P-value floating point precision PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Emma Knowles posted on Thursday, July 06, 2017 - 6:02 am
Hi all,

My apologies if this has already been raised but I couldn't find a solution to this problem in another thread. I am running a mediation analysis and the p-values for, for example, the indirect effect are printed in the output as 0.000. Is there a way to get a non-truncated p-value? As far as I can tell saving the results does not include the pvalues.

Thanks,

Emma
 Bengt O. Muthen posted on Thursday, July 06, 2017 - 6:56 pm
Q1: No. But 3 decimals should be enough. You know then that the p-value is less than 0.001.
 Emma Knowles posted on Friday, July 07, 2017 - 5:46 am
Hi Bengt,

Thanks so much for your response. I was curious about this because the manuscript in question contains many statistical tests and reviewers like to raise the issue of multiple testing correction. So in some cases knowing that the p-value is less than 0.001 is not sufficient. However, it is helpful to know that Mplus cannot give any more information than this.

Emma
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: