What's wrong with this formula?

[asb3stos]

Hi everyone, i'm making a formula for Spearsmans Rank Correlation and
I've made a realtively simple formula for a part of it but it's not
working and I can't spot what's wrong with it.

=1-(6*G98)/(ROWS(B83:B102))*(POWER((ROWS(B83:B102),2))-1)

Can anyone help me, i'm dying here! Cheers!
(the references are all correct, maybe it's the brackets?)

 

[Mallycat]

What are you trying to do with ROWS()? This will count the number of
rows in a range. Are you meaning to use SUM()? or something else?

Matt

 

[CLR]

Maybe.........

=1-(6*G98)/(ROWS(B83:B102))*(POWER(ROWS(B83:B102),2))-1

Vaya con Dios,
Chuck, CABGx3

 

[David Biddulph]

Hi everyone, i'm making a formula for Spearsmans Rank Correlation and
I've made a realtively simple formula for a part of it but it's not
working and I can't spot what's wrong with it.

=1-(6*G98)/(ROWS(B83:B102))*(POWER((ROWS(B83:B102),2))-1)

Can anyone help me, i'm dying here! Cheers!
(the references are all correct, maybe it's the brackets?)

You want (n^3-n) under the divide, so you need to put brackets around the
whole of that divisor (n*(n^2-1))
=1-(6*G98)/(ROWS(B83:B102)*(POWER(ROWS(B83:B102),2)-1))

In your formula I think you're multiplying by n^2-1 where you intended to
divide.

 

[asb3stos]

Thanks for your help guys, i've got it working now thanks to you, i'll
post the full formula here in a few minutes.

 

[asb3stos]

Check this calcualtor i've made for Spearman's Rank Correlation, it's a
bit shabby sure, but it's a simple as i can get it. This just gives the
correlation between two sets of data.
http://www.filelodge.com/files/room30/831080/Spearmans Rank Correlation Calculator.xls