I'm only following people for 6 months. I want to see what difference I can detect now and what difference I'd be able to detect if I recruit for another 4 months. Is that making sense?
Hi Charlotte,
Say these groups are A, B & C. At any point of time, you could perform the 2 tests (t-test & F-test) as follows:
a) A vs B
b) A vs C
c) B vs C
If for any of the above three comparisons, the null hypothesis (equality) is rejected then the one with lower mean is superior to the other.
Hope this helps.
- Badges
- Science topic