Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Statistical significance
Message posted by Jim Poindexter on March 21, 2000 at 12:00 AM (ET)
How can One test statistical significance of correlations when answering research questions without establishing hypotheses?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Statistical significance
Message posted by jammaster on March 22, 2000 at 12:00 AM (ET)
just ask your statistics-teacher,
he knows.
Re: Statistical significance
Message posted by Jack Tomsky on March 23, 2000 at 12:00 AM (ET)
If the null hypothesis is that the population correlation coefficient is zero, use a t test. If the null hypothesis is that the correlation is some specified nonzero value, then the transformation, (1/2)*ln[(1+r)/(1-r)], which is approximately normal, can be used for the test.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.