The Political Bias Test

In My Humble Opinion
41

Yeah, I don’t see how this thing could work even if it were entirely factual. How is asking you your political persuasion tell you about your bias? And how much is it bias if you’re just guessing, because you don’t know the answer?

The only way I could see a bias test working would be to present you with the actual information first, then ask you questions about it. That way you’d see if your bias was interfering. As such, I will take the test with Google at the ready to check for any question I don’t know the answer to.

42

FWIW, I scored a 0% on bias and was 72% correct on my political knowledge.

43

Wait a second. You guys are incorrect about the GMO question. Here it is:

“Are genetically modified foods (GMO’s) unsafe for humans due to being genetically modified?”

It says nothing about consuming the foods. It also says nothing about scientific consensus. It’s just a simple question.

I know the answer they’ll want is Yes, but I would definitely have picked “we don’t know yet.” If they’d given better choices, I’d go with something like “moderate to high assurance they are safe.”

Also, the following is not a fact based question:
“Rate whether you think this explanation is a major reason the economy is not doing better than it is, a minor reason, or not a reason at all.” Since it was “foreign aid spending is too high,” and my understanding is that it is relatively low compared to our GDP, I went with barely a reason at all.

Same here:
“The average US citizen would be better off if a larger number of highly educated foreign workers were legally allowed to immigrate to the US each year.”

And:
“How does the US compare with Sweden and Denmark according to [social mobility]? The US has…” We apparently don’t know, because we keep measuring relative vs. actual mobility. Though I assume the answer they want is that it is higher.

Also, this one appears to be bad because the answer is not an option:
“Does permitting adults without criminal records or histories of mental illness to carry concealed handguns increase or decrease violent crime?” There is apparently no statistical link. I thus went with “we don’t know as of yet,” but that’s not the same thing.

Yeah: a lot of the stuff was false. I looked up all my answers (since I didn’t know the answer to any of them, because literally none of them are important to my political beliefs), so I know what the answers were. It claims a political bias of 40%, but all of my wrong answers were just based on taking the results of a Google search, save 1 where I misread the question.

It’s also very clear by the time the thing is over that it’s all just an ad.

Still, since I said I’d give answers, I shall: 13/18 and 40.2% bias.

44

As for the ones I got “wrong”:
The GMO one, as I misread and thought it said safe.

8-biggest companies as of 2014:
I found a list online posted in 2014 and counted. 5/8 were US. It claims 7.

social mobility: see above. As I said, it’s not really known. There were cites going both directions, and a review saying we just don’t know.

US richer than EU: I say “most” fits better than “almost all,” but that’s debatable.

Death penalty: the studies I found say it does not do anything, not that we don’t know.

Just for giggles, I went back and changed my answers for the one I misread, to see how it changes my bias score: It becomes 22%. If it’s that variable, this is not a very good test.

45

0% biased, but I don’t know much. Also I think that test was pretty biased in itself.

46

0% Bias and 78% correct. I think I did the same thing with the answers I got wrong-- tried to guess what they were trying to trick us with.

There was one questions where we had to choose between “most” and “nearly all”. I chose most because I figured it included “nearly all”, but I was marked incorrect.

47

0 bias and 77% correct. I too couldn’t tell the difference between “most” and “nearly all.” I figured it was about 90%. Is that most or nearly all? Am I wrong?

48

12/18, no bias.

But IMHO this is an absolutely terrible test. How many of the questions should reasonably well-informed people know the answers to, off the top of their head? Maybe a third of them?

They needed questions where persons with biases one way or the other would think they knew what the answer was, rather than thinking, “Why the hell would I know whether the share of power from renewables increased by 1% or 2% between 2009 and 2013? And how is making the wrong guess between the two going to demonstrate bias on my part?”

49

:slight_smile: I think I had to read that question 3 times before I could convince myself they were serious.

50

Two fundamental false assumptions upon which this is predicated:

  1. People can and should be “unbiased,” or magically “objective.” In fact different people are in different social/economic situations, and therefore naturally should have different interests (i.e., biases)

  2. Bias is necessarily the cause of misinformation (or lack of information). In fact, sometimes people just don’t know all the facts, regardless of their bias. No one has the time to be 100% informed about everything.

51

Yep. My general tendency, when answering a question where I don’t know the answer and am wild-guessing, is to go for the middle amount. When asked whether America is the most popular destination for immigrants, I was like, “I think so, but I dunno, maybe there’s like a big boom going on in Australia or something, or maybe hella lot of people are going from Vietnam to China,” so I chose the middle option.

It was wrong, and the quiz decided that showed my bias against strict immigration control or something like that.

I’m not impressed.

52

But if you think about it, asking you question that you knew the answer to wouldn’t be able to suss out bias. The idea is to purposely ask questions you don’t know the answer and have to guess. If you’re biased, you’re going to guess the wrong answer that fits your political leanings. But I think a lot of us were guessing in such a way that we were trying to anticipate the trick, often guessing against what our politically leanings might be. That defeats the purpose. At any rate, the best way to do this would be to give you a very limited time answer the questions. The more you think about them, the less likely they are going to expose your bias.

53

To add to that… the higher your correct score is, the more error there is in measuring your bias. If you got all the questions correct because you knew all the answers, there would be no way to measure your bias.

54

My bias score was 63% and I disagree with it. I don’t factually KNOW the answers to all of the questions asked. There was no “I don’t know” button or “how confident are you in this answer?” slider. I have NO idea what % of our GDP is spent on foreign aid. I’d give you a ballpark maybe, but that’s not good when the answers were literally 1 percentage point apart.

What makes me personally NOT very biased is that I look stuff up I don’t know. I ask questions. So yeah, my guesses show what I’m GUESSING but it doesn’t show that if I were in a position of power, I wouldn’t use my best guesses to make policy. I’d look up facts and figures. My actions are not an accurate reflection of my best guess on info I don’t know.

55

You’re not supposed to know the answers. They chose those questions because most people wouldn’t know. But none of what you wrote doesn’t mean you’re not biased. It just means that you may take measures to minimize the effect of that bias.

Which brings us to the common refrain: The people who are dangerous are the ones who don’t know what they don’t know.

56

Or, as I did, guess the middle answer.

57

But when the difference between the offered answers is very small (as it was for a number of the answers), it isn’t a good measure of bias.
The test is an interesting idea, poorly implemented.

58

And I did something similar. Like I said, above, they should give you a very limited time to answer so you don’t overthink it. That’s how you get at your bias anyway-- what does your get tell you.

59

Here’s the problem with the test as written. Imagine a question where my gut tells me “big.” If presented with the options 10%, 50%, and 75%, my bias will point me to 75% because that’s big- especially as compared to the other options.
But when the choices are 96%, 97%, and 98% - all of those are big. They’re all right in the same range; they round to the same place. They all fit comfortably in my definition of “big” so my bias toward the correct answer doesn’t dictate which one of them is the most likely - my understanding of how tests are written does.

60

The problem with picking the middle answer is that sometimes, things are true.

Choosing a point midway between the correct answer and a wrong answer makes you… drumroll please… wrong. Reality doesn’t award points for even-handedness.

You can see this in anti-vaccination groups: They scream to high heavens that they’re not anti-vaccination, they’re pro-safe-vaccination, which is a deliberate lie about how safe vaccines are in and of itself, and then they propose a “middle path” through the vaccination schedule which, guess what, still results in a lot of unvaccinated kids running around. They’re even-handed, unbiased, and utterly, totally wrong.