I am working on a quantitative analysis of Stack Overflow. I am particularly interested in the causal relationships in the system.

Examples of causal questions that I can think of are:

  • Does daily limit on reputation points cause users to contribute to the website constantly?

  • Does putting up a bounty for a question cause that question to get more answers?

  • Does being able to edit answers/questions cause high-quality answers/questions?

  • ...

Depending on the causal question, I am analyzing the data on Stack Overflow to verify the causal effect in the system.

If there is any causal question you have about the system that you would like to be verified with a data analysis. Please let me know. I will be happy to share my results with the community.

  • 5
    If you can answer these questions with any kind of certainty, Jeff will hire you immediately.
    – Jon Seigel
    Mar 23 '10 at 17:52
  • :), Well I am trying to answer as much as I can. I hope I can discover interesting knowledge about SO. Mar 23 '10 at 17:54
  • 2
    Sounds very similar to this other question: meta.stackexchange.com/questions/42396/… Mar 23 '10 at 18:10
  • Yes, exactly. Though our analysis are not limited to exogenously manipulated variables. Mar 23 '10 at 19:43
  • Just in case anyone is interested, we performed an analysis on StackOverflow dataset and our paper summarizing our results got accepted to the ACM/SIGKDD Workshop on Social Media Analytics. Here is the link to the paper: cs.umass.edu/~hoktay/pub/soma2010.pdf Feel free to drop me a line if you have any questions May 28 '10 at 22:30

I'd be interested in how users habits change after they hit 10k rep.

In other words do they spend less time on the site when they get to the last rep boundary and does their answer:vote ratio change.

  • Similarly after users hit 3k (It's a long jump from there to 10k, does that affect motivation & cause contributions to slow?)
    – Alconja
    Mar 24 '10 at 5:35
  • Do users get a specific badge when they hit 10k rep or similarly 3k rep? Or how can we identify the exact time at which a user reaches 10k/3k? Mar 27 '10 at 21:46

Do Badges help encourage the specific badge behavior, as per their design goal.


Does lowering the amount of rep on questions cause better questions?

  • "..rep on questions ..." is that reputation on questions? Do people get reputation when they ask questions or am I missing something? Mar 23 '10 at 18:31
  • @Huseyin - only when those questions get upvotes. It used to be +10 per vote, now it's only +5 for questions.
    – user27414
    Mar 23 '10 at 18:35
  • @Huseyin Which is a major rule change only a few days old, and certainly interesting to watch whether it causes any measurable changes in how questions are asked.
    – Pekka
    Mar 23 '10 at 21:26
  • That's really great! Thank you very much for the information. It would be really interesting to analyze that—though we will need to wait a little and gather some data since it is a new change in the system. Great! Mar 23 '10 at 23:05

Does receiving downvotes cause users to improve their posts? Do those improvements cause upvotes (or removal of downvotes)?

How do those answers vary with the reputation of the user in question?

  • 2
    Oh, I just noticed the age of this question (after answering). Is it still relevant? Dec 2 '15 at 20:25

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .