Closed ShoumoBhattacharyaLab closed 11 months ago
Hi Shoumo Thanks for using DescTools. The Dunnett test as it is implemented in DescTools only returns the two-sided p-values. I simply didn’t think of one-sided tests, when coding it. If you need one-sided tests you can make use of the package PMCMRplus, which also has a function dunnettTest() allowing to pass the known ‘alternative’ argument.
In future I might also include this in DescTools::DunnettTest()… 😉
Best, Andri
Hello Andri
I have a request - Dunnett test is parametric, is it possible to incorporate a similar but non-parametric test (one control - many comparisons, multiple testing correction) for data that is not normally distributed? Or is such a test already available in DescTools?
Best wishes
Shoumo
Shoumo Bhattacharya MD MSc (Biochem) FRCP FESC FMedSci | Professor of Cardiovascular Medicine & Fellow of Green Templeton College | Radcliffe Department of Medicine & Wellcome Centre for Human Genetics | University of Oxford @.*** | www.well.ox.ac.uk/bhattacharya
On 9 Aug 2023, at 18:00, Andri Signorell @.***> wrote:
Hi Shoumo Thanks for using DescTools. The Dunnett test as it is implemented in DescTools only returns the two-sided p-values. I simply didn’t think of one-sided tests, when coding it. If you need one-sided tests you can make use of the package PMCMRplus, which also has a function dunnettTest() allowing to pass the known ‘alternative’ argument.
In future I might also include this in DescTools::DunnettTest()… 😉
Best, Andri
— Reply to this email directly, view it on GitHubhttps://github.com/AndriSignorell/DescTools/issues/125#issuecomment-1671815797, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BBW5ZOV2T4VMAE5V2AV52SDXUO637ANCNFSM6AAAAAA3FYNAII. You are receiving this because you authored the thread.Message ID: @.***>
Hi Shoumo
I have no knowledge of such a test, sorry.
Best, Andri
Von: ShoumoBhattacharyaLab @.> Gesendet: Freitag, 11. August 2023 09:13 An: AndriSignorell/DescTools @.> Cc: Andri Signorell @.>; State change @.> Betreff: Re: [AndriSignorell/DescTools] DunnettTest (Issue #125)
Hello Andri
I have a request - Dunnett test is parametric, is it possible to incorporate a similar but non-parametric test (one control - many comparisons, multiple testing correction) for data that is not normally distributed? Or is such a test already available in DescTools?
Best wishes
Shoumo
Shoumo Bhattacharya MD MSc (Biochem) FRCP FESC FMedSci | Professor of Cardiovascular Medicine & Fellow of Green Templeton College | Radcliffe Department of Medicine & Wellcome Centre for Human Genetics | University of Oxford @. <mailto:@.> | www.well.ox.ac.uk/bhattacharya http://www.well.ox.ac.uk/bhattacharya
On 9 Aug 2023, at 18:00, Andri Signorell @. <mailto:@.> > wrote:
Hi Shoumo Thanks for using DescTools. The Dunnett test as it is implemented in DescTools only returns the two-sided p-values. I simply didn’t think of one-sided tests, when coding it. If you need one-sided tests you can make use of the package PMCMRplus, which also has a function dunnettTest() allowing to pass the known ‘alternative’ argument.
In future I might also include this in DescTools::DunnettTest()… 😉
Best, Andri
— Reply to this email directly, view it on GitHubhttps://github.com/AndriSignorell/DescTools/issues/125#issuecomment-1671815797, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BBW5ZOV2T4VMAE5V2AV52SDXUO637ANCNFSM6AAAAAA3FYNAII. You are receiving this because you authored the thread.Message ID: @. <mailto:@.> >
— Reply to this email directly, view it on GitHub https://github.com/AndriSignorell/DescTools/issues/125#issuecomment-1674302699 , or unsubscribe https://github.com/notifications/unsubscribe-auth/ABSGDZGNP5UULJNIQC2EJMTXUXLPBANCNFSM6AAAAAA3FYNAII . You are receiving this because you modified the open/close state. https://github.com/notifications/beacon/ABSGDZH63JH67BPRLD5X55LXUXLPBA5CNFSM6AAAAAA3FYNAIKWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTTDZPKOW.gif Message ID: @. @.> >
Please let me know if the DunnettTest implementation is one-tailed or two-tailed - I cannot find this in the documentation. Many thanks Shoumo