0
$\begingroup$

Following is the experimental design - There are two treatment group and one control. Need to conclude if treatment 1 is better than treatment 2. Is there a statistical test that helps in answering this question

$\endgroup$
0
$\begingroup$

Yes. The test depends on what metric you want to measure. If it is website conversions you'd use a two sample test for proportions. See this for web experiments

$\endgroup$
  • $\begingroup$ Here we need to do multiple comparison. T1 Vs C and T2 Vs C for this i can do individual t-test. how do I compare the difference between (T1Vs C) and (T2 Vs c) . Thanks $\endgroup$ – Naveenan Mar 4 '16 at 21:24
  • $\begingroup$ If it's a randomized trial you can directly compare t1 with t2. Why do you want to compare both with c and then take the difference? $\endgroup$ – wabbit Mar 7 '16 at 14:28
  • $\begingroup$ Yes experiment is randomized. This information is helpful. I can compare T1 directly with T2. One question that is bothering me about - for this experiment design whether we need to use bonferroni correction. $\endgroup$ – Naveenan Mar 8 '16 at 15:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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