Question 977304
<font face="Times New Roman" size="+2">


Throw out the Z-score and the raw score since both of those are below the mean, whereas the 57th percentile is at least a little bit above the mean.  But your T-statistic of 70 doesn't mean much without some knowledge of the size of the sample from which the statistic was derived.


Be that as it may, I was able to bound the problem thus:  The minimum sample size must be 4 because you gave four, albeit differently stated, test results.  The maximum sample size I calculated to be 25, because if the sample size is larger than 25 the raw score is larger than 100 which I assume to be the upper limit on the raw score.


Either way, the minimum raw score for a t-statistic of 70 is 58 which is way larger than the 57th percentile.


Having said all that, this question is magnificently stupid and made so by the inclusion of a t-statistic and z-score comparison.  Z-scores are for when you know *[tex \Large \sigma] and the t-distribution is for when you don't know *[tex \Large \sigma].  Since the use of these two sample statistics is mutually exclusive, why would anyone ask you to do something so dumb as to compare the relative magnitude of the two for a given set of data?


John
*[tex \LARGE e^{i\pi}\ +\ 1\ =\ 0]
My calculator said it, I believe it, that settles it

*[tex \Large \ \
*[tex \LARGE \ \ \ \ \ \ \ \ \ \