What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 130 Stack Exchange communities.

Source: In which ways do unregistered users participate?

Data source: February 6, 2012 NDA dump

The results here are only for StackOverflow, as it pretty well dominates the network. The final month of February, 2012 was excluded because of partial data. The data presented covers all questions and answers.

In case it isn't obvious in the following graphs, the percentage lines (green) are plotted on the right-side secondary axis (note: the scales are very different).


Unregistered vs. Registered Questions/Month

  • The steep dropoff to zero for unregistered questions is due to requiring question-askers to register first, which was implemented at the end of September, 2011. The unregistered stats going forward are of course moot, but I figured it would be interesting to include for historical sake.
  • Even ignoring the tail where the unregistered rate was added into the registered rate, it does look a bit like the question-asking rate overall is increasing exponentially.

Unregistered vs. Registered Answers/Month

  • Even looking through the feature changes list, it's unclear why the answer rate suddenly stagnated around March, 2011. The tail has to be ignored because of the increase in user registrations.
  • The unregistered rate, although a bit hard to see from the graph alone, has been relatively constant since November, 2010. The drop in percentage share is due to the increase of the registered rate (which makes sense, again because of increased user registrations).
  • While the graph says nothing about post quality, the unregistered user participation rate is low enough that it may be worth considering abandoning unregistered answers entirely.

Unregistered vs. Registered Edits/Month

  • These stats include accepted suggested edits (which are real edits after they're accepted).
  • Not surprisingly, it looks like unregistered users edited questions more frequently than answers.
  • The big increase in edits in early 2011 is due to the implementation of the +2 rep bonus for an accepted suggested edit.
  • On average, about half of all posts get edited, so it's critical to have that area of the application work well for users.

If you'd like to have a look at all statistics I collected for this question, you can download the Excel spreadsheet here.

share|improve this question
    
This is the final report for this project. Thank you all very much for participating! While I only presented some results in their own post, there are still many SEDE queries I created that are directly linked from the suggestions (see the aforementioned link) which you can peek at in the future. Thanks! –  Jon Seigel May 20 '12 at 16:34
    
+1, very interesting. Excuse my appalling Stack Overflow knowledge, but what causes the explosion between 2009-07 - 2009-11? Is this when SO left beta or what? –  Matt May 20 '12 at 16:37
    
@Matt: It's only an explosion in unregistered participation; I'm not exactly sure why as there's very little documentation from that time, and I was new then as well. –  Jon Seigel May 20 '12 at 16:43
1  
They're my army of sock puppets. How do you think I gained all that rep? Through actual expertise? –  Pëkka May 20 '12 at 16:53
1  
Bunch of stuff happened in that timeframe, including the launches of Meta, Super User, and then this... Nothing that jumps out at me as obvious though. –  Shog9 May 20 '12 at 17:10

You must log in to answer this question.

Browse other questions tagged .