Most of us believe we "hit" statistical significan
Post# of 154721
(Total Views: 845)
Posted On: 02/24/2021 9:18:11 AM
Most of us believe we "hit" statistical significance in the PE .
In average, we believe that SOC had a 34.2 % of deaths (44.
and VX had 16.4% (42.9) with a death improvement of 52.2%.
This would produce a p-value of 0.0000809 (yes, 4 zeroes). Not delving too much into this. BorelFields already posted a very complete analysis of this data.
OK, imho we are being a bit optimistic. But, please take any number you like for SOC deaths. Say 30%, please do remember that midway our trial we had a 23.1% overall death rate (45/195) at a time that most trials were reporting SOC deaths exceeding 30%)
With this (SOC=30%) we have a 2-tailed p-value of 0.01377 (38.5% death reduction)
So, do we have GRANTED that we meet statistical significance ??? No. There are combinations of events that can result in us not getting there. For example, in the second part of out trial the SOC survival improved dramatically AND VX decreased idem (please do remember that the overall death rate at the end of the trial is very similar to that of the first part).
Or, in the second part of trial, the SOC got mostly severe and we got mostly critical ... and opposite in the first cohort.
Or, VX worked in the first part of the trial and by some "act of the devil" did not work as effectively in the second part ...., while the good doctors were recovering SOC patients at a much better rate.
But ... but ... you get my point ???. Entienden ???
And yes, we longs are a very optimistic bunch ... but we are not delusional.
Only COD afflicts us
In average, we believe that SOC had a 34.2 % of deaths (44.
and VX had 16.4% (42.9) with a death improvement of 52.2%. This would produce a p-value of 0.0000809 (yes, 4 zeroes). Not delving too much into this. BorelFields already posted a very complete analysis of this data.
OK, imho we are being a bit optimistic. But, please take any number you like for SOC deaths. Say 30%, please do remember that midway our trial we had a 23.1% overall death rate (45/195) at a time that most trials were reporting SOC deaths exceeding 30%)
With this (SOC=30%) we have a 2-tailed p-value of 0.01377 (38.5% death reduction)
So, do we have GRANTED that we meet statistical significance ??? No. There are combinations of events that can result in us not getting there. For example, in the second part of out trial the SOC survival improved dramatically AND VX decreased idem (please do remember that the overall death rate at the end of the trial is very similar to that of the first part).
Or, in the second part of trial, the SOC got mostly severe and we got mostly critical ... and opposite in the first cohort.
Or, VX worked in the first part of the trial and by some "act of the devil" did not work as effectively in the second part ...., while the good doctors were recovering SOC patients at a much better rate.
But ... but ... you get my point ???. Entienden ???
And yes, we longs are a very optimistic bunch ... but we are not delusional.
Only COD afflicts us
(27)
(0)