I did a quick edit to "I see" versus "oh really" and noticed that while four people had answered it, it didn't have a single upvote (until I added mine). I think that is a bit odd because it got a lot more answers than the typical question on ELL, which should mean it was more interesting than the typical question.

I created a simple SEDE query that returns zero score questions with more than two answers and was surprised by the number. Zero score doesn’t necessarily mean no upvotes though, and I haven’t had time to look through them. The one with the most answers and no votes is Why does Harry Potter put everything on his plate except peppermints in this passage?.

I don't mean this as a criticism of the people who didn't upvote; everyone is entitled to vote or not vote how they like. I would just like to understand why that question didn't earn any upvotes. Was it the topic? Was it the spelling error? Did someone lose their keys and undo their upvote by accident?

I think it's a lot easier to upvote answers. If they are correct and reasonably well-written, it seems obvious they should get an upvote. It's a lot harder to upvote questions. Sometimes the fluency of the author causes the question to not make a good first impression. Sometimes it takes a little effort to tease out what makes the question interesting. What sort of advice would you give question authors if they wanted to earn your upvote?

We've discussed this before, but the members of the community change over time so I think it would be good to revisit it. Some related prior discussions:

Questions with (multiple) Answers but no votes

Flaws in the voting model in re. questions

  • I must admit that if I cast a vote for a question, it is much more likely to be a down or close vote, or both. I don't know if that makes me a curmudgeon. Commented Feb 13 at 14:45
  • 1
    @MichaelHarvey What qualities do make you upvote questions? I've been giving more weight lately to whether the questions have answers that are interesting than to whether they are interesting to me (if the question is otherwise well-formed).
    – ColleenV
    Commented Feb 13 at 15:28
  • 2
    I tend to upvote well-written, clear, focused questions that show signs of research and at least some thought on the part of the questioner, and particularly where the answer is not obvious. Commented Feb 13 at 21:18
  • 1
    @MichaelHarvey I think it's important for a question to include some attempt to explain what the author is thinking. I don't mind too much if the answer seems obvious to me as a native speaker. I think some answers are only obvious if you know them :)
    – ColleenV
    Commented Feb 13 at 21:23
  • 2
    I'm unsure your purpose here. Are you asking people who read that question and didn't upvote it why they didn't upvote it? Or asking for opinions why? Or is lack of upvotes for otherwise well-received questions a topic you want to draw attention to generally? Or something else?
    – gotube Mod
    Commented Feb 14 at 10:35
  • 2
    @gotube I would like to know what sort of guidance we can give askers to earn those upvotes. This question had a reasonable amount of effort put into it in my opinion, and spurred 4 answers, so I'm wondering what more it needed to do to get upvoted. It became a hot network question, so it's not a great example any longer I guess. I think it would be good to just discuss as a community our standards for upvoting.
    – ColleenV
    Commented Feb 14 at 11:31
  • 2
    It's become the norm the only ELL questions which earn >2 upvotes are those that ride on the Hot Network Ferris wheel. It is the folk outside the ELL community who get excited and cast votes. They tend to be original questions which spur users to exclaim "Oh yeah, I'd never thought 'bout that before" Qs with a quirk, a twist on something that native speakers take for granted. Users who answer assiduously are nearly always the same native speakers, they generally post good answers so it's easier to upvote their contributions rather than the askers. Unless your name is a certain Tom...
    – Mari-Lou A
    Commented Feb 14 at 19:10
  • A tangent, but I'm always surprised at people who answer questions, then don't upvote them. If I can't stomach to upvote a question, I won't answer it. Overwhelmingly answerers get the most upvotes --as it should be since they provide the value-- but I think expecting there should be at least one upvote per answer is reasonable.
    – gotube Mod
    Commented Feb 14 at 21:02
  • 1
    @gotube it’s reasonable to answer a question that you think doesn’t deserve an upvote. I’d find it odd for someone to downvote or close vote a question they answered though.
    – ColleenV
    Commented Feb 15 at 2:47
  • There's nothing like threads in Meta purporting to tell me how to vote to make me do the exact opposite. Commented Feb 15 at 13:29
  • 1
    @MichaelHarvey I hope you don't feel like that's what I am trying to do with this post. I don't like being pressured to behave a certain way either. I'm genuinely interested in different people's approaches to voting. The only "wrong" way to cast a vote on SE is to target a user instead of their posts.
    – ColleenV
    Commented Feb 15 at 14:30
  • 1
    I very much prize the anonymity of the voting system on SE, and I particularly resist any pressure to explain downvotes. I try to cast votes purely to make questions or answers rise or sink, and not to reward or punish. It is absolutely wrong to target a user. Commented Feb 15 at 14:34
  • 1
    @MichaelHarvey I think anonymous voting is absolutely essential to this system. My husband drove Uber for a while, and the rider could immediately see what a driver rated them and could retaliate. Rider ratings in that system were pretty useless because driver rating under 4.8 of 5 gets them throttled and eventually kicked off the system. When I do take the time to explain my downvote it never helps. I get accused of picking on people. Also, I tend not to want to spend more effort on a post than the author did.
    – ColleenV
    Commented Feb 15 at 14:43

1 Answer 1


Well, there is a bit of a prompt for the upvote. The hover-over text says "This question shows research effort; it is useful and clear." Ideally this should be the reason for an upvote.

