I believe it. I've got friends on both sides of the political aisle. I can't believe some of the stuff my friends think is true, and they think the same of me sometimes.
I think confirmation bias is really strong. We're all best off if we take care to ingest a balance of news. (One of my techniques: I read 'RealClearPolitics' daily. It contains outrageous articles from both sides, which I hope gives me a more balanced perspective.)
This is interesting, but I think it's important to note one thing. Susceptibility to fake news != rate of believing fake news. This experiment seems to have measured susceptibility.
>For instance, in the first study stories were either favorable or unfavorable about either former President Barack Obama or current President Donald Trump. Favorable stories reported that the target had donated $50 million of their personal fortune to selected charities, while the unfavorable stories reported that the target was facing criminal charges related to voter fraud in the 2016 Presidential election. Participants then made "legitimacy judgments" of each story, rating how much they believed the story to be true, reliable, and trustworthy.
I'm pretty left-wing, so my biases would presumably lead me to believe the favorable story about Barack Obama, and disbelieve the favorable story about Donald Trump. [1] How does that work out in practice? Well, at first glance it seems pretty accurate. Trying my best to abstract away from the context of this study, I'd give about 2% credence to the story about Trump, and maybe 35% to the story about Obama (does Obama even have $50M of personal wealth?). Hypothesis confirmed, right!? Not exactly.
It's true that I'm answering these questions according to my existing beliefs about each politician. But those beliefs aren't simply the result of "bias". I came to those beliefs for reasons. For example, I know that Trump is infamously stingy. [2] Him donating 50 million in personal wealth would be a real shock to me, and for good reason. I know a bit less about Obama, but I don't have a strong reason to think he's stingy, and I do vaguely recall some news indicating that he has made a number of large charitable donations. [3]
So yes, both liberals' and conservatives' judgments are based on their beliefs, and if you put those people in a position where they have to use those beliefs to judge claims in the absence of other evidence, they are liable to believe lies. But one of these groups believes the truth - that Obama is more charitable than Trump - and the other does not. In other words, much of the liberal susceptibility to fake news can be explained simply by pointing at the justifiably higher credence they have in many pro-liberal stories.
[1] I'm relatively well informed, so I'd give ~0% credence to either story about criminal charges. But if one of these stories were true, it would probably be true about the person who's already under investigation for his actions in the 2016 election.
That has got to be one of the most fascinating meta-comments I've ever read, give that the article goes on to say:
> The researchers measured individual differences in thinking styles and found that regardless of the political party identification, when high need for cognition individuals were presented with fake news stories that were consistent with their ideology, they were even more likely than everyone else to judge the story as legitimate, and when they were faced with fake news story that were inconsistent with their ideology, they were even less likely to consider the news legitimate than everyone else.
One possible explanation is that highly logical and rational people are not only more likely to disbelieve politically-inconsistent news stories along tribal lines, but they are also more likely to seek out further disconfirming information, thus exaggerating their disbelief of politically-inconsistent stories. This is consistent with research showing that people who score high in need for cognition tend to build information rich social networks, but of course this can be problematic when your rich social networks are still operating in an echo chamber.
> However, liberals aren't off the hook, as they are statistically more likely to use investment in the righteousness of their political viewpoints to believe politically-consistent news stories, and their higher level of need for cognition to delegitimize politically-inconsistent news stories. The researchers found that liberals who scored higher in a measure of "collective narcissism"-- which measures a tendency to invest in, and perceive superiority of, your political views--showed exaggerated legitimacy judgments for the politically-consistent (e.g., anti-Trump) fake news stories.
>One possible explanation is that highly logical and rational people are not only more likely to disbelieve politically-inconsistent news stories along tribal lines, but they are also more likely to seek out further disconfirming information, thus exaggerating their disbelief of politically-inconsistent stories. This is consistent with research showing that people who score high in need for cognition tend to build information rich social networks, but of course this can be problematic when your rich social networks are still operating in an echo chamber.
I haven't said anything in conflict with this. My point is that there's a difference between susceptibility to believing fake news and the rate at which one actually believes fake news. Surely you wouldn't deny that "highly logical and rational people" are more likely to have correct political opinions than those who are the opposite?
Rather, the point is that both liberals and conservatives develop a set of beliefs that allows them to make snap judgments. So they're equally susceptible to believing an individual piece of fake news that is targeted at them. But as a matter of fact, the things that liberals believe tend (but are not always) more true than those believed by conservatives. This is not an accident, but can rather be explained by the fact that the very beliefs that make up the biases that liberals hold are often informed by some knowledge of what the facts are.
