|
On July 31 2009 05:35 Djabanete wrote:Show nested quote +On July 31 2009 05:23 Dazed_Spy wrote:On July 31 2009 05:15 Djabanete wrote:On July 31 2009 04:44 travis wrote:
This was easy because I am not stupid and put gigantic ranges for everything that I didn't know. That means you are hugely underconfident. You might want to consider some self-help books, this can be a real impediment to leading a normal life. Best of luck. Also, read the first page? lol how is this even an issue of confidence? Its an issue of being a blind little twat or seeing reality for what it is. These questions arent taught anywhere. They arent going to be found anywhere in daily life. Almost every single person here should be able to immediately go "hm never bothered to read into this EVER, therefore my chance of answering it correctly is statistically non existent". If you think you can get the answer right despite knowing nothing on the subject, thats not being overly confident, its called a self induced delusion. I'm confident in my ability to reason, thus I answered as Travis did, in a logical rational manner. Show nested quote +On July 31 2009 00:37 Strayline wrote:On July 31 2009 00:20 Carnivorous Sheep wrote:
Uhm, you didn't actually limit us. So I could do 1 - 1000000 for all my intervals...? You need to think about if you would take the bet both ways, if you put 0 - a google plex you wouldn't actually take the bet against your range even if you got 9 dollars back if you put in one. Like I said, read the first page. The test isn't about setting bounds that you know include the correct answer, it's about setting bounds that you're 90% sure include the correct answer. In other words, you should be willing to take a bet either that you're right or that you're wrong (with the odds adjusted appropriately). It's harder than one might think. Unless someone knows mathematical probability such as game theory, its not so much adjusting for odds appropriately as it what I said, admitting weather you legitimately know the answer or not and responding in kind...
|
South Africa4316 Posts
On July 31 2009 05:23 Dazed_Spy wrote:Show nested quote +On July 31 2009 05:15 Djabanete wrote:On July 31 2009 04:44 travis wrote:
This was easy because I am not stupid and put gigantic ranges for everything that I didn't know. That means you are hugely underconfident. You might want to consider some self-help books, this can be a real impediment to leading a normal life. Best of luck. Also, read the first page? lol how is this even an issue of confidence? Its an issue of being a blind little twat or seeing reality for what it is. These questions arent taught anywhere. They arent going to be found anywhere in daily life. Almost every single person here should be able to immediately go "hm never bothered to read into this EVER, therefore my chance of answering it correctly is statistically non existent". If you think you can get the answer right despite knowing nothing on the subject, thats not being overly confident, its called a self induced delusion. I'm confident in my ability to reason, thus I answered as Travis did, in a logical rational manner.
So the logical rational manner of being 90% sure is choosing an absurd number? How old was MLK when he died... if you're only 90% sure that he was between 1 and 200 years old, then you are underconfident.
A more realistic example is the diameter of the moon. Once again, you could say 1-100,000km, but if you're only 90% sure of that, then you're wildly underconfident. Thinking about it logically, the moon is smaller than the earth, so it can't have a diameter larger than that of the earth. The furthest air plane flight I think realistic is about 14,000km, but it's probably about 18,000km-24,000km from pole to pole. 21,000km is thus half the circumference of the Earth, which means the diameter would be 21,000km/3.14 = 7,000km. So just thinking about it logically, I'm basically 95%+ sure that moon's diameter is not bigger than 7000km. So I'll decrease it to say 4000km (still too big in my opinion, if you think about illustrations of the moon and Earth, but I want to be 90% sure), and the minimum I'll just have to guess. Is it possible that the moon's diameter is 10 times smaller than the Earth's? It's possible but not very likely... Still, I'd say 500km. So then I'd be 90% confident that the moon's diameter is between 500km and 4000km. So there you go. I have absolutely no idea what the moon's diameter is, but I can make a judgement in which I'm 90% confident. That would be a logical way of doing it.
