Hey jamie9.
Can you show us what you have tried?
15 test subjects are asked if they prefer brand A, brand B, or have no preference. 9 prefer brand A, 3 prefer brand B and 3 have no preference. What is the P-value for a sign test of no difference between brands?
Thanks in advance for any help guys, I'm really stuck
Yes sorry if this doesn't make sense...
Okay so I was thinking for no difference H0: µ = 7.5 and H1: µ ≠ 7.5.
Then I wanted to use a z test and test for ≥ 7.5 where B(15, 0.5) but the standard deviation wasn't given so I assumed I was on the wrong track?
When you say sign test, I thought you wanted to do a non-parametric test for this data.
Given this data you have three choices and it might make sense if you allocated brand A to -1, no preference to 0, and brand B to +1 and then used a test that checked whether the median of this sample was statistically equal to 0 (given some confidence level) or whether you could reject this claim (or hypothesis) with the data.