***** To join INSNA, visit http://www.insna.org *****
I have Twitter data (i.e. followings, followers, Tweets, etc) for three countries' ministries (at least 35 ministries from each country). I am interested to check if the three countries differ statistically in terms of followings, followers, and Tweets. Which test should I be running? Can I also conduct some correlations analysis on this data, for example, correlation between followers and Tweets?
Second, what kind of sampling strategy would you recommend to select a representative sample from the tweets for conducting semantic analysis? (Combined tweets of all the ministries are more than 170,000).
Thank you for help in advance.