Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 0 down vote favorite

If you have performed a statistical test using the standard 95% confidence threshold then a p-value of < 0.05 will indicate significance.

I was reading a blog post about typesetting some correlation data using the R library xtable. The author uses a series of ifelse statements to replace p-values with stars to indicate whether the result was less than 0.001, 0.01, or 0.05. I was wondering if anyone could comment on the validity of this?

If you perform a statistical test at a 95% confidence level and you get a p-value of < 0.01 does that mean that you could say that the result is significant to the 99% level. Or can you only assume a stronger confidence level if you explicitly performed the test with that criteria?

share|improve this question

1 Answer

up vote 3 down vote accepted

Calculation of a p-value does not require any CI to be mentioned. If you have the z-score, you can calculate the p-value by integration over the normal distribution from -inf to the z-score. Or, you could look it up in z-tables.

That's why if you perform a statistical test at a 95% confidence level and you get a p-value of < 0.01 that does mean you could say that the result is significant to the 99% level too.

share|improve this answer
Of course! I was forgetting when you perform a test you get back the z-score and then compare it with the value in table. It's been a long time since I've actually done that manually. – Alastair Nov 27 '12 at 17:30
It's actually tough to do manually if we get statistic that follow other distributions. Calculating from z-score is easy though. – Blain Waan Nov 27 '12 at 17:36

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.