Woo! We can now perform the tests; however, this isn’t very useful if we cannot interpret what they tell us!
To report our results, we need to:
- Report the p value
- Is the p value significant or not?
If it is significant
- Report our Cramer’s V value
- What strength is this? (Use the guide below to interpret)
- Describe some notable comparisons from our observed and expected count tables
Cramer’s V Interpretation Guide
Zero – No association between variables
0.2 – Weak association between the variables
0.4 – Medium association between variables
0.6 – Strong association between variables
0.8 – Strange (possible but check calculations)
1 – Variables are perfectly associated
Stop and Reflect:
What about the results from the previous sprint?
Revisit the previous sprint and jot down the above criteria.
Apply Your Thinking:
Using the dataset we have been working on, choose your own 2 variables to run a Chi-Square and Cramer’s V test on.
Do you have any cells with an expected count less than 5? If so, what do we do with this?
- What is your p value? Sig or not?
- Strength of the Cramer’s V if applicable