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.

I saw the ad for Startups.com and decided to check it out. It seems to me like the audience it's attracting is very stingy with their votes! Do they need to RTFM better? Do they need more anthropology done on them?

But more importantly, how can I find out if my gut feeling is correct: Do/can we have these kinds of statistics available on any / all stack* sites?

share|improve this question
    
Your question is about a StackExchnage site and belongs here: meta.stackexchange.com –  random Nov 12 '09 at 12:16
    
Also, why not just ask on their site? –  random Nov 12 '09 at 12:29
3  
Isn't his question about stackoverflow and serverfault too? –  Dominic Rodger Nov 12 '09 at 12:46
1  
The question is about why Onstartup users are stingy with their upvotes. SO and SF were just weaseled in. Continues on asking about StackExchange. –  random Nov 12 '09 at 13:13
    
Actually, this question is about whether Onstartup users are stingy with their upvotes, especially compared to StackOverflow. I'd have thought such statistics would be most likely to be available on stack overflow (for instance, i've just found statoverflow.com/sandbox) –  Rob Fonseca-Ensor Nov 13 '09 at 12:08

2 Answers 2

Votes per user per day isn't a useful measure unless it takes user activity into account, and there's no real way to tell how engaged a user is. A user who isn't around anymore isn't going to vote on new content, no matter how much they participated in the past.

If you're interested in how much the community votes as a whole, votes per post are a good measure. This Data Explorer query shows the total number of upvotes and downvotes for all time (excluding votes on deleted posts) divided by the total number of non-deleted posts. Cross-site queries are pretty expensive, here's the data as of 2014-08-28:

                    up    down
skeptics           12.29  1.66
meta                7.95  1.57
workplace           7.81  1.14
cstheory            8.00  0.82
academia            8.11  0.52
programmers         7.71  0.85
scifi               7.91  0.56
rpg                 6.89  0.31
space               6.89  0.29
aviation            6.86  0.22
matheducators       6.65  0.21
history             5.93  0.84
earthscience        6.21  0.38
outdoors            6.19  0.21
pets                6.15  0.23
mathoverflow        5.99  0.38
travel              6.04  0.30
sustainability      6.09  0.24
codegolf            5.85  0.34
bricks              5.95  0.17
movies              5.61  0.45
christianity        5.02  0.95
tex                 5.74  0.06
boardgames          5.35  0.19
mathematica         5.35  0.19
french              5.23  0.25
japanese            5.13  0.31
genealogy           5.19  0.13
moderators          5.03  0.23
gaming              4.59  0.64
cogsci              4.69  0.51
beer                4.96  0.23
parenting           4.90  0.27
puzzling            4.72  0.41
reverseengineering  4.80  0.32
german              4.89  0.22
ebooks              4.81  0.23
security            4.63  0.38
hermeneutics        4.30  0.70
italian             4.72  0.23
expatriates         4.74  0.20
sports              4.38  0.45
audio               4.59  0.21
music               4.56  0.22
chess               4.53  0.21
gardening           4.64  0.09
photography         4.48  0.22
politics            3.87  0.83
anime               4.36  0.33
tridion             4.45  0.07
ux                  4.30  0.20
judaism             4.10  0.34
money               4.08  0.33
astronomy           4.10  0.31
english             3.80  0.54
russian             3.94  0.36
bicycles            4.02  0.25
cooking             4.01  0.25
islam               3.30  0.91
linguistics         3.93  0.24
softwarerecs        3.76  0.40
scicomp             4.03  0.12
martialarts         3.85  0.27
cs                  3.68  0.37
crypto              3.74  0.31
biology             3.75  0.27
networkengineering  3.72  0.30
quant               3.59  0.42
writers             3.84  0.17
productivity        3.83  0.14
startups            3.79  0.14
gamedev             3.51  0.39
datascience         3.59  0.23
fitness             3.56  0.26
spanish             3.61  0.18
codereview          3.62  0.16
freelancing         3.56  0.19
philosophy          3.16  0.57
ham                 3.52  0.16
chinese             3.34  0.28
hinduism            3.24  0.38
craftcms            3.52  0.04
physics             3.15  0.38
poker               3.28  0.22
blender             3.30  0.11
webapps             2.84  0.54
pm                  3.20  0.16
bitcoin             2.86  0.49
electronics         2.94  0.37
unix                3.12  0.14
opendata            3.10  0.11
diy                 2.92  0.14
stats               2.94  0.12
homebrew            2.94  0.09
buddhism            2.78  0.18
ell                 2.73  0.21
StackOverflow.Br    2.65  0.29
dsp                 2.71  0.19
robotics            2.77  0.12
chemistry           2.68  0.21
windowsphone        2.62  0.26
raspberrypi         2.67  0.13
graphicdesign       2.56  0.23
dba                 2.55  0.18
tor                 2.57  0.15
android             2.33  0.38
garage              2.52  0.14
math                2.48  0.17
sqa                 2.47  0.12
arduino             2.42  0.16
gis                 2.44  0.10
salesforce          2.39  0.12
SuperUser           2.26  0.22
ubuntu              2.19  0.24
apple               2.22  0.20
StackOverflow       2.17  0.23
webmasters          2.16  0.15
patents             2.01  0.28
joomla              2.01  0.17
ServerFault         1.87  0.21
sound               1.85  0.07
expressionengine    1.57  0.04
drupal              1.36  0.23
wordpress           1.20  0.25
magento             1.08  0.13
sharepoint          1.13  0.05
share|improve this answer
    
these stats are skewed as these don't account for deleted posts. As an example, at Programmers about 2/3 of my downvotes (~20,000) are on deleted posts –  gnat Aug 28 at 13:01
    
@gnat I think that voting stats specifically for non-deleted posts are useful — the bias comes from the variation in deleting bad posts (especially closed questions) between sites. Also, you are an outlier, even for Programmers. –  Gilles Aug 28 at 13:07
    
real outlier seems to be Community ♦. As for the skew, it's not difficult to estimate. Total votes up and down per user are public and include those on deleted posts. Just sum these for all the users and compare against total of votes up and down per post –  gnat Aug 28 at 13:21
    
Observation: the five largest sites, measured by total number of questions (SO, Math, SU, SF, Ubuntu), are way down on the list. But the 6th largest, TeX, does remarkably well in the voting department. –  Weapon of Choice Aug 28 at 18:16

You can't really answer that question because votes are anonymous, but you can have a look at the down- and upvotes given on the not deleted posts in SEDE.

Unfortunately you have to run that query on every site to make a comparison, because cross-site queries are not (yet?) supported.

Here is the basic query that gets the data for you:

select creationdate
, sum(case when votetypeid = 2 then 1 else 0 end) as Upvotes
, sum(case when votetypeid = 3 then 1 else 0 end) as Downvotes
from votes
where votetypeid in (2,3)
group by creationdate
order by creationdate

A nice graph as result for Stack Overflow:

Enter image description here

And compare that with StartUps:

Enter image description here

share|improve this answer
1  
FYI: While I'll admit my SQL is weak, and I don't know MS' flavour at all, this query (From this answer ) seems to indicate cross-site queries are doable, if possibly awkward. –  Billy Mailman Aug 27 at 21:48
    
Oh, cool, I'll give that a testdrive. You need to do the heavy lifting your self by the looks of it but I see how it can work. –  rene Aug 28 at 6:39
    
SEDE stats are skewed as these don't account for deleted posts –  gnat Aug 28 at 13:02

You must log in to answer this question.

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