 # Book 6 page 14 Quest # 32

Can someone clarify this for me? I thought the Z stat was calculated using: x-u/std. dev am i incorrect??? correct ans is C

can you post the whole thing please?

Which of the following statements about hypothesis testing involving the z stat is FALSE? A calculated z stat for testing a sample mean is z = (x-u)/std dev/ sq rt n B p value is the probability of getting a test stat (z) at least as far from the hypothesized value as the one calculated C a z test is sometimes used in place of a t test for tests concerning a mean when sample size is small D if the confidence level is set a 95%, the chance of rejecting the null hypothesis when in fact it is true is 5%

C is the false statement above.

rjs157 Wrote: ------------------------------------------------------- > Can someone clarify this for me? > > I thought the Z stat was calculated using: > > x-u/std. dev > > am i incorrect??? > > correct ans is C After reading your question:: Option A was asking for ‘sample mean’ so we need to use the z-value = (X - mu)/ [STD/SQRT(n)] to take into account for the sampling error while estimating the confidence intervals on populations.

I realize that C is the best answer now because you always use a t stat for small samples… but I always like to rule out all the other answers so, what’s the equation for z STAT? maybe i’m thinking of a z “score” which is X-u/std. dev…

“Option A was asking for ‘sample mean’ so we need to use the z-value = (X - mu)/ [STD/SQRT(n)] to take into account for the sampling error while estimating the confidence intervals on populations.”

yup. since you’re testing a sample MEAN, you must scale the variance by the sample size ie. use standard error as opposed to standard deviation C is definetly the answer anyway… just think, t-stats provide a wider band than z-stats

thanks guys