15

Low rep and anonymous visitors who vote on posts see a message that their vote was recorded, but the information isn't visible anywhere. There is some reporting of this available to a limited number of users in the Moderation Tools (10K users), but that has no explanation, and it seems incomprehensible, so <10K users aren't missing much on that one. That reporting is also focused on site activity rather than your own posts.

It seems like this information on my own posts might be useful feedback. Is there a way for all users to access those vote counts on their own posts?


Note: this is a self-answered question in response to the issue being raised in another recent Meta question. Other answers are also welcome.

1

1 Answer 1

8

How can I access the information?

SEDE (Stack Exchange Data Explorer) is a provision for querying meta information. If you want to create your own queries, there is basic help here: https://data.stackexchange.com/help.

There is a huge existing base of queries people have contributed. There is some really cool and useful stuff there, worth checking out. For Super User, start here: https://data.stackexchange.com/superuser/queries?order_by=popular

The query, Feedback on my posts, gives you the anonymous and low rep votes on your own posts.

To use it, enter your userID, which you can find on your profile page for the relevant site. The search box at the top will contain user:userid; that number is what you enter. Your userID is site-specific; if you want to query your own posts on another site, you will need to enter your userID for that site.

Make sure it's displaying the correct site icon at the bottom, or enter the name of the site you want. It reports posts for the site you select. Also, you'll get someone else's posts on a different site if the query isn't set to the site you want; someone else will have the same userID on a different site.

Then click Run Query.

It produces a list of your posts (identified as links so you can directly access them), with vote counts identified as "Helpful" and "Unhelpful" (to distinguish them from upvotes and downvotes).

Only about a third of my own posts were reported. It could be that many didn't receive any outside votes.

How useful is it?

Usefulness is mixed. It's useful if you use it in a useful way. :-)

For a number of reasons, the anonymous and low rep voting doesn't always track with the qualified voting, and it is often wildly mixed -- as many people thought a post was great as thought it sucked. The raw vote counts don't always tell you much.

The value is when you can use the information as a sanity check, or to add insight. For example:

  • Validity of a downvotes. Say a number of your answers receive one downvote as the only voting. The answers seem fine to you, and you suspect malicious voting. If the anonymous & low rep voting is predominantly positive, that would support your suspicion. If the voting is predominantly negative, many other, impartial people also had a problem with the answers, suggesting the downvote was warranted.
  • Meaning of zero votes. A number of my own answers received no qualified votes of any kind. The posts seemed fine to me and I assumed it was because nobody cared about the question. However, the script showed a bunch of votes, all negative. That means those answers were universally assessed as unhelpful to the point of motivating negative votes. That suggests I was looking in the wrong direction for an explanation.
  • Identify posts that could benefit from improvement. If a post received little or no attention from qualified voters but a lot of "helpful" votes from others, it means people found value in it. Look for ways to improve the post, and the question to which an answer is attached. If the post received a lot of "unhelpful" votes, try to figure out why, and look for ways to make it better.

In general, the value will be in cases where the voting was predominantly in one direction, and especially where that is very different from the qualified voting, or shows a pattern when there was no qualified voting.

8
  • also relevant are the Tutorial and the Database schema documentation for the public data dump and SEDE
    – rene
    Jul 18, 2019 at 9:42
  • I edited that query you mentioned for Superuser, giving the credits to the original source.
    – CaldeiraG
    Jul 18, 2019 at 10:46
  • @CaldeiraG, LOL. I linked to the original script created in 2012, which defaults the site to SO. The ones you found and edited are exact copies created today by an anonymous user, that defaults to SU. It looks like you went to the queries list instead of the direct link, sorted it by Recent, and those were at the top. Maybe somebody who saw this post went and played with the script and saved it. And now there are two new copies of those copies you edited for attribution. :-)
    – fixer1234
    Jul 18, 2019 at 11:31
  • 3
    @fixer1234 the anonymous who created was actually me, unregistered :)
    – CaldeiraG
    Jul 18, 2019 at 11:33
  • I got myself registered after I found out I wasn't logged in there
    – CaldeiraG
    Jul 18, 2019 at 11:34
  • 5
    Curiously my most "anonymously downvoted" posts are on answers where I've stated "you can't do that" implying that for at least some people the downvote is effectively "this might be right but wasn't the answer I wanted". As such there are some answers that are "unhelpful" per that query but rather than being wrong are simply not really what people wanted to hear.
    – Mokubai Mod
    Jul 18, 2019 at 15:46
  • @Mokubai, might that be one of the reasons 125 rep is required to downvote? Yeah, interpreting those votes needs to consider the context.
    – fixer1234
    Jul 18, 2019 at 18:16
  • The Data Explorer is a very useful tool, but the top queries on it are both ancient and incredibly broken, so without some moderation of which queries are kept and whether they actually work, they are the worst possible advertisement for its capabilities. Jul 31, 2019 at 17:44

You must log in to answer this question.

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