strange p value

29 views
Skip to first unread message

Aron Arnold

unread,
Jan 21, 2018, 7:36:38 PM1/21/18
to sofastatistics
Hello,
I wondered if someone could help me.
When I run a Mann-Whitney U test, I get this strange p value: "7.655e-3"
When I add or delete items from my Excel spreadsheet and run the test in exactly the same way, I get a correct value. But when I run it with all the data that I have to analyse, I get this 7.655e-3.
Could someone tell me what I'm doing wrong?
Thank you in advance :)
All the best,
Aron

J David Eisenberg

unread,
Jan 21, 2018, 11:08:47 PM1/21/18
to sofasta...@googlegroups.com
7.655e-3 is the way SOFA (and most programming languages) does
scientific notation; it’s the same as 7.665 times ten to the -3. As
you can see, I can't type a superscript in this email; 7.665 x 10 ^ -3
is as close as I could get. That is why SOFA and programming languages
use the "e" notation (think of the "e" as standing for "tens
exponent") -- to get around the problem of not having superscripts on
old-style computer terminals and printers.

So, 7.665e-3, or 7.665 times ten to the -3, is 0.007665.
> --
>
> ---
> You received this message because you are subscribed to the Google Groups
> "sofastatistics" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sofastatistic...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

J David Eisenberg

unread,
Jan 21, 2018, 11:09:30 PM1/21/18
to sofasta...@googlegroups.com
Sorry about the mistype of 7.655 as 7.665 in a couple of places. I
need new glasses :)

Aron Arnold

unread,
Jan 22, 2018, 5:58:22 AM1/22/18
to sofastatistics
Thank you so much for your answer!!!
As I'm not that familiar with statistics software I didn't know this was a conventional way of encoding exponents.
So that's great new - my p value is definitely smaller than 0,05 :)
Reply all
Reply to author
Forward
0 new messages