I’ve built a model on excel to calculate chi-squared on each of my 8 questions within questionnaire. i'm using chi-squared to understand relationship between scores provided and organisation. |
I'm getting an calc. error (#DIV/0) for the p-value because no one from either category scored a zero "0" or a five "5". |
Excel is throwing an error in the total expected value which is throwing the =CHISQ.TEST (p-value) calc. into error. |
For the =CHISQ.TEST calc. do i therefore need to discount zero and five as categories and reduce my degrees of freedom by 2? |
Is that an acceptable method? |
Observed Total | 0 | 1 | 2 | 3 | 4 | 5 | |
CAT 1 | 0 | 1 | 1 | 4 | 14 | 5 | 25 |
CAT 2 | 0 | 1 | 1 | 0 | 1 | 1 | 4 |
0 | 2 | 2 | 4 | 15 | 6 | 29 | |
Expected Total | 0 | 1 | 2 | 3 | 4 | 5 | |
CAT 1 | 0 | 2 | 2 | 3 | 13 | 5 | 25 |
CAT 2 | 0 | 0 | 0 | 1 | 2 | 1 | 4 |
0 | 2 | 2 | 4 | 15 | 6 | ||
Chi Sq | 0 | 1 | 2 | 3 | 4 | 5 | |
BAE | #DIV/0! | 0.304 | 0.304 | 0.088 | 0.088 | 0 | #DIV/0! |
MOD | #DIV/0! | 1.901 | 1.901 | 0.552 | 0.552 | 0 | #DIV/0! |
#DIV/0! | 2.205 | 2.205 | 0.640 | 0.641 | 0.000 | #DIV/0! |