I’ve made the point that we tend to scrutinize posts/comments more carefully for flaws when they argue against beliefs that we hold, which results in subtly flawed arguments supporting a majority position being voted up despite being flawed, while a similarly flawed argument against a majority position will more likely be discovered as flawed and voted down. This results in the appearance that there are more valid arguments supporting the majority position than there really are, and drives away those who argue against the majority position because they think they are being treated unfairly.
Does anyone disagree with this, or think that Less Wrong voters have already adequately compensated for it? Or, if you think this is a real effect, but shouldn’t be called groupthink, what is the right name for it?
Real effect but not full-blown groupthink. For instance, I wouldn’t expect people on LW to be accused of “disloyalty”. I’ve said once or twice that I din’t think of myself primarily as rationalist, to no raised eyebrows.
I have not always been aware of the technical meaning of groupthink. But even then I was careful to distinguish between group cohesion one one hand, and irrational commitment to failing courses of action (or poisoned beliefs) on the other.
I suspect that we’re not criticizing the CraigsList project as harshly as we should. I suggested as much, but there was little follow-up. This strikes me as counterproductive; such a project should be subjected to the harshest tests and launched only if it survives unscathed.
The phrase “insider bias” comes to mind, but I’m not sure it captures what you mean either.
I suspect that we’re not criticizing the CraigsList project as harshly as we should. I suggested as much, but there was little follow-up.
I made one or two skeptical comments on that page, but I didn’t criticize the project as much as it deserved. It seemed so obviously nugatory. I didn’t want to spend the time and effort thinking about something that I wasn’t at all interested in, especially when everyone else seemed so excited about it. I just figured the people who really care about it can do it if they want.
There is a type of selection bias at work here, where people who aren’t at all interested in something may not feel the desire to spend time fighting the tide, which then makes the tide appear stronger. There are probably also people who feel this way about cryonics. (Not me: I think it’s awesome.)
I have not always been aware of the technical meaning of groupthink. But even then I was careful to distinguish between group cohesion one one hand, and irrational commitment to failing courses of action (or poisoned beliefs) on the other.
There seems too much discussion on this site regarding definitions. I call them “word arguments” and tend to skip them when I sense their presence. It seems people like to categorize things, in which case a word argument is an attempt to write dictionaries so everything make sense. I much prefer to try and understand what the other person is attempting to say, perhaps even without some specific, technical word to describe it, and continue the discussion in profitable directions.
In which case, the proper response to accusations of groupthink would be, “I disagree for reasons X and Y. Now to cover your othre points...” Or, in my estimation, even better, “Why does it seem like groupthink to you?” You don’t even have to ask them for their definition of the word to understand what they want to tell you, so long as they spell it out. And you don’t necessarily have to answer the accusation of groupthink, then, either, especially if their reasons are not contained in the definition of the word. Instead, you answer the points themselves: what the person was trying to convey in the first place.
It’s not I think that we don’t suffer from confirmation bias—of course we do, and I doubt that we do properly compensate for it—it’s that saying “groupthink” or “confirmation bias” without providing evidence doesn’t advance the debate any, and seems to constitute no more than a way of sneering at people for not agreeing with you.
It’s not I think that we don’t suffer from confirmation bias—of course we do, and I doubt that we do properly compensate for it—it’s that saying “groupthink” or “confirmation bias” without providing evidence doesn’t advance the debate any, and seems to constitute no more than a way of sneering at people for not agreeing with you.
There seems too much discussion on this site regarding definitions. I call them “word arguments” and tend to skip them when I sense their presence. It seems people like to categorize things, in which case a word argument is an attempt to write dictionaries so everything make sense. I much prefer to try and understand what the other person is attempting to say, perhaps even without some specific, technical word to describe it, and continue the discussion in profitable directions.
In which case, the proper response to accusations of groupthink would be, “I disagree for reasons X and Y. Now to cover your other points...” Or, in my estimation, even better, “Why does it seem like groupthink to you?” You don’t even have to ask them for their definition of the word to understand what they want to tell you, so long as they spell it out. And you don’t necessarily have to answer the accusation of groupthink, then, either, especially if their stated reasons are not contained in the definition of the word. Instead, you answer the points themselves: what the person was trying to convey in the first place. And you might be able to suggest they use a better word or phrase, once you understand them.
Perhaps they are not trying to convince you that there is more bias than you think (which would require evidence), but just reminding you that it does exist and is probably not being adequately compensated for? If we agree that a problem exists, then we should welcome such reminders (at least until we find some other way to solve the problem).
If you’re referring to what I think you are (can you perhaps point me to another instance where you’ve made this point?), I simply disagree that the respective arguments in question are “similarly flawed”.
One circumstance in which you can usually be safe in dismissing a minority argument is when the minority arguer has ignored the arguments already put forth for the majority position. If you want to be a contrarian, that’s fine, but you can’t expect to persuade the majority to your point of view without addressing the reasons they believe what they believe.
I came across the information that on the individual level, this is called belief bias:
Belief bias is a cognitive bias in which someone’s evaluation of the logical strength of an argument is biased by their belief in the truth or falsity of the conclusion. This effect has been demonstrated in psychological experiments, and is independent of reasoning ability.
(ETA: A search of LW and OB shows that we’ve never discussed this particular bias before, at least not by name.)
