This question was closed without grading. Reason: Other
Aug 11, 2012 07:15
11 yrs ago
2 viewers *
English term
spend, spent
English
Medical
Mathematics & Statistics
clinical trials
All statistical type I error will be spent on the hypothesis tests performed on the data collected in the treatment period
[what does it mean in this context?]
[what does it mean in this context?]
Responses
5 | eliminated | B D Finch |
3 +1 | use a spending function | DLyons |
3 -1 | passed on | ErichEko ⟹⭐ |
References
might help | Cristina Crişan |
Responses
-1
7 mins
English term (edited):
spent
passed on
The errors of Type I will be passed on to the test, i.e. made as input for tjhe said test.
1 hr
eliminated
It means that Type 1 errors (i.e. those leading to rejection of a true null hypothesis) will be eliminated by means of the tests in question.
Peer comment(s):
neutral |
DLyons
: "Reduce" perhaps, but "eliminate" is too strong.
1 hr
|
Rationally, I agree with you. However, I think that "eliminated" is what the writer means.
|
+1
3 hrs
use a spending function
Probably arising in the context of deciding whether to stop the trial early.
Reference:
http://books.google.ie/books?id=KW9esgo759EC&pg=PA217&lpg=PA217&dq
http://www.niaid.nih.gov/about/organization/dcr/BRB/Documents/bvaluejcgs.pdf
Peer comment(s):
neutral |
B D Finch
: That does look possible.
3 days 3 hrs
|
Thanks Barbara. It's not entirely clear from the extract but that's my best guess. I notice it's posted in Polish also and all three of us independently came up with the same suggestion.
|
|
agree |
Doroteja
20 days
|
Thanks Doroteja.
|
Reference comments
5 hrs
Reference:
might help
There are two ways of designing an O'Brien-Fleming design, a popular design for conducting interim analyses of clinical trials. The first method is to use an error (or alpha) spending function, which essentially gives you a "budget" of error you can spend at each interim analysis.
http://realizationsinbiostatistics.blogspot.ro/2007/09/my-ob...
Lan and DeMets developed a very flexible
boundary procedure that accommodates unequal
timing, additional looks, even extending the trial
– Called “alpha spending” functions
– Researchers can choose to “spend” their alpha (i.e.,
conduct interim analyses) any way they want
– Ensures that total alpha “spent” is no more than 0.05
(or whatever was specified)
http://www.icssc.org/Documents/AdvBiosGoa/Tab 25.00_InterimA...
http://realizationsinbiostatistics.blogspot.ro/2007/09/my-ob...
Lan and DeMets developed a very flexible
boundary procedure that accommodates unequal
timing, additional looks, even extending the trial
– Called “alpha spending” functions
– Researchers can choose to “spend” their alpha (i.e.,
conduct interim analyses) any way they want
– Ensures that total alpha “spent” is no more than 0.05
(or whatever was specified)
http://www.icssc.org/Documents/AdvBiosGoa/Tab 25.00_InterimA...
Something went wrong...