Upgrade your account to a Community Supporter account and remove most of the site ads.

Reply to thread

Message: <blockquote data-quote="Cadence" data-source="post: 7653143" data-attributes="member: 6701124">This seems like it would be a great source of hundreds of discussion points for a first course in statistics!I think I would actually just avoid talking about the kind of sample you have - you have a non-random convenience sample of some not-well defined larger population.  If that population is supposed to be all gamers who post on line then you are suffering from a huge source of bias due to undercoverage because your sampling frame is systematically missing all of the company web-sites. Its even farther away from all gamers.  Is it better to just say you have a census of "The most popular independent RPG sites and Blogs"?As far as what you're measuring.  You are only getting the count of "people who post about games" if your algorithm only counts the number of distinct posters over the 90 days once each.  Further you would need to assume that most people don't have at least two boards they use regularly.      You are even farther from getting the number of "people who play each game" (which one of the charts has as a label) since a lot of us seemingly post in threads for systems that aren't related to what we are currently playing.  Going with "Most discussed RPGs" like you say in several places seems appropriately nebulous enough that it isn't misleading.And then once you have your sample and a rough research question you get all the arguments about what data is most appropriate.  If it's threads, is a one post thread as good as a 1,000 post one?  If it's number of posts, what happens when a few 3.5ers who are on-line at the same time manage to crank out 20 pages of posts in a day and a half arguing about wizards versus fighters?   If there are lots of PF on-line games and OD&Ders like to play in person should all on-line games be excluded, or are you just going for total on-line traffic about the games?   If its threads, how does "All D&D count"?   If its post, how does a post talking about 4e as a counter example in a thread about a PF question count?   What happens if I don't like the auto-linking ENworld does when I type Pathfinder so that I use PF instead?   For posts, how does posting they wish they could give XP count versus actually gettting XP?  etc...Hmmmm....  I'm getting more and more glad it isn't my project to run and that I just get to see the cool results.   (Which I really enjoy, thanks!).</blockquote>

Verification