4
$\begingroup$

Update

I have requested this thread to be locked.

End of update

I am listing the four answers that came up in the thread "What should be our first Polystats Project?" as separate answers in this question. Please vote up or downvote the answers depending on which one question you think we should consider for our first project. BTW, this question is CW so that downvotes and upvotes have no rep consequences.

A few points:

  • We all should contribute to the project even if our favorite project question is not chosen by the community.
  • I will keep the poll question open for 3 days (i.e., till 22nd September, 00 hours UTC) at which time I will ask the mods to lock the question.
  • If there is a tie I will flip a coin to choose the specific question that we will look at.

Previous Questions reg Polystats-Project-1:

Question 1: What should be our first Polystats Project?
Question 2: What should be the data collection strategy for our first Polystats Project?

Relevant Meta Thread:

How about starting a Polystats Project?

$\endgroup$

migrated from stats.stackexchange.com Jun 19 '12 at 19:07

This question came from our site for people interested in statistics, machine learning, data analysis, data mining, and data visualization.

4
$\begingroup$

Are there reputation effects in SE communities and to what extent are reputation effects contingent on various variables such as size of the community, number of views per day, no of questions per day etc?

$\endgroup$
2
$\begingroup$

Quantify the quality of answers and questions (what makes a question/answer a good one?)

$\endgroup$
2
$\begingroup$

Is there any evidence for herd behavior? (i.e., Is the probability of a question/answer being voted up/down dependent on the existing up/down totals?)

$\endgroup$
1
$\begingroup$

Classify how the communities differ.

$\endgroup$
1
$\begingroup$

Create a better metric than '% accept rate' to identify users who 'give back to the community'.

$\endgroup$
0
$\begingroup$

Can we predict how long a user will remain active on the site? (nice Survival Analysis exercise.)

$\endgroup$
0
$\begingroup$

Can we cluster the tags somehow, or construct a similarity between them? (with an eye towards identifying redundant tags, perhaps.)

$\endgroup$
0
$\begingroup$

Does rep really correspond to expertise? (I guess this is similar to the other question, "are there rep effects?" or "snowball effect", but one possible outcome of this research would be a new way to compute rep.)

$\endgroup$

You must log in to answer this question.