Found this question and answer searching "confirmation bias" and "accepted answer". I think the green check box might be too close to the up / down-vote controls. I don't question the existing up votes for questions, but I had observed that accepted answers tend to get the most votes -- which I'll admit makes a lot of sense for all the reasons cited in the answer to the linked question. A lot of factors can go into why someone will up-vote.

The only reason I raise this question is that I caught myself feeling an unbidden urge to up-vote the accepted answer on a question I had just up-voted, even though I hadn't really read the answer to see if it helped me in my particular situation or was particularly well written. I think, particularly for long time users of SE sites, the green check and reputation bumps begin to become Pavlovian stimuli, creating a positive feedback reward in our minds. We respond favorably to the shade of green, perhaps.

I'm not suggesting that we change any mechanics or stats, just that the placement of the green check be reconsidered.

Does it seem to you that the green check lends some authority to an answer beyond it's content sometimes?

What if the green check were moved to the right side of a question so that the effect of its proximity to the up/down arrows was limited?

Alternately, what if we changed the color of the check from green to brown, red, purple or some disassociated color? Would that create confusion or reduce the usability of the design or help dissipate confirmation bias?

  • 1
    The whole concept of having a tick there is for confirmation bias... – user392547 Aug 27 '18 at 13:20
  • @Chair - or simply to indicate which answer helped solve the problem posed by the questioner? Confirmation bias seems like something to be avoided if the best overall answer is the one we hope floats to the top. I'm not saying don't show which answer was accepted, just that the color and placement of the check may be artificially skewing votes because of the positive feedback. Perhaps my query is better aimed at the positive association of the indicator rather than the confirmation bias it creates. – J E Carter II Aug 27 '18 at 13:29
  • But indicating which answer helped to solve the problem is confirmation bias! SE wants to tell readers that they should attach a lot of importance to a particular answer because it helped the OP. Anyways, I'm against this. Green ticks look nice and feel correct. Green stands for correct stuff, and though I'm one of those guys who complains about accepted answers being wrong, changing the colour of the tick mark is not the way to solve the problem. – user392547 Aug 27 '18 at 13:32
  • Fair point, @Chair - what if the green check were distanced from the up/down arrows somewhat but still clearly associated with an answer? Do you think that would still convey "correctness" but potentially break the perhaps unfounded urge to up-vote? – J E Carter II Aug 27 '18 at 13:58
  • 4
    The more problematic bias is that the user-accepted answer is forced to the the top of the highest-voted sort, both physically and psychologically. First seen = "best answers rise to the top"(sic) – Robert Cartaino Aug 27 '18 at 15:03
  • Is it always @RobertCartaino? I hadn't noticed that was always the case.... but agreed, that is also a form of biased information display. Maybe it's just me, I don't experience the same reflexive urge to always up-vote the top listed answer (in the cases where there is no accepted answer and it sorts by votes) as I do when I see that green check. – J E Carter II Aug 27 '18 at 17:44

You must log in to answer this question.

Browse other questions tagged .