I’m sorry if my English is bad, my native is Russian. I have a dataset of UK general election results and want to compare two groups which are the seats gained by Labour party (36) and the remaining seats (614, with exclusion of NI seats it is 596) on some criterias (education, health, so on, based on 2011 Cencus. I’m newbee to statistics.

My questions are:

1) I’m not sure if the assumpution of independecy is valid in this case? I assumsed it is because two sets are not related with each other as it would be with research of ‘before and after treatment’ case

2) What method (if research is meaningful) should i choose for comprasion? Should i run Welch-U ranked data test, which i did and it shows very high number of degrees of freedom? Or Mann-Withney, which i suppose would be irrelevant because of median of remaining seats set will always be ~ 300-320? Or randomly choose 30 (lab gained seats) of 614 (596) ? Some kind of weighting?

i would appreciate any answers, any research, any papers with any level of complexity for newbee. Thank you for answer.