The Student Room Group

Statistics - what does x bar - ns mean?

Hi! I'm doing some past papers and there was this question that confused me. So I left it and went back to it after I finished the rest of the paper. I looked at the mark scheme later, and it said that you have to use xbar-ns, but I haven't come across this formula before? I'd really appreciate if someone could explain what it's used for and what it means! Thank you in advance :biggrin:

Edit: I know that xbar is the mean and s is the sample standard deviation. I just don't know when it should be used and what for
(edited 7 years ago)
Original post by JustARandomer123
Hi! I'm doing some past papers and there was this question that confused me. So I left it and went back to it after I finished the rest of the paper. I looked at the mark scheme later, and it said that you have to use xbar-ns, but I haven't come across this formula before? I'd really appreciate if someone could explain what it's used for and what it means! Thank you in advance :biggrin:

Edit: I know that xbar is the mean and s is the sample standard deviation. I just don't know when it should be used and what for


A link to the paper/question would be useful :h:
for the question, n was used as 2 and I'm not sure how the 2 came about either. So the answer came out as 45.8-2(24), where 45.8 is the mean and 24 is the standard deviation


Original post by JustARandomer123
for the question, n was used as 2 and I'm not sure how the 2 came about either. So the answer came out as 45.8-2(24), where 45.8 is the mean and 24 is the standard deviation


Ah, that is something of a nasty question.

You may not have heard explicitly of the 68-7-95-99.7 rule but you will have seen it in action. It just says that 68% of data is within 1 standard deviation of the mean, 95% is within 2 standard deviations and 99.7% is within 3 (which is why they've used 'n=2' and 'n=3' when it seems they've plucked it out of thin air). But anyway, they use that to find a range in which 95% of the data probably lies and since a lot of values are negative it's not good enough to model a distribution where all of the values are positive. And in this case, you're using s, the estimate for standard deviation rather than standard deviation itself.

https://en.wikipedia.org/wiki/68%E2%80%9395%E2%80%9399.7_rule
(edited 7 years ago)
Original post by SeanFM
Ah, that is something of a nasty question.

You may not have heard explicitly of the 68-7-95-99.7 rule but you will have seen it in action. It just says that 68% of data is within 1 standard deviation of the mean, 95% is within 2 standard deviations and 99.7% is within 3 (which is why they've used 'n=2' and 'n=3' when it seems they've plucked it out of thin air). But anyway, they use that to find a range in which 95% of the data probably lies and since a lot of values are negative it's not good enough to model a distribution where all of the values are positive. And in this case, you're using s, the estimate for standard deviation rather than standard deviation itself.

https://en.wikipedia.org/wiki/68%E2%80%9395%E2%80%9399.7_rule


oh, that makes more sense :P thank you!

Quick Reply

Latest