calculate the p-value for a negative t-value

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Camper - 17 Apr 2006 23:24 GMT

My question is how can I calculate a p-value for a negative t-value?

In several tests the t-test value is negative and I cannot use the standard
TDIST(x,df,1) function. Is it correct to calculate the p-value of negative
t-values as:
1 - absolute(P[x])?

If the above is correct, is this the correct adaption to the TDIST function:
=IF(x>0,TDIST(x,deg_freedom,tails),(1-(TDIST(ABS(x),deg_freedom,tails))))

Signature

Mahalo for your help,
Camper

Reply to this Message

Mike Middleton - 18 Apr 2006 07:57 GMT

Camper -

Here are some excerpts from my book, "Data Analysis Using Microsoft Excel:
Updated for Office XP":

Left-Tail Test
p-value =IF(t<0,TDIST(ABS(t),df,1),1-TDIST(t,df,1))

Right-Tail Test
p-value =IF(t>0,TDIST(t,df,1),1-TDIST(ABS(t),df,1))

Two-Tail Test
p-value =TDIST(ABS(t),df,2)

- Mike
www.mikemiddleton.com

> My question is how can I calculate a p-value for a negative t-value?
>
[quoted text clipped - 7 lines]
> function:
> =IF(x>0,TDIST(x,deg_freedom,tails),(1-(TDIST(ABS(x),deg_freedom,tails))))

Reply to this Message

Camper - 18 Apr 2006 19:26 GMT

Aloha Mike,

That answers my question and is exactly what I needed. Mahalo for the
reference to the book. I will have to track down a copy for my
library/reference collection.

Signature

Mahalo for your help,
Camper

> Camper -
>
[quoted text clipped - 24 lines]
> > function:
> > =IF(x>0,TDIST(x,deg_freedom,tails),(1-(TDIST(ABS(x),deg_freedom,tails))))

Reply to this Message