P value calculation

AngryBadger

Apologies if this question is particularly stupid. I'm trying to use a T-test to validate/support some conclusions I've drawn based on the experimental data I have. I can't figure out how to calculate the P-value once I've worked out the "t" statistic.

I'm obviously missing something simple, but it also seems like the relevance of the P-value is totally arbitrary?

Thanks in advance for any help.

GuanYin

AngryBadger wrote:

Apologies if this question is particularly stupid. I'm trying to use a T-test to validate/support some conclusions I've drawn based on the experimental data I have. I can't figure out how to calculate the P-value once I've worked out the "t" statistic.

I'm obviously missing something simple, but it also seems like the relevance of the P-value is totally arbitrary?

Thanks in advance for any help.

You don't calculate P values, you look them up on a statistics table, or use a calculator.

AngryBadger

psi wrote:

You don't calculate P values, you look them up on a statistics table, or use a calculator.

As I said something simple

But presumably the P-value is related to something then? As in what do I need to work it out using tables or a calculator?

GuanYin

AngryBadger wrote:

As I said something simple

But presumably the P-value is related to something then? As in what do I need to work it out using tables or a calculator?

Google is your friend.

http://www.statsoft.com/textbook/sttable.html#t

EmmB

Yep just calculate Z, the test statistic...there can be different formulas for this depending on what type of data you're using. Then look up the corresponding P-value on that table, the p-value is the area under the normal curve..p-value can be interpreted as measure of statistical significance, the smaller the p-value the greater the statistical significance.

AngryBadger

Sirs I am in your debt. Much obliged.