> Surely you wouldn't deny that "highly logical and rational people" are more likely to have correct political opinions than those who are the opposite?
That's a complicated question, really... I guess I don't know what qualifies as a "correct political opinion". Presumably every person believes they have a "correct political opinion", because if they did not then they would change it. Obviously some number of these people are wrong. So for any given opinion do we just take into account what the majority of "highly logical and rational people" believe? (letting alone I don't know how you would do anything other than vaguely qualify that group). What if there is no majority, because there are seven different opinions on the same topic? That means for any given opinion, the majority believe it is incorrect. Are, then, all political opinions incorrect, or perhaps varying degrees of incorrect? This doesn't seem unreasonable to me, but still rather impossible to quantify.
> But as a matter of fact, the things that liberals believe tend (but are not always) more true than those believed by conservatives.
Is this true? How could we tell? We obviously cannot axiomatically prove it, so we must infer it based on some sample of both groups, presented with "truths". Because we're talking about political opinions, we can assume those truths are limited to this arena. In that case, we must also provide an objective, and potentially a moral framework... which gets fairly messy.
I do agree with your use of the words "likely" and "tend", as they convey a notion of probability, which seems like a reasonable way to approach this. It does seem likely to me that if you presented the set of "highly logical and rational people" with a political issue and collected their opinions, and then did the same with the inverse set, and I had to place a bet on which group's decision I would like more, I would bet on the first group. However, if you took that same set and compared it to, say, "all individuals with knowledge of the field", I would likely bet on the second set, due to their superior context.
You allude to context when you mention snap judgement. I don't necessarily agree, however, that liberals have more knowledge (or perhaps "superior context") than conservatives. In some arenas this is definitely true, but the inverse is also obviously true.
Another interesting line of reasoning: if we accept that there is equal susceptibility between both liberals and conservatives to fake news, it follows that "highly logical and rational people" are not, in fact, more likely to have (more) correct political opinions. Let's say, for example, most "highly logical and rational people" are conservatives, and let us further stipulate that significantly more fake news is pandered to conservatives than liberals. This would indicate that the group predicted to have the more correct political opinions definitely does not. Of course, the inverse is also true, which really means that no correlation between general rationality and correctness of opinions can be drawn. This isn't entirely unsurprising... Ben Carson is one example of this, IMO, someone who is clearly brilliant and rational, but perhaps not entirely correct in all his political opinions. It seems more reasonable to select for experience and domain knowledge over general rationality, which again I think is something you're sort of suggesting when you say that liberals are informed by more knowledge of what the facts are, but to split domain knowledge groups into liberal vs. conservative is probably one of the most ineffective splits possible, and I would be surprised if more "fact knowledge" could be determined as held more by one group than another within a reasonable margin of error.
Sorry this got a bit rambly. Just kind of working through my thoughts.
There's a lot of good stuff here. But there seems to be a thread of "correct political opinions are purely subjective." This is sometimes true I suspect, but there are two cases where it isn't:
Some "opinions" are actually just awareness of or acceptance of objective fact. To stick with the theme: Obama either gave or did not give $x to y charity in year z. Two people can hold differing subjective opinions, but one is objectively wrong.
The other case is when a political "opinion" seems subjective but is really based on a set of objective facts. This is where debate in the current climate goes to hell. I've found that people I disagree with often hold rational positions based on the "facts" they believe to be true. I just don't think any of their facts are accurate. One of us in that case is objectively correct, but resolving that takes an authority we both believe to be canon. That doesn't exist.
Not that I think giving away stuff is a desirable qualification for the presidency, but...
Trump provided his private plane to transport Andrew Ten, who had been refused by commercial airlines due to complicated life-support equipment. Trump provided Jennifer Hudson with a place to live after her family was murdered. Trump has donated his entire presidential salary, to various causes, most recently alcoholism research.
The big donor though is neither Obama nor Trump. It's Mitt Romney.
>The big donor though is neither Obama nor Trump. It's Mitt Romney.
For sure, I don't deny that. I'm not claiming anything about whether being liberal makes you "good" or "bad" w.r.t. charitable donations. I'm claiming that the liberals in this study might know some additional facts that would explain why they're more likely to believe the positive story about fake news in this case. It would certainly be interesting to retest this, but use Obama and Romney as the names instead.
I don't see how this study is valid. Participants were shown "unfavorable stories [that] reported that the target was facing criminal charges". But is it necessarily bias to think that Trump is more likely to commit criminal behaviour before reading the story, and therefore more likely to think such a story is true?
If I see a story that "an unseeded female tennis player beat the top-seeded male tennis player" shouldn't I rate it less likely to be true than a story with the genders reversed?
