For the biggest ones: How many of those active users are bots, advertisers, and scammers? I'd guess about half on Facebook.
Also, is it considered "active" if you have a dormant account but have the app installed on your phone and it still watches what you're doing? What if you only use it to communicate with family because it's the only internet they understand?
Further, what about duplicate accounts or "secretive" secondary accounts so you can click on the depraved stuff you like without that showing in your public feed?
I feel like the real numbers for the big ones are massively inflated by issues like these.
The Fediverse is small enough to as of yet not be affected. Once it gets large enough, it will have all of this, too.
with respect to bots, as of this time I don't think it's a problem that can be fully solved, although I do think over a long enough timeline the fediverse is probably the best suited to handle that problem.
I wanted to see a visualization of the relative size comparison, so I used the data that was available on Wikipedia, but this data is approximate at best.
The Fediverse is by design affected by inflated numbers. If one user uses three different services, the user is counted three times. However, for the Fediverse it doesn't really matter - that number of total users is just as irrelevant like the total number of used email addresses.
Also, is it considered "active" if you have a dormant account but have the app installed on your phone and it still watches what you're doing?
Almost all platforms use "monthly active users" - anyone that uses it at least once per month is considered an active user. If you have an app installed but don't use it, it doesn't count. Some platforms also provide a daily active users metric.