Why am I getting different values of degrees of freedom for the s.

[jeng]

I am trying to use excel to do some simple data analysis (t test). I am
analyzing a few different sets of data with the same number of samples each.
3 in a control group and 3 in a treatment group. By my understanding of
degrees of freedom...there are 4 degrees of freedom (3+3)-2. For some reason
though, for the different data (with 3 samples in each group) I am getting a
different number of degrees of freedom (ie. 2 instead of 4). If I have the
same number of samples in each analysis condition, shouldn't my degrees of
freedom be the same??

[Mike Middleton]

jeng -

Excel's Analysis ToolPak (Tools | Data Analysis) has three separate t-tests.
You did not specify which one you are using.

With 3 observations in each of the two samples, there will be df = 2 for the
paired-sample test, df = 3 for the unequal-variance test, and df = 4 for the
equal-variance test.

- Mike
http://www.mikemiddleton.com

"jeng" <jeng@discussions.microsoft.com> wrote in message
news:A3C97D43-A48C-4D9F-B1BA-9AECF14F4B73@microsoft.com...
>I am trying to use excel to do some simple data analysis (t test). I am
> analyzing a few different sets of data with the same number of samples
> each.
> 3 in a control group and 3 in a treatment group. By my understanding of
> degrees of freedom...there are 4 degrees of freedom (3+3)-2. For some
> reason
> though, for the different data (with 3 samples in each group) I am getting
> a
> different number of degrees of freedom (ie. 2 instead of 4). If I have
> the
> same number of samples in each analysis condition, shouldn't my degrees of
> freedom be the same??