Part of the point of the exercise is that we're not always completely rational, and that we are more confident in our own quick judgements than we should be. For instance Mozart. I knew Mozart was a romantic composer, and I knew baroque was in the 16th and 17th centuries, so romantic era had to be the 18th century. But instead of taking something I'd be 90% confident with, I did a few quick calculations and then said 1710-1750, when he was actually born in 1756. There's no way I should be 90% confident in that answer. I could have gotten the centuries of the baroque wrong, or gotten wrong that he was at the start of the 18th century, instead of the end. I could have been wrong that he wasn't at the very start of the 18th century. Instead of making a good judgement, I was too confident in my knowledge and made my range too small. That's the point of the exercise. That when people make estimates that they are 90% confident in (not 100%) their accuracy tends to be much lower. In fact, I read a study about three days ago that showed exactly how accurate people are. When people are 100% confident, they tend to be about 85% accurate, while people who are 90% confident are only 60% accurate. The point is that we are too confident in our own jdugements, and even though this test has it's faults, it makes that point well.
|
On July 31 2009 05:43 Dazed_Spy wrote:Show nested quote +On July 31 2009 05:35 Djabanete wrote:On July 31 2009 05:23 Dazed_Spy wrote:On July 31 2009 05:15 Djabanete wrote:On July 31 2009 04:44 travis wrote:
This was easy because I am not stupid and put gigantic ranges for everything that I didn't know. That means you are hugely underconfident. You might want to consider some self-help books, this can be a real impediment to leading a normal life. Best of luck. Also, read the first page? lol how is this even an issue of confidence? Its an issue of being a blind little twat or seeing reality for what it is. These questions arent taught anywhere. They arent going to be found anywhere in daily life. Almost every single person here should be able to immediately go "hm never bothered to read into this EVER, therefore my chance of answering it correctly is statistically non existent". If you think you can get the answer right despite knowing nothing on the subject, thats not being overly confident, its called a self induced delusion. I'm confident in my ability to reason, thus I answered as Travis did, in a logical rational manner. On July 31 2009 00:37 Strayline wrote:On July 31 2009 00:20 Carnivorous Sheep wrote:
Uhm, you didn't actually limit us. So I could do 1 - 1000000 for all my intervals...? You need to think about if you would take the bet both ways, if you put 0 - a google plex you wouldn't actually take the bet against your range even if you got 9 dollars back if you put in one. Like I said, read the first page. The test isn't about setting bounds that you know include the correct answer, it's about setting bounds that you're 90% sure include the correct answer. In other words, you should be willing to take a bet either that you're right or that you're wrong (with the odds adjusted appropriately). It's harder than one might think. Unless someone knows mathematical probability such as game theory, its not so much adjusting for odds appropriately as it what I said, admitting weather you legitimately know the answer or not and responding in kind...
Right, you first figure out if you know the answer (and the test is designed in such a way that people generally won't), and then you try to intuitively throw out a confidence interval. If the confidence interval is absurdly wide, you're doing it wrong, because the very definition implies that you should be 10% sure that the correct answer is not within the bounds you provide. In other words, you're not supposed to get every question "right". You're supposed to gamble. The objective of the test is not to figure out whether you're right or not, but whether you're good at gambling on your knowledge; and unless you're perfect at gambling on your knowledge (that is, over a large number of samples you have 90% of answers within your confidence intervals), you will either trust your knowledge too much (overconfidence) or not enough (underconfidence).
I think the test would be better if people had to provide 50% confidence intervals. Then people wouldn't get so hung up on getting the "right" answer and it would be easier to grasp the point of the test. It comes across as a trivia test but it actually isn't.
edit: massive ninja attack
|
8/10
1) 30-45 years
2) 2485-4971 miles
3) 4-18 countries
4) 1-100 books
5) 1243-2485 miles
6) 11000-250000 pounds
7) 1730-1800
8) 90-500
9) 3100-6000 miles
10) 36000-59000 feet.
Got the airplane and elephant wrong. But I tried to think logically and state ranges that i was reasonably certain of. In the cases I had no idea whatsoever (old testament, elephant, and somewhat the airplane, i tried to widen the ranges).
|
On July 31 2009 06:17 LaLuSh wrote:
5) 1243-2485 miles
Haha, where did this one come from? I'm curious since the other answers you gave were all nice round numbers 
edit: Maximum estimate = (minimum estimate x2) - 1? I'm so confused.
|
My (and Chill's for the record) objection is that some of the answers are, as he says, "shocking."