IMO it should not be used to document the fact that you "had fun" reading the question or are particularly interested in it. The up- and down-vote systems were originally built to help answerers sort or filter questions by how answerable they were. See my thoughts on this question's inverse: This isn't really a need without a high volume of questions. So in our SE both the up- and down-votes aren't really useful, except to help a new user build their initial reputation score.

To translate and amplify that tooltip text:

  • "...shows research effort": This shouldn't just be some sort of work ethic that rewards diligence. For one thing, it's about avoiding duplicate questions: the user might say "I saw this or that similar Q, but my Q differs because ___. Also, especially in this SE, sometimes users have been confused by their outside research. Perhaps they read some bit of guidance and have misunderstood it, or misapplied it, or it was just plain bad. A good answer gives us that source as they found it, so we can address the misunderstanding, not just the confusion that came from it.
  • "...it is useful": Honestly, this one doesn't matter that much for this SE either. The Music SE that I'm also on puts a high priority on proving that each question might be of value as a reference to future readers. But ELL is more comfortable helping individuals with their immediate problem even if no one else will have the same one. Still, some questions might truly not be useful, like if the questioner is asking a bit of trivia or in bad faith. I think we could read this guidance as "the question is useful to the questioner."
  • "... and clear": Of course, if a question isn't clear enough to be answerable, it should be closed, not downvoted. But we could award upvotes to especially clear questions.

On ColleenV's suggestion, let's turn the above into hypothetical advice about "how to ask a great question."

  • Start with research. It doesn't have to be exhaustive research, and it's okay if you haven't found answers that others consider obvious, but start with a bit of a look around. Do a bit of googling, and search within the ELL site. Sometimes what you find will not answer your initial question, but will change it or confuse you even more. Congratulations! Your changed question is a better one, and will get better answers because it's more specific. (Usually.)
  • Make it clear. If your question is about something you found, include what you found in your question. Make sure you give enough context. For instance, if you found a strange word or phrase in a book, quote enough of the surrounding sentences that it can be clear what's going on. Make sure that your post really asks a question, and make sure it's only one question. Avoid giving too much information, either, so that it's hard to tell what the question is.
  • Ask for answers that help you. This isn't usually a problem, because most people come here for help. But occasionally we might think of a question out of idle curiosity, or just want to make a point. For instance, "Why is the plural of 'goose' 'geese' but the plural of 'moose' is just 'moose'?" The answer is "just because," and that's not very helpful. Well, actually there's probably a good answer based on etymology, but... I can't come up with a better example right now of a question that doesn't have any substantive answer. Or you might have caught an inconsistency in a grammar book and ask "Isn't that inconsistent?" If you're not actually confused, and just want to point it out, then it's not a good question. General rule of thumb: if you don't need answers, don't ask the question. There are other venues to simply discuss something, like the chat channel here, or Reddit.
  • The up- and down-vote systems were originally built to help answerers sort or filter questions by how answerable they were No, people who are interested in the question getting answered are allowed to upvote questions regardless of how "answerable" they are. Regardless, I was asking what qualities you personally look for in a question that makes you upvote, not what you think all voters should use as a standard. If you were talking to someone who is asking a question, what advice would you give them for getting your upvote?
    – ColleenV
    Commented Mar 7 at 20:58
  • @ColleenV Well, I guess just what I mentioned here, though I could reword it in the form of advice. One challenge is always that a high percentage of askers are newcomers and there's a lot of "structure" to SE that they haven't yet noticed. Commented Mar 7 at 21:29
  • Ok, so why didn't my question earn your upvote? I would be interested in how you would reframe this advice for askers though. Also, ELL is in the top 20ish sites for questions, so I disagree with your assertion that we don't need to rank questions.
    – ColleenV
    Commented Mar 7 at 21:33
  • 1
    @ColleenV Hehe good point, I just answered a question without upvoting it. I guess maybe for me personally it's because I've lost faith in the votes being useful. But it definitely meets the criteria I listed. Meanwhile, I'm editing this answer. Commented Mar 7 at 21:36
  • I am not demanding your upvote :) I'm just trying to get us to think about why many of us upvote more on answers than questions.
    – ColleenV
    Commented Mar 7 at 21:38

You must log in to answer this question.

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