The US map is somewhat useless unless you normalize it for population (at the least, total population, although population of Internet or Twitter users would be better). It's not interesting that California has a lot more people with Twitter opinions on gun control than Idaho, given that California has a lot more people, period.

There is, as always, a relevant xkcd [1]

[1] http://www.xkcd.com/1138/

Also, time of day. It turns out people talk about guns a lot more when they are not sleeping.

Totally, it looks as if the nation is breathing (i kinda like that for aesthetic reasons). If we bin the counts by day, a lot of the temporal resolution will be lost. What do you suggest?

You made a good point. But it makes more sense to divide by the number of tweets (within the given time interval) from the state instead of the user count which would not normalize it.

We will get the necessary data (tweet flow) to normalize the counts. Thanks for your comment.

map updated. normalized by tweet flow. Thanks again.

Just wanted to point out that searching on the term 'gun' probably skews towards those who beleive in stricter legislation.

We mainly want to filter this keyword to see how the discussion of "gun" is affected by the events following Sandy Hook tragedy. The opinions classification is not done over the same dataset.