I agree. This is why I said (below) that the study seems to measure susceptibility to believing fake news, not the rate at which one actually believes fake news. If one walks around with a liberal bias, one will believe some wrong things, and believe some fake news, but one will still be overall have more correct beliefs than someone who walks around with a conservative bias.
(I would not call myself a liberal, for what it's worth - this is a point about the positions typically held by these two groups in US politics.)
Because we actually have data and some scientific understand about gender differences, that question is not very representative of bias. Change gender with race like: "an unseeded white player beat the top-seeded African American tennis player" - Should you rate it less likely to be true if the race was reversed?
Trump vs Obama becomes less a bias test the more factual information a participant has. The test I can only guess assume that participants do not actually have factual information about either president, thus all being equal they should not have anything concrete to base a logic conclusion.
15 comments
[ 3.4 ms ] story [ 47.0 ms ] threadI think confirmation bias is really strong. We're all best off if we take care to ingest a balance of news. (One of my techniques: I read 'RealClearPolitics' daily. It contains outrageous articles from both sides, which I hope gives me a more balanced perspective.)
>For instance, in the first study stories were either favorable or unfavorable about either former President Barack Obama or current President Donald Trump. Favorable stories reported that the target had donated $50 million of their personal fortune to selected charities, while the unfavorable stories reported that the target was facing criminal charges related to voter fraud in the 2016 Presidential election. Participants then made "legitimacy judgments" of each story, rating how much they believed the story to be true, reliable, and trustworthy.
I'm pretty left-wing, so my biases would presumably lead me to believe the favorable story about Barack Obama, and disbelieve the favorable story about Donald Trump. [1] How does that work out in practice? Well, at first glance it seems pretty accurate. Trying my best to abstract away from the context of this study, I'd give about 2% credence to the story about Trump, and maybe 35% to the story about Obama (does Obama even have $50M of personal wealth?). Hypothesis confirmed, right!? Not exactly.
It's true that I'm answering these questions according to my existing beliefs about each politician. But those beliefs aren't simply the result of "bias". I came to those beliefs for reasons. For example, I know that Trump is infamously stingy. [2] Him donating 50 million in personal wealth would be a real shock to me, and for good reason. I know a bit less about Obama, but I don't have a strong reason to think he's stingy, and I do vaguely recall some news indicating that he has made a number of large charitable donations. [3]
So yes, both liberals' and conservatives' judgments are based on their beliefs, and if you put those people in a position where they have to use those beliefs to judge claims in the absence of other evidence, they are liable to believe lies. But one of these groups believes the truth - that Obama is more charitable than Trump - and the other does not. In other words, much of the liberal susceptibility to fake news can be explained simply by pointing at the justifiably higher credence they have in many pro-liberal stories.
[1] I'm relatively well informed, so I'd give ~0% credence to either story about criminal charges. But if one of these stories were true, it would probably be true about the person who's already under investigation for his actions in the 2016 election.
[2] https://www.washingtonpost.com/politics/trump-boasts-of-his-...
[3] https://www.forbes.com/sites/danalexander/2017/02/06/obama-d...
> The researchers measured individual differences in thinking styles and found that regardless of the political party identification, when high need for cognition individuals were presented with fake news stories that were consistent with their ideology, they were even more likely than everyone else to judge the story as legitimate, and when they were faced with fake news story that were inconsistent with their ideology, they were even less likely to consider the news legitimate than everyone else.
One possible explanation is that highly logical and rational people are not only more likely to disbelieve politically-inconsistent news stories along tribal lines, but they are also more likely to seek out further disconfirming information, thus exaggerating their disbelief of politically-inconsistent stories. This is consistent with research showing that people who score high in need for cognition tend to build information rich social networks, but of course this can be problematic when your rich social networks are still operating in an echo chamber.
> However, liberals aren't off the hook, as they are statistically more likely to use investment in the righteousness of their political viewpoints to believe politically-consistent news stories, and their higher level of need for cognition to delegitimize politically-inconsistent news stories. The researchers found that liberals who scored higher in a measure of "collective narcissism"-- which measures a tendency to invest in, and perceive superiority of, your political views--showed exaggerated legitimacy judgments for the politically-consistent (e.g., anti-Trump) fake news stories.
I haven't said anything in conflict with this. My point is that there's a difference between susceptibility to believing fake news and the rate at which one actually believes fake news. Surely you wouldn't deny that "highly logical and rational people" are more likely to have correct political opinions than those who are the opposite?
