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.

When Stack Exchange analyzed user comment friendliness using Amazon's Mechanical Turk during the Summer of Love, was any research done on whether the OP's Gravatar and/or use of a real name influences the friendliness of comments on your contribution?

If so, could you share your findings with us?

I'd be interested in seeing whether showing your (or, well, someone's) real-world face and/or name tends to make people treat you more courteously than a default Gravatar, or whether there seems to be no significant difference.

Edit: the Excel workbook referenced in the blog post doesn't seem to contain question IDs. With IDs, one could find this out through the API. Could this be added to the data?

share|improve this question
People are always nice to me, but that may just be because I'm such a charming individual. –  Tim Stone Oct 21 '12 at 20:32
@Tim yeah. And you're exceptional anyway! Plus you use your real name - I added that to the question. I'll bet people are less nice to user122345 type users asking bad questions than real name ones –  Pëkka Oct 21 '12 at 20:35
I would also like to know if humans are treated better than creatures of other species, such as unicorns. –  BoltClock's a Unicorn Oct 21 '12 at 20:36
I don't have anything conclusive to offer. Anecdotally, some days I'm pretty well respected and other days I'm a clueless girl who interprets things emotionally with feelings. So. –  Anna Lear Oct 21 '12 at 20:40
-1, I do not like your gravatar face. Interesting question, although I'm personally pretty skeptical about the analysis that was done being meaningful at all. –  Josh Caswell Oct 21 '12 at 20:49
@BoltClock owls are treated better than viscachas, that much is for sure. We learned that the hard way over at ELU. –  ЯegDwight Oct 21 '12 at 20:52
@BoltClock'saUnicorn I propose that anytime anyone downvotes a unicorn, an IE6 compatibility question is raised. That should sort that out. –  dash Oct 21 '12 at 20:52
Another good question to raise is does posting your own photo make you be nicer to others? It would be interesting to see what percentage of comments, rude and polite, were left by people who put up a real photo or use a real name vs those who use the default avatar or a fake photo..... –  jmort253 Oct 21 '12 at 21:10
Having a picture associated with your account (and I know I don't but seriously, no one wants to see my face) may actually do as you suggest; for example, putting a photo in your wallet makes it more likely that it gets returned to you: lifehacker.com/5830113/…. From that link we can conclude that putting your avatar up as a really cute baby may make it much harder for people to downvote you ;-) –  dash Oct 21 '12 at 21:14
I would like to see three categories for names: user12345, a nickname or joke name (Pekka, Bill the Lizard) that some can connect to a real name but not all can, and a real name. Similarly for avatars, a default geometry one, a not-a-picture-of-me-but-it-says-something like mine, and an actual picture of someone who appears to be the account owner. I think that if there is a "helping a real person effect" it is stronger for the last one in each list than for the middle, while the middle will be stronger than the first ones. –  Kate Gregory Oct 21 '12 at 21:19
@dash: I've seen at least one user who uses the same baby picture as that Lifehacker article. It's not helping a lot in that case, I'm afraid. –  Josh Caswell Oct 21 '12 at 21:33
@Kate Pekka is indeed a real first name. :) I get your point though, your categorization (and thinking of a "real person effect" rather than separate factors) sounds really good –  Pëkka Oct 21 '12 at 21:39
@AnnaLear: People with cats in their profile always get treated better than anyone else. –  Robert Harvey Oct 21 '12 at 21:59
interesting idea, but there's possible confounding variables; it's quite likely that someone with a gravatar is more familiar with SO, thus posting more properly formed questions –  Ben Brocka Oct 21 '12 at 22:13
For that matter - what about parrots with people in their profiles? –  Pëkka Oct 22 '12 at 7:13

2 Answers 2

I can tell you personally that, unless someone has something offensive in their picture or username, I believe that their profile pic has no effect on me at all, unless perhaps it's subliminal:

answer my post @_@

The effort that folks put into making their question understandable, forming complete sentences and using decent capitalization and punctuation far trumps the effect that any user name or profile picture may have.

share|improve this answer
It's the entire package that has an effect, not just 1 thing, so yes, people who write clearly generally will get better treatment. I also suspect it's all subliminal. I don't think anyone says "Hey, I'll be a jerk to that guy because he looks like a cartoon character...". Plus, this raises the question of whether or not people who use different types of photos tend to behave differently to attract or encourage others to react badly to them... In other words, maybe it's not the photo, but the fact that unprofessional people are possibly more likely to use unprofessional photos... –  jmort253 Oct 21 '12 at 22:28
I'd agree that more personalization / personification of a profile doesn't help you get treated any nicer, but I am wondering if the lack of it might earn you a bit more hostility if you asked a vague, poorly formatted and broadly unanswerable question. –  Tim Post Oct 22 '12 at 0:29

I'd be interested in seeing whether showing your (or, well, someone's) real-world face and/or name tends to make people treat you more courteously than a default Gravatar, or whether there seems to be no significant difference.

Unless you make it into a double-blind trial - which means posting the same questions, answers or comments by users with and without pictures, so you could compare reactions, which would be rather tricky as it would duplicate content - you'd only measure correlation, and correlation is not the same as causation.

Even if it was proven that users who show their real faces tend to be upvoted more than those who never bothered to upload pics and still use identicons, we wouldn't know if they are upvoted because of that.

It is not unlikely that the use of real pictures vs. identicons is correlated to other factors, such as

  • whether you plan to participate in the forum in long run or only want to get a few technical answers

  • whether you are naturally more inclined to form social relationships (and therefore have a habit of posting your personal pic everywhere you can), which reflects in your communication style


I guess this would make sense only when comparing among new users with or without gravatars. (new = less than x amount of rep and/or time on the site)

If there's 10 new users who uploaded pics (someone who does this straight away exhibits an even stronger motivation to do so than somebody like me - if I recall, I didn't upload my picture until I'd already used SO for a few months) and 10 users who didn't, I bet you there's already something different between the latter and the former, in terms of their personality type and what not.

share|improve this answer
All very true and fair points. But seeing whether there is a correlation would be interesting either way –  Pëkka Mar 21 at 13:33
@Pëkka I agree - if it was significant, this would be interesting by itself, even if we couldn't tell exactly how it works –  Morawski Mar 21 at 13:37

You must log in to answer this question.

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