I'm doing a study into the factors which cause dogs to develop noise phobias.
One of my factors is the breed of dog. I have seven groups of dog breeds and have a number of dogs which do and do not have noise phobia for each of these seven groups.
I carried out a chi squared test (because I have two categoric variables: dog breed and yes/no noise phobic) and found that breed was statistically significant in the chance of the animals having noise phobia.
In order to find where the significance lies here I am looking to do a post-hoc analysis but I have no idea which statistical test I should be using here or how to go about it.