If I said to you "How many of each type of animal did Moses take on the Arc?"
and you said "Two"
and I said "Are you 99% confident?"
you would say "of course!"
but I would say "haha you're way to confident, it was Noah who took who took animals on the Arc!"
In my opinion, in the above example you were being perfectly logical about how certain you were. It was another psychological factor about how human beings follow stories they hear that made you wrong. I tricked you and I don't think the "You should always allow for a X% chance that I'm tricking you!" is a fair argument. The situation is not a game of cards, you are administering a test and should therefore be somewhat "fair" about the basic setup.
|
On July 31 2009 00:58 Chill wrote:
This is stupid because the answers in themselves are shocking. Sure you should account for that in your interval, but the test is pitted against you. If they were questions with reasonable answers in everyday things this would hold a lot more weight for me.
I didn't find any of the answers shocking
|
On July 31 2009 06:20 Strayline wrote:
My (and Chill's for the record) objection is that some of the answers are, as he says, "shocking."
If I said to you "How many of each type of animal did Moses take on the Arc?"
and you said "Two"
and I said "Are you 99% confident?"
you would say "of course!"
but I would say "haha you're way to confident, it was Noah who took who took animals on the Arc!"
In my opinion, in the above example you were being perfectly logical about how certain you were. It was another psychological factor about how human beings follow stories they hear that made you wrong. I tricked you and I don't think the "You should always allow for a X% chance that I'm tricking you!" is a fair argument. The situation is not a game of cards, you are administering a test and should therefore be somewhat "fair" about the basic setup.
I'm confused, which question in the OP is a trick question?
Your question is one in which everybody thinks they know the answer, and would know the answer, except that you're not asking what they think. None of the questions in the OP is trying to trick you, they're just estimation questions chosen so that people won't know the answer and are forced to guess. All of the questions are on a terrestrial scale and the answers are distances, times, dates, and two-digit numbers --- it's not like asking about weird physical quantities or things nobody's ever heard of.
What questions would you use?
|
I had 4/10 and under-guessed most of the ones I had wrong. Interesting test, although I wouldn't say that this very basic test proves much.
By the way, I was slightly annoyed that you asked for non-metric estimates, as that meant I had to convert my estimates to non-metric. Since I'm too lazy to get a calculator, this may have further influenced my amount of wrong estimates.
|
I'm amazed at the amount of people taking this test seriously O.O
Oh less than 1% of the people cannot calculate precisely off the top of their head what 90% off something I don't know exactly is. That is sooo shocking. I definitely proved something new and important there. ¬¬
|
If you're interested in a book discussing this from an economics / public choice perspective click
here. It deals with the "weird question" issue too in some ways. Caplan uses US survey data to show the average person is wrong on a LOT of clear-cut (you would think) issues. Errors don't "balance out" by the "miracle of aggregation" in large enough populations. This, he argues, is a problem for democracy.
|
On July 31 2009 00:13 azndsh wrote:
It is a well-known fact in that people tend to think that they know more than they really do.
Is it a well known fact, or do we tend to THINK it's a well known fact? eh???
|
Edit: damn browser.
Overconfidence is an established cognitive bias, demonstrated in a number of settings.
|
South Africa4316 Posts
On July 31 2009 06:20 Strayline wrote:
My (and Chill's for the record) objection is that some of the answers are, as he says, "shocking."
If I said to you "How many of each type of animal did Moses take on the Arc?"
and you said "Two"
and I said "Are you 99% confident?"
you would say "of course!"
but I would say "haha you're way to confident, it was Noah who took who took animals on the Arc!"
In my opinion, in the above example you were being perfectly logical about how certain you were. It was another psychological factor about how human beings follow stories they hear that made you wrong. I tricked you and I don't think the "You should always allow for a X% chance that I'm tricking you!" is a fair argument. The situation is not a game of cards, you are administering a test and should therefore be somewhat "fair" about the basic setup.
Which questions do you think are out to trick you though? The only two I can see that are surprising is the deepest point in the ocean, and the weight of an empty boeing.
