The problem is that there is bandwagon behavior: a comment with a negative score will tend to be downvoted further. (This also applies on the positive side.)
Does this happen? I think my behaviour is actually the opposite, I sometimes upvote comments that I think have been downvoted unfairly even though I probably wouldn’t consider them worth an upvote otherwise.
It would be startling if conformity bias didn’t operate here, and I don’t see much evidence that it doesn’t.
That said, I don’t think these are mutually exclusive. I often upvote after (IMO) unearned downvotes, and sometimes downvote after unearned upvotes, but that doesn’t mean I’m not subject to the bandwagon effect.
The bandwagon effect is real, I think. My own behavior is to pay particular attention to heavily upvoted or downvoted comments to see if I can see the reason for the excitement. If I can’t find a reason, I will often vote the opposite way, as you do. But I usually find the reason. And then I can’t resist adding my voice to the crowd’s.
And since the upvoting and downvoting is silent and anonymous, the reasons for it should and I think do(#) tend to resemble the reasons of democratic voting, which reasons were discussed in Bryan Caplan’s Myth of the Rational Voter—the point of the book being that the reasons tend to be irrational. The result is a phenomenon that is overall irrational, with occasional exceptions.
Granted, it might not be much of an improvement if voters had to add an explanation, since humans are nothing if not fantastic rationalizers.
(#) I say I think, because since the voting is silent and anonymous, no one but the voters can actually know, so anyone else is forced to come up with a hypothesis which fits the voting pattern.
Does this happen? I think my behaviour is actually the opposite, I sometimes upvote comments that I think have been downvoted unfairly even though I probably wouldn’t consider them worth an upvote otherwise.
It would be startling if conformity bias didn’t operate here, and I don’t see much evidence that it doesn’t.
That said, I don’t think these are mutually exclusive. I often upvote after (IMO) unearned downvotes, and sometimes downvote after unearned upvotes, but that doesn’t mean I’m not subject to the bandwagon effect.
The bandwagon effect is real, I think. My own behavior is to pay particular attention to heavily upvoted or downvoted comments to see if I can see the reason for the excitement. If I can’t find a reason, I will often vote the opposite way, as you do. But I usually find the reason. And then I can’t resist adding my voice to the crowd’s.
And since the upvoting and downvoting is silent and anonymous, the reasons for it should and I think do(#) tend to resemble the reasons of democratic voting, which reasons were discussed in Bryan Caplan’s Myth of the Rational Voter—the point of the book being that the reasons tend to be irrational. The result is a phenomenon that is overall irrational, with occasional exceptions.
Granted, it might not be much of an improvement if voters had to add an explanation, since humans are nothing if not fantastic rationalizers.
(#) I say I think, because since the voting is silent and anonymous, no one but the voters can actually know, so anyone else is forced to come up with a hypothesis which fits the voting pattern.