I agree 100%, and it occurs to me that the problem somewhat feeds on itself since there is a natural temptation to downvote in retaliation; to upvote “allies”; and so forth.
we tend to scrutinize posts/comments more carefully for flaws when they argue against beliefs that we hold
It seems likely that, given limited time to read comments, any given member will stick to the more controversial (and therefore interesting) comment threads, leading to an apparent piling on when disagreement is expressed.
I’ve made the point that we tend to scrutinize posts/comments more carefully for flaws when they argue against beliefs that we hold, which results in subtly flawed arguments supporting a majority position being voted up despite being flawed, while a similarly flawed argument against a majority position will more likely be discovered as flawed and voted down. This results in the appearance that there are more valid arguments supporting the majority position than there really are, and drives away those who argue against the majority position because they think they are being treated unfairly.
Does anyone disagree with this, or think that Less Wrong voters have already adequately compensated for it? Or, if you think this is a real effect, but shouldn’t be called groupthink, what is the right name for it?
Real effect but not full-blown groupthink. For instance, I wouldn’t expect people on LW to be accused of “disloyalty”. I’ve said once or twice that I din’t think of myself primarily as rationalist, to no raised eyebrows.
I have not always been aware of the technical meaning of groupthink. But even then I was careful to distinguish between group cohesion one one hand, and irrational commitment to failing courses of action (or poisoned beliefs) on the other.
I suspect that we’re not criticizing the CraigsList project as harshly as we should. I suggested as much, but there was little follow-up. This strikes me as counterproductive; such a project should be subjected to the harshest tests and launched only if it survives unscathed.
The phrase “insider bias” comes to mind, but I’m not sure it captures what you mean either.
I made one or two skeptical comments on that page, but I didn’t criticize the project as much as it deserved. It seemed so obviously nugatory. I didn’t want to spend the time and effort thinking about something that I wasn’t at all interested in, especially when everyone else seemed so excited about it. I just figured the people who really care about it can do it if they want.
There is a type of selection bias at work here, where people who aren’t at all interested in something may not feel the desire to spend time fighting the tide, which then makes the tide appear stronger. There are probably also people who feel this way about cryonics. (Not me: I think it’s awesome.)
There seems too much discussion on this site regarding definitions. I call them “word arguments” and tend to skip them when I sense their presence. It seems people like to categorize things, in which case a word argument is an attempt to write dictionaries so everything make sense. I much prefer to try and understand what the other person is attempting to say, perhaps even without some specific, technical word to describe it, and continue the discussion in profitable directions.
In which case, the proper response to accusations of groupthink would be, “I disagree for reasons X and Y. Now to cover your othre points...” Or, in my estimation, even better, “Why does it seem like groupthink to you?” You don’t even have to ask them for their definition of the word to understand what they want to tell you, so long as they spell it out. And you don’t necessarily have to answer the accusation of groupthink, then, either, especially if their reasons are not contained in the definition of the word. Instead, you answer the points themselves: what the person was trying to convey in the first place.
It’s not I think that we don’t suffer from confirmation bias—of course we do, and I doubt that we do properly compensate for it—it’s that saying “groupthink” or “confirmation bias” without providing evidence doesn’t advance the debate any, and seems to constitute no more than a way of sneering at people for not agreeing with you.
There seems too much discussion on this site regarding definitions. I call them “word arguments” and tend to skip them when I sense their presence. It seems people like to categorize things, in which case a word argument is an attempt to write dictionaries so everything make sense. I much prefer to try and understand what the other person is attempting to say, perhaps even without some specific, technical word to describe it, and continue the discussion in profitable directions.
In which case, the proper response to accusations of groupthink would be, “I disagree for reasons X and Y. Now to cover your other points...” Or, in my estimation, even better, “Why does it seem like groupthink to you?” You don’t even have to ask them for their definition of the word to understand what they want to tell you, so long as they spell it out. And you don’t necessarily have to answer the accusation of groupthink, then, either, especially if their stated reasons are not contained in the definition of the word. Instead, you answer the points themselves: what the person was trying to convey in the first place. And you might be able to suggest they use a better word or phrase, once you understand them.
Perhaps they are not trying to convince you that there is more bias than you think (which would require evidence), but just reminding you that it does exist and is probably not being adequately compensated for? If we agree that a problem exists, then we should welcome such reminders (at least until we find some other way to solve the problem).
You can do that on a case-by-case basis, but the problem is that accusations of groupthink are too easy.
Edit: Come to think of it, it wouldn’t be a good boo-light if it wasn’t easy to make the accusation.
If you’re referring to what I think you are (can you perhaps point me to another instance where you’ve made this point?), I simply disagree that the respective arguments in question are “similarly flawed”.
One circumstance in which you can usually be safe in dismissing a minority argument is when the minority arguer has ignored the arguments already put forth for the majority position. If you want to be a contrarian, that’s fine, but you can’t expect to persuade the majority to your point of view without addressing the reasons they believe what they believe.
I came across the information that on the individual level, this is called belief bias:
(ETA: A search of LW and OB shows that we’ve never discussed this particular bias before, at least not by name.)
I agree 100%, and it occurs to me that the problem somewhat feeds on itself since there is a natural temptation to downvote in retaliation; to upvote “allies”; and so forth.
It seems likely that, given limited time to read comments, any given member will stick to the more controversial (and therefore interesting) comment threads, leading to an apparent piling on when disagreement is expressed.