Rather, the point is that both liberals and conservatives develop a set of beliefs that allows them to make snap judgments. So they're equally susceptible to believing an individual piece of fake news that is targeted at them. But as a matter of fact, the things that liberals believe tend (but are not always) more true than those believed by conservatives. This is not an accident, but can rather be explained by the fact that the very beliefs that make up the biases that liberals hold are often informed by some knowledge of what the facts are.
That's a complicated question, really... I guess I don't know what qualifies as a "correct political opinion". Presumably every person believes they have a "correct political opinion", because if they did not then they would change it. Obviously some number of these people are wrong. So for any given opinion do we just take into account what the majority of "highly logical and rational people" believe? (letting alone I don't know how you would do anything other than vaguely qualify that group). What if there is no majority, because there are seven different opinions on the same topic? That means for any given opinion, the majority believe it is incorrect. Are, then, all political opinions incorrect, or perhaps varying degrees of incorrect? This doesn't seem unreasonable to me, but still rather impossible to quantify.
> But as a matter of fact, the things that liberals believe tend (but are not always) more true than those believed by conservatives.
Is this true? How could we tell? We obviously cannot axiomatically prove it, so we must infer it based on some sample of both groups, presented with "truths". Because we're talking about political opinions, we can assume those truths are limited to this arena. In that case, we must also provide an objective, and potentially a moral framework... which gets fairly messy.
I do agree with your use of the words "likely" and "tend", as they convey a notion of probability, which seems like a reasonable way to approach this. It does seem likely to me that if you presented the set of "highly logical and rational people" with a political issue and collected their opinions, and then did the same with the inverse set, and I had to place a bet on which group's decision I would like more, I would bet on the first group. However, if you took that same set and compared it to, say, "all individuals with knowledge of the field", I would likely bet on the second set, due to their superior context.
You allude to context when you mention snap judgement. I don't necessarily agree, however, that liberals have more knowledge (or perhaps "superior context") than conservatives. In some arenas this is definitely true, but the inverse is also obviously true.
Another interesting line of reasoning: if we accept that there is equal susceptibility between both liberals and conservatives to fake news, it follows that "highly logical and rational people" are not, in fact, more likely to have (more) correct political opinions. Let's say, for example, most "highly logical and rational people" are conservatives, and let us further stipulate that significantly more fake news is pandered to conservatives than liberals. This would indicate that the group predicted to have the more correct political opinions definitely does not. Of course, the inverse is also true, which really means that no correlation between general rationality and correctness of opinions can be drawn. This isn't entirely unsurprising... Ben Carson is one example of this, IMO, someone who is clearly brilliant and rational, but perhaps not entirely correct in all his political opinions. It seems more reasonable to select for experience and domain knowledge over general rationality, which again I think is something you're sort of suggesting when you say that liberals are informed by more knowledge of what the facts are, but to split domain knowledge groups into liberal vs. conservative is probably one of the most ineffective splits possible, and I would be surprised if more "fact knowledge" could be determined as held more by one group than another within a reasonable margin of error.
Sorry this got a bit rambly. Just kind of working through my thoughts.
Some "opinions" are actually just awareness of or acceptance of objective fact. To stick with the theme: Obama either gave or did not give $x to y charity in year z. Two people can hold differing subjective opinions, but one is objectively wrong.
The other case is when a political "opinion" seems subjective but is really based on a set of objective facts. This is where debate in the current climate goes to hell. I've found that people I disagree with often hold rational positions based on the "facts" they believe to be true. I just don't think any of their facts are accurate. One of us in that case is objectively correct, but resolving that takes an authority we both believe to be canon. That doesn't exist.
Trump provided his private plane to transport Andrew Ten, who had been refused by commercial airlines due to complicated life-support equipment. Trump provided Jennifer Hudson with a place to live after her family was murdered. Trump has donated his entire presidential salary, to various causes, most recently alcoholism research.
The big donor though is neither Obama nor Trump. It's Mitt Romney.
For sure, I don't deny that. I'm not claiming anything about whether being liberal makes you "good" or "bad" w.r.t. charitable donations. I'm claiming that the liberals in this study might know some additional facts that would explain why they're more likely to believe the positive story about fake news in this case. It would certainly be interesting to retest this, but use Obama and Romney as the names instead.
Simple application of Goodhart's law.
If I see a story that "an unseeded female tennis player beat the top-seeded male tennis player" shouldn't I rate it less likely to be true than a story with the genders reversed?
(I would not call myself a liberal, for what it's worth - this is a point about the positions typically held by these two groups in US politics.)
Trump vs Obama becomes less a bias test the more factual information a participant has. The test I can only guess assume that participants do not actually have factual information about either president, thus all being equal they should not have anything concrete to base a logic conclusion.