1. He died ver young, but for you to be 90% sure you should include for that possibility.
2. Just a plain question. Africa is about 8000km from one end to the other, so with all the squiggles included it could easily be 4000km.
3. Another plain answer. We all know of at least four countries, but you should expect there to be quite a few small countries. 4-25 would be a safe bet here I think.
4. Once again, can't see how this can be deceitfully tricky. For some people it's common knowledge.
5. Once again, not very surprising. Compared with the earth, it just about fits.
6. This one I think is a bit surprising. I was sure it was heavy, so I guessed 50k-250k. Still, I could have gone even higher on my margins to be surer.
7. Nothing surprising about this one.
8. Maybe surprising if you assume that humans have the longest gestation period. Without knowing that humans have the longest gestation period, I can't see how you can assume that though. I'd expect it to be longer than humans, so 180-900 days would probably be my guess.
9. Straight-forward I think. Shorter than I expected, but it should be in your range.
10. Surprising to me. In fact, I don't get how it works... If the earth's diameter is roughly 12,000km (wiki), then surely you can't get a place that's deeper than 6,000km? Supposedly this place is 11,000km deep, so I don't really get it.
In the end, I really think there's only one question that's a "trick" question, that even if you think about it logically, you can get it wrong. The others you have varying degrees of knowledge on, and you should adjust your ranges to accomodate that.
On July 31 2009 06:36 VIB wrote:
I'm amazed at the amount of people taking this test seriously O.O
Oh less than 1% of the people cannot calculate precisely off the top of their head what 90% off something I don't know exactly is. That is sooo shocking. I definitely proved something new and important there. ¬¬
This test is actually a proper test, designed to show something with minimum difficulty. And your second statement is stupid. If you didn't look at my profile, and I asked you to guess my age, but provide a minimum and maximum age, would you be able to come up with a range that you'd be 90% sure about? So, with very little information, you can make a judgement that you'd have 90% confidence in.
This test tests the same thing. When people have little information, can they provide ranges which they are 90% (not 100%) sure are accurate. It's not about how many answers you can get right or wrong, or how much you know. It's about having a justified sense of confidence in your answers. So yes, it is surprising. If I told thousands of people to make 10 predictions that the are 90% sure will come true, and the average is to get 3 predictions right, then that would prove something new and important. Not that this test is new, if I'm not mistaken it was designed somewhere between 1972-1984 (I'm about 70% sure of that)
|
On July 31 2009 06:36 VIB wrote:
I'm amazed at the amount of people taking this test seriously O.O
Oh less than 1% of the people cannot calculate precisely off the top of their head what 90% off something I don't know exactly is. That is sooo shocking. I definitely proved something new and important there. ¬¬
I'm amazed at the amount of people maligning the test before they bother to figure out what it's testing and what it does or doesn't prove. Hint: it's not an intelligence/trivia test.
Daigomi keeps ninja'ing the crap out of me. (He even writes longer posts than me.)
|
On July 31 2009 06:02 Daigomi wrote:Show nested quote +On July 31 2009 05:23 Dazed_Spy wrote:On July 31 2009 05:15 Djabanete wrote:On July 31 2009 04:44 travis wrote:
This was easy because I am not stupid and put gigantic ranges for everything that I didn't know. That means you are hugely underconfident. You might want to consider some self-help books, this can be a real impediment to leading a normal life. Best of luck. Also, read the first page? lol how is this even an issue of confidence? Its an issue of being a blind little twat or seeing reality for what it is. These questions arent taught anywhere. They arent going to be found anywhere in daily life. Almost every single person here should be able to immediately go "hm never bothered to read into this EVER, therefore my chance of answering it correctly is statistically non existent". If you think you can get the answer right despite knowing nothing on the subject, thats not being overly confident, its called a self induced delusion. I'm confident in my ability to reason, thus I answered as Travis did, in a logical rational manner. So the logical rational manner of being 90% sure is choosing an absurd number? How old was MLK when he died... if you're only 90% sure that he was between 1 and 200 years old, then you are underconfident. I wasn't aware that we were allowed to pick a range of his age, I thought we had to pick a specific answer and give a rate of our confidence for it. I.E: My answers [on all] were im 90% sure that I dont know what the answer was. I was right everytime!
