To calculate the probability
of observing the outcome by chance, one needs also to calculate the
number of degrees
of freedom.
When a comparison is made between one sample, with categories, and
another sample, as in our coin flip example, a simple rule is that the
degrees of freedom equal the product of (number of columns minus one)
x (number of rows minus one). This calculation is done without counting
the row and column containing the totals.
That is, (c-1) x (r-1).
|
Number of heads
|
Number of Tails
|
Totals
|
Coin A |
5
|
5
|
10
|
Coin B |
7
|
3
|
10
|
Totals |
12
|
8
|
20
|
For the data above, this gives (2 - 1) x (2 - 1) = 1.
Another way of looking at this computation is to ask for the minimum
number of figures that must be supplied, in addition to all the totals,
that would allow us to complete the table. Here, any one figure in any
of the four cells, along with the row and column totals, enables us
to complete the table. Thus, there is one degree of freedom in this
2 x 2 table.