Close Home Forum Sign up / Log in

Statistics Help Needed Please!!

B

Hi,

I am really struggling with statistics and I don't think it is even that difficult but I would be very grateful for any help!

So, I have as my dependent variables a continuous list of policy scores which I developed, these can be considered continuous.

I have some independent variables which are categorical e.g. political party, county council or unitary authority etc

How can i test if there is a relationship between them? Which tests do I used? I did all my work using chi-squared but apparently that is when they are both categorical so is wrong!

Also, I have one other set of continuous independent variables (population data) - I used Pearson's correlation - is that correct?

Thanks so much in advance!!!

Ben

C

If I'm understanding your data, for the first question I'd use a Kruskal Wallis test, your policy scores would be the outcome variable, and the political party (for example) the grouping variable. This would show whether there is a difference between the policy score dependeing on the political party. You could use and ANOVA if the data is continuous AND scale data (i.e. a score of 2 means the double of a score of 1).

For the relationship between the population data and the policy score, I'd use a Spearman's correlation rather than Pearson's correlation (unless the data is scale data).

Hope this helps.

B

Thanks so much! This has been unbelievably useful! I know it seems obvious but I've spent two days trying to figure this out!!

18888