Study finds political bias skews perceptions of verifiable fact

From what I can see the real problem is that most people don’t care about facts anymore. They just want to hear what they think anyway and the political parties are too happy to feed that.

When you talk to left wingers and right wingers about the economy of the last ten years the left guys will tell you it was paradise until 2016 and then everything went to hell. The right wingers will tell you it was misery until 2016 and then the boom started. When you look at stock indices, unemployment or other charts you will see a steady rise over ten years. You won’t even see that the president has changed. But pointing that out is futile. Most people don’t want facts.

I can't believe they thought it was that high. I mean general unemployment is no where near 27% and you'd suspect that with threat of losing visa status, immigrants are more likelier to take a job they dont exactly grok.

> One trick to circumvent this is to report the OPPOSITE conclusion first, and then say, 'actually, it was the opposite' after the listener says 'well, duh.'

Study finds political bias improves perception of verifiable facts.

I don't think I heard a lot of "well, duh".

> In reality, we don't know until we test... We have a lot of stories and biases that we carry all over the place, some are useful, most are shorthand, and many will lead us to wrong conclusions.

And even when we test, we don't really know, because we don't know whether we're talking about the same thing. What does "immigrant" mean? Are you still an immigrant after having lived in the US for 30 years? Is a naturalized citizen still an immigrant, or does he cease to be an immigrant when he becomes a citizen? Do you consider the children of an immigrant to be an immigrant? If so, for how many generations? Are you absolutely sure that everybody you talk to answers these questions the same way you do?

I'm often surprised how vague these political surveys are and how loose they play with language. And I'm even more surprised when they then use the data to make strong calls.

I'm pretty sure if you define the words, you get different results, e.g. ask about "immigrants in the last 10 years", or say "immigrants who have not become citizens that immigrated in the last 80 years". When you only use vague terms, you're not really asking what people believe the numbers to be, you're asking what people understand the term to mean AND what they believe the numbers to be for that understanding.

A current example is asking about racism, where it's no longer just academia where you might encounter something like "systemic effects disadvantaging local minority populations" as the working definition, while the average person might understand it closer to "discrimination based on race". You'll get very different responses based on what definition the person is working with, even when you ask about the same metric, e.g. "how often do you witness racism in your work place?"

I think the most interesting bit about this is how wrong _everyone_ is on the highlighted metrics.

More data here: https://scholar.harvard.edu/files/stantcheva/files/slides_ae...

The extreme _regional_ wrongness about economic mobility is particularly interesting; people in places with higher mobility tend to somewhat downplay it; those in places with lower mobility tend to massively upplay it.