Re: P - Values
I agree with you Piersante, that p-values and CIs are closely related as they are calculated using the same information. However the way this information is presented in a CI makes it much easier to consider clinical significance than using a p-value.
Regards Rod Jackson
On 12/01/11 9:19 AM, "Piersante Sestini" <[log in to unmask]">[log in to unmask]> wrote:
On 10/01/2011 22.03, Rod Jackson wrote:
Re: P - Values Hi – The interpretation of whether a finding is likely to clinically significant or not, is one of key arguments for using CIs rather than p-values.
I don't think that CIs tell us more than P about clinical significance (in fact, they are closely correlated). What tell us about clinical effect are the outcome sizes (with or without intervention). P and CIs only tell us about how much we can trust the estimated effect size,
cheers
Piersante Sestini