I've been on the StackExchange boards for quite some time now and it seems that people don't upvote other's answers anymore when browsing the site. At least from my experience.

Let's take this example. When I answered this rather simple question it had 3 views. It had 30 views and not a single viewer upvoted my answer although I think it basically answered the question with links to two related articles for more information.

In this other example there is a clear and simple answer that would - imho - deserve an upvote.

Note: It's not about that I demand reputation for that or anything, it's just an example. But an upvoted answer would at least get the question out of the "unanswered" tab.

So my question is: Shouldn't users be encouraged to upvote good information whenever they see it? At least that's what I like to do when I browse other topics, even if I don't understand much - a useful answer deserves votes, not just by the OP.

link
well, I always read the responses and upvote when they are good; the same for the question when it is interessant – kokbira Apr 25 '11 at 14:06
1  
we also have a reminder we now display when registered users arrive via web search.. see the screenshot I added here meta.stackoverflow.com/questions/89045/… – Jeff Atwood Apr 28 '11 at 4:01
@Jeff Oh, that's nice, something similar had come to my mind before. Thanks for adding this and taking input from the community. – slhck Apr 28 '11 at 7:58
feedback

2 Answers

up vote 28 down vote accepted

Understand that

  • 90% of traffic is from web searches

  • the way we calculate views is very strict by IP

  • it takes a minimum of 15 rep to cast any votes at all

So of those 30 views, 27 (90%) of them are statistically going to be from users who cannot vote even if they wanted to.

link
Wow the math adds up pretty well. Thanks Jeff! – Kyle Apr 24 '11 at 21:08
1  
I just noticed this phenomenon from "new" questions that couldn't really attract traffic from web searches. But I guess the 15 rep cap plays a part there too. – slhck Apr 24 '11 at 22:14
Interesting stats. – boehj Apr 25 '11 at 1:17
4  
@slhck "new" questions do appear in web searches and so should attract traffic that way the same as any other. The SE network is quite aggressively scanned by search engines IIRC. Take a question from near the top of the new questions list (maybe something ~30mins old), shove something similar in to Google, and see what happens! – DMA57361 Apr 26 '11 at 8:43
feedback

The two main reaons why people who read a question or answer and don't vote are:

  1. They shoot in from an external search like Google or Yahoo! and often do not have the reputation to upvote questions or answers that helped them solve their problem.

  2. They are already capable of voting, but not sure on the topic matter and do not want to cast a vote in an area they're not confident in or they're not able to test the validity of an answer.

link
Valid point with #2, but as for searches: As far as I've seen, most of the time this happens to questions that are relatively new (~4 hours) but still gather many views. – slhck Apr 24 '11 at 18:13
2  
A number of times I've done a google search when researching to answer a question that's new, and google returns the question itself in the search results. We get indexed pretty rapidly. – Majenko Apr 25 '11 at 10:38
Case #3: People like me or @studiohack (can) run out of votes in roughly about an hour. I need to rate limit my voting :| ( or have number of votes allocated increased - but this is a rare case anyway) – Sathya Apr 25 '11 at 15:49
@Sathya That's interesting, I didn't know about a cap for casting votes! – slhck Apr 25 '11 at 20:44
2  
@slhck There's a Badge for thatâ„¢ – Sathya Apr 26 '11 at 14:09
1  
#2 is it for me. I never upvote things I'm largely uncertain about, even if they're great answers. (I won't be casting upvotes on Unix questions anytime soon.) – Shinrai Apr 28 '11 at 17:36
feedback

You must log in to answer this question.

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