Isaac Dialsingh <[log in to unmask]> commented:
It is a common outcome for reviewers of journals to request p-values up
to the first non zero digit. For example, 0.000 might not be accepted ...
David W Smith replied:
Statistical programs usually provide either 3 or 4 digits. ...
The statistician should reply to such reviewers with information not on how many digits were computed but how many are believable. Even "exact" tests are based on assumptions of randomness, exchangeability and whether the data are representative of a population. How much could the value be altered by changing one data point (sensitivity)? David's direct answer (with which I agree) to the question taken out of context raises the danger of papers continuing to crescendo on the p, omitting the "so what does this mean on the ground?".
Allan Reese
***********************************************************************************
This email and any attachments are intended for the named recipient only. Its unauthorised use, distribution, disclosure, storage or copying is not permitted. If you have received it in error, please destroy all copies and notify the sender. In messages of a non-business nature, the views and opinions expressed are the author's own and do not necessarily reflect those of the organisation from which it is sent. All emails may be subject to monitoring.
***********************************************************************************
|