|
South Africa4316 Posts
On July 31 2009 06:50 Dazed_Spy wrote:Show nested quote +On July 31 2009 06:02 Daigomi wrote:On July 31 2009 05:23 Dazed_Spy wrote:On July 31 2009 05:15 Djabanete wrote:On July 31 2009 04:44 travis wrote:
This was easy because I am not stupid and put gigantic ranges for everything that I didn't know. That means you are hugely underconfident. You might want to consider some self-help books, this can be a real impediment to leading a normal life. Best of luck. Also, read the first page? lol how is this even an issue of confidence? Its an issue of being a blind little twat or seeing reality for what it is. These questions arent taught anywhere. They arent going to be found anywhere in daily life. Almost every single person here should be able to immediately go "hm never bothered to read into this EVER, therefore my chance of answering it correctly is statistically non existent". If you think you can get the answer right despite knowing nothing on the subject, thats not being overly confident, its called a self induced delusion. I'm confident in my ability to reason, thus I answered as Travis did, in a logical rational manner. So the logical rational manner of being 90% sure is choosing an absurd number? How old was MLK when he died... if you're only 90% sure that he was between 1 and 200 years old, then you are underconfident. I wasn't aware that we were allowed to pick a range of his age, I thought we had to pick a specific answer and give a rate of our confidence for it. I.E: My answers [on all] were im 90% sure that I dont know what the answer was. I was right everytime!
Haha
On July 31 2009 06:48 Djabanete wrote:Show nested quote +On July 31 2009 06:36 VIB wrote:
I'm amazed at the amount of people taking this test seriously O.O
Oh less than 1% of the people cannot calculate precisely off the top of their head what 90% off something I don't know exactly is. That is sooo shocking. I definitely proved something new and important there. ¬¬ Daigomi keeps ninja'ing the crap out of me. (He even writes longer posts than me.)
It's a bad habit 
|
On July 31 2009 06:48 Daigomi wrote:
10. Surprising to me. In fact, I don't get how it works... If the earth's diameter is roughly 12,000km (wiki), then surely you can't get a place that's deeper than 6,000km? Supposedly this place is 11,000km deep, so I don't really get it.
The answer is in feet :p
|
On July 31 2009 06:50 Dazed_Spy wrote:Show nested quote +On July 31 2009 06:02 Daigomi wrote:On July 31 2009 05:23 Dazed_Spy wrote:On July 31 2009 05:15 Djabanete wrote:On July 31 2009 04:44 travis wrote:
This was easy because I am not stupid and put gigantic ranges for everything that I didn't know. That means you are hugely underconfident. You might want to consider some self-help books, this can be a real impediment to leading a normal life. Best of luck. Also, read the first page? lol how is this even an issue of confidence? Its an issue of being a blind little twat or seeing reality for what it is. These questions arent taught anywhere. They arent going to be found anywhere in daily life. Almost every single person here should be able to immediately go "hm never bothered to read into this EVER, therefore my chance of answering it correctly is statistically non existent". If you think you can get the answer right despite knowing nothing on the subject, thats not being overly confident, its called a self induced delusion. I'm confident in my ability to reason, thus I answered as Travis did, in a logical rational manner. So the logical rational manner of being 90% sure is choosing an absurd number? How old was MLK when he died... if you're only 90% sure that he was between 1 and 200 years old, then you are underconfident. I wasn't aware that we were allowed to pick a range of his age, I thought we had to pick a specific answer and give a rate of our confidence for it. I.E: My answers [on all] were im 90% sure that I dont know what the answer was. I was right everytime!
You're clearly overconfident in your instruction-following abilities.
|
South Africa4316 Posts
On July 31 2009 07:00 SonuvBob wrote:Show nested quote +On July 31 2009 06:48 Daigomi wrote:
10. Surprising to me. In fact, I don't get how it works... If the earth's diameter is roughly 12,000km (wiki), then surely you can't get a place that's deeper than 6,000km? Supposedly this place is 11,000km deep, so I don't really get it. The answer is in feet :p
Ahaha, oops I noticed that, but my conversions were just terrible. Still, then it's surprisingly not deep :p
|
|
|
|
|
|
EPT
