My apologies for the really basic nature of this question, but I haven't done any statistics in years and am just getting back into the swing of things - doing a postgrad course after years in corporate.

We have collected data on a population of n dogs, doing a focal sampling count for 6x different behaviour patterns.

We carried out these observations for each dog:

1st time with the observer in position A

2nd time with observer in position B

Our hypothesis is that the observer's position B will increase the count of these behaviour patterns.

I did a X-squared test which came back suspiciously high (7.09, meaning that the probability of this difference being by chance is less than 1:100.

Was I correct in using a X-squared test? Is there a more appropriate test for this hypothesis?