Microsoft Office Tutorials and References

In Depth Information

**Testing to the T**

A t-distribution is similar to a normal distribution. The plotted shape is a bell

curve. However a t-distribution differs particularly in the thickness of the

tails. How much so is dependent on the degrees of freedom. The degrees of

freedom roughly relate to the number of elements in the sample, less one.

All t-distributions are symmetrical around 0, as is the normal distribution. In

practice, however, you always work with the right half of the curve —

positive t-values.

To use the TDIST function, follow along:

1. Position the cursor in the cell where you want the result to appear.

2. Enter
=TDIST(
to start the function.

3. Enter a value for t, or click a cell that has the value.

4. Enter a comma (
,).

5. Enter the degrees of freedom.

6. Enter a comma (
,).

7. Enter one of the following:

•1for a one-tailed test

•2for a two-tailed test

8. Enter a
).

If the t-value is based on a paired test, then the degrees of freedom is equal to

1 less than the count of items in either sample (remember, the samples are

the same size). When the t-value is based on two independent samples, then

the degrees of freedom = (count of sample-1 items – 1) + (count of sample-2

items – 1).

The TINV function produces the inverse of TDIST. That is, TINV takes two

arguments — the probability and the degrees of freedom — and returns the

value of t. To use TINV, follow along:

1. Position the cursor in the cell where you want the result to appear.

2. Enter
=TINV(
to start the function.

3. Enter the probability value (or click on a cell that has the value).

4. Enter a comma
,).

5. Enter the degrees of freedom.

6. Enter a
).