libaiyi
★    

China,
2018-07-12 05:20

Posting: # 19037
Views: 1,790
 

 Confidence Interval and p value are Con­tradictory [General Sta­tis­tics]

Hi,all

The output of a BE study shows that the confidence interval of all three PK parameters are between 80%-125%. But the p value of AUC0-t's treatment effect is bigger than 0.05, which is in contrast to the conclusion from CI, and the period effect and carryover effect are all proved equal. So, what would be the reason of this contradictory conclusion? All the output are from SAS output.

Thanks in advance.;-)
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-07-12 11:09

@ libaiyi
Posting: # 19038
Views: 1,574
 

 100% included in 90% CI?

Hi libaiyi,

» The output of a BE study shows that the confidence interval of all three PK parameters are between 80%-125%. But the p value of AUC0-t's treatment effect is bigger than 0.05, which is in contrast to the conclusion from CI, …

I don’t see a contradiction. If the 90% CI contains 100%, the p value will always be >0.05. Only if not, it may be <0.05 (see this lengthy thread).

» … and the period effect and carryover effect are all proved equal.

You can’t prove anything with statistics. Do you mean that the p values are not significant?

Cheers,
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. ☼
Science Quotes
libaiyi
★    

China,
2018-07-13 08:48

@ Helmut
Posting: # 19042
Views: 1,483
 

 100% included in 90% CI?

Hi,Helmut

» I don’t see a contradiction. If the 90% CI contains 100%, the p value will always be >0.05. Only if not, it may be <0.05 (see this lengthy thread).

Sorry I wrote it wrong, it supposed to be that the effect is smaller than 0.05, but I understand now about the meaning of p value.

» You can’t prove anything with statistics. Do you mean that the p values are not significant?

Yes I did mean that the p values are not significant.
Thanks a lot!:party:
zizou
★    

Plzeň, Czech Republic,
2018-07-17 23:51

@ Helmut
Posting: # 19078
Views: 1,412
 

 100% included in 90% CI?

Hi.
» I don’t see a contradiction. If the 90% CI contains 100%, the p value will always be >0.05. Only if not, it may be <0.05 (see this lengthy thread).

Correct & (as mentioned in the mentioned thread) if the 90% CI does not contain 100%, the p value will always be <0.1.

Best regards,
zizou
Activity
 Thread view
Bioequivalence and Bioavailability Forum |  Admin contact
19,694 posts in 4,181 threads, 1,355 registered users;
online 5 (0 registered, 5 guests [including 5 identified bots]).
Forum time (Europe/Vienna): 13:48 CEST

A central lesson of science is that to understand complex issues
(or even simple ones), we must try to free our minds of dogma and
to guarantee the freedom to publish, to contradict, and to experiment.
Arguments from authority are unacceptable.    Carl Sagan

The BIOEQUIVALENCE / BIOAVAILABILITY FORUM is hosted by
BEBAC Ing. Helmut Schütz
HTML5