sdm-help-list
89 Subscribers 
sdm-help-list > Appropriate threshold in the absence of precise p-values
Showing 1-2 of 2 posts
Mar 22, 2018 02:03 PM | hennavartiainen
Appropriate threshold in the absence of precise p-values
Hi!
I'm currently working on a meta-analysis using this excellent tool, and was wondering if it would be acceptable to use "n" as the t-value for the studies that do not report the precise p-value (i.e. no Z- or T-values either, merely p< .01 or similar).
Thank you for your help in advance!
I'm currently working on a meta-analysis using this excellent tool, and was wondering if it would be acceptable to use "n" as the t-value for the studies that do not report the precise p-value (i.e. no Z- or T-values either, merely p< .01 or similar).
Thank you for your help in advance!
Mar 23, 2018 02:03 PM | Jo Cutler
Appropriate threshold in theabsence of precise p-values
Hi
I'm not one of the software developers but think I'm right in saying that you can use the calculator given on the SDM website (https://www.sdmproject.com/utilities/?sh...) to covert a threshold p-value (e.g. p<0.01) to a minimum t-value (i.e. the t for which p = 0.01) giving a conservative estimate of the effect.
Jo
I'm not one of the software developers but think I'm right in saying that you can use the calculator given on the SDM website (https://www.sdmproject.com/utilities/?sh...) to covert a threshold p-value (e.g. p<0.01) to a minimum t-value (i.e. the t for which p = 0.01) giving a conservative estimate of the effect.
Jo