@domotorp This is probably rather naive question, but since you linked to query for SO, I will ask anyway: You know that you have to switch site to MathOverflow, right? (Apart from that, I think that questions from single user might be rather small sample to draw some relevant conclusions from them.) — Martin Sleziak 7 hours ago
4:24 PM
I would expect to see bigger fluctuations on site with more questions, so I tried it on math.SE: data.stackexchange.com/math/query/552109/…
To be honest, I am not sure how to interpret those results. (And not even whether the query does what it is expected to do.)
I should probably leave this to some people with better SQL skills. (I would try a bit more if I had more spare time. There some work that should be done during the weekend.)
BTW I wanted to link to the question rather than to a comment above. All this was in connection with:
2
![MathOverflow Meta](http://cdn.sstatic.net/Sites/mathoverflowmeta/img/apple-touch-icon.png)
I wonder if there is any data on whether questions receive more attention in some months, while less in some others.
Is it better to ask during Christmas or the summer break, at the beginning or at the end of the semester?
I hope that somebody with better SQL skills will try to give you some more useful data, but results of my (very simplistic) experiments are here. — Martin Sleziak 11 secs ago
Perhaps this room might be a good place to ask for help - but it was frozen for inactivity two weeks ago:
SEDE - Stack Exchange Data Explorer
Discussion related to the usage of Stack Exchange's Data Explorer
5:03 PM
2 hours later…
7:07 PM
3 hours later…
9:58 PM
@domotorp I guess that probably the safest choice is not to 2009 (since MO did not exists from the beginning fo 2013). And we also want to cut 2016 away. (We only want to use full years. We do not want the results to be skewed by recent posts - there are more recent posts without an answer than the general average.)
As I said, it would probably be better to split this by week than have results for every single day.
Still, the fact character of the graph changes close to the current date (it is more visible for math.SE) makes me suspicious that I am doing something wrong and that the query does not calculate what I wanted it to.
And even if it does, it is rather simplistic. Having at least one answer is not the same as being answered. (It should probably be restricted to answers with positive score. But that would definitely make the query more complicated.)
If I run it for StackOverflow, there is again a change near the current date. As I said, this look rather suspicious. But I am unable to spot where the mistake is.
However, whenever considering some statistics like this, I always recall the quote from the end of this answer.
"I don't feel like slicing the data by tags, time of day, day of week, month of year, phases of Moon, etc. The predictive power of this data is rather weak. The site is growing; its dynamics are different during school semesters and during holidays; some prolific answerers join, others quit or go on hiatus. Questions vary by difficulty and the level of specialization, which is something that tags do not capture. But if someone feels like slicing the data -- go ahead and fork the query."
"A question does not have to get lots of views to get answered: it needs to be seen by 1 person with the sufficient expertise, interest in the question, and time available. There is no way to predict when or if such a person will see the question. Choosing proper tags and informative title, writing a clear question in easy-to-read format, setting a bounty if needed -- all of these contribute to success more than pondering statistical tea leaves."
BTW as I mentioned in a comment, I think that adding statistics might be a reasonable thing to do. I guess that when somebody searches on meta for questions in this tag, then this belongs to the type of questions they are expecting to find.
« first day (673 days earlier) ← previous day next day → last day (2765 days later) »