At the corner market, you can buy rice by
1 answer:
You might be interested in
I think like that but I’m not sure
These questions should be asked more clearly.
If 3% is the standard deviation of the average then a 1% decrease in score is a z=-1/3, which is small in absolute value, so we can't reject the claim. In this interpretation we'd answer: -0.33, no.
But if we'll assume 3% is the standard deviation of the individual samples, it's a different story. We use percent as our unit. The standard deviation of the average is
.
So we get a z (really a t) here of
That's kinda borderline, a one sided t test with 29 dfs will give about the same probability as the normal distribution for a tail z=1.8, p=.036 aka p=3.6%. Typically we'd choose a 1% or 5% rejection threshold before we started; this one is in between.
