[BebopKid]

I'm doing a survey to see if there is any randomness to the numbers people choose between 1 and 20.

I started with one hypothesis and after asking 43 people, the results do not match my expectations.

I'm hoping that I can improve my sample size by asking for people on the forum to help.

I'll post my original expectations and a summary of the data next week. I don't want to taint the responses.

Thanks in advance.

[Al_U_Card]

17 (Actually my "3rd" respnse, but I like to be cagey....lol)

[mike777]

Humans are very very often fooled by randomness.

My understanding has always been humans are great at pattern recognization but very lousy at understanding probabilities or true randomness.

In other words humans have lousy ability for randomness.

See Taleb's books on the subject.

http://www.thestalwart.com/the_stalwart/20...taleb_book.html

[mikeh]

In terms of non-random human behaviour, some 30 years ago someone wrote a word on a piece of paper, concealed from me, and then asked me to pick a number between 1 and 10...out loud. I did so. I was asked to double it, and then to double that number and so on for several repetitions. I was then asked to name a vegetable, and I said 'carrot'.

The hidden word was.... 'carrot'.

I have since done this to perhaps 20 people over the years (you can tell that I am the life of the party ) and almost (but not all) have said 'carrot'... my last victim was only a few weeks ago.

I have no idea if it works in cultures other than Western Canada, altho the person who showed me this said that he'd read that about 70% of North Americans responded carrot on this test.

It has always struck me as weird. Heck, carrots are neither my favourite nor least favourite vegetable... and I didn't watch Bugs Bunny much as a kid, growing up in the UK

[Gerben42]

I expect a non-random distribution from this test, especially caused by people who are trying to be random by not picking their "first choice".

[helene_t]

Edit: see new updated post.

This post has been edited by helene_t: 2007-July-13, 03:41

[Gerben42]

Maybe we should wait for more answers until the data are significant? Data is significant? What's correct? I'm confused.

[helene_t]

Gerben42, on Jul 12 2007, 03:34 PM, said:

Maybe we should wait for more answers until the data are significant? Data is significant? What's correct? I'm confused.

Good question. According to Pinker's "Words and rules", it used to be "are", but "is" has become so widespread that some consider it correct.

I agree that more data points may alter the conclusion. It is important that voters don't read the results until they have voted, though.

Btw, there seems to be specific kinds of brain damages that cause voters to differ from the norm. In the "vegetable=carrot" example, subjects with William's syndrome would give unusual answers like "Pak Soi" or "Savoy cabbage". Of course which vegetable is most typical depends on where you live. The point is that subjects with William's syndrom tend to provide atypical examples.

[Al_U_Card]

Did you notice that no one picked 1 or 20? Is that because the question was worded to ask for numbers "between"?

[mycroft]

"Data are" is correct. However if you say that, you are either:

1) An English teacher (and likely an old one, having learned Latin) or a relative.
2) A Computer geek (who knows what a datum is, and probably also the difference between data and statistics, and why - not just that - there are three different "averages"). s/Computer/Math - the boundaries are pretty slim anyway.
3) An ex- or current- role-playing (or war-) Gamer, and you still grit your teeth when you hear "I rolled a dice and got 6". ASL players, with their DRM and drm, get to count double for this one.

I am all three. "Data are" still sounds wrong to my ears. Ah well, alea jacta est, I suppose.

Michael.

[hotShot]

Gerben42, on Jul 12 2007, 03:34 PM, said:

Maybe we should wait for more answers until the data are significant? Data is significant? What's correct? I'm confused.

It depends...
If you are referring to Lt. Com. Data use "is".

[Fluffy]

When picking the random number I was looking for something close to 10 (the average) but not 'too close' (standard deviation maybe).

My first impulse was 13, then I found that number to be related with luck and discarded it. I ended up picking 17 influenced by the fact it is prime.

Numbers below 10 seemed too simple for me.

[luke warm]

Fluffy, on Jul 12 2007, 04:35 PM, said:

~~~Numbers below 10 seemed too simple for me.

which probably explains why i picked 8

[Aberlour10]

I picked up "11" (my favorite # on the roulette table)

[Gerben42]

Thunder rolled. He rolled a 6

[helene_t]

I see two paterns:
- We vote for big numbers
- We avoid round and semi-round numbers (5,10,15,20)

Then again, there are so many pattern to test so even a completely random sequence is almost bound to show some kind of patern.

[Al_U_Card]

The only thing that is random is.......our reasons...lol

[GeeGee]

mikeh, on Jul 11 2007, 11:36 PM, said:

In terms of non-random human behaviour, some 30 years ago someone wrote a word on a piece of paper, concealed from me, and then asked me to pick a number between 1 and 10...out loud. I did so. I was asked to double it, and then to double that number and so on for several repetitions. I was then asked to name a vegetable, and I said 'carrot'.

The hidden word was.... 'carrot'.

I have since done this to perhaps 20 people over the years (you can tell that I am the life of the party ) and almost (but not all) have said 'carrot'... my last victim was only a few weeks ago.

I have no idea if it works in cultures other than Western Canada, altho the person who showed me this said that he'd read that about 70% of North Americans responded carrot on this test.

It has always struck me as weird. Heck, carrots are neither my favourite nor least favourite vegetable... and I didn't watch Bugs Bunny much as a kid, growing up in the UK

I got this one as well.....

I was asked 'what's 3 x 2?' '6', 'what's 5 + 1?', '6', 'what's 2 + 2 + 2?', '6'. 'Name a vegetable', 'carrot'. Just automatic, then the piece of paper turns up with the word 'carrot' on it.

So I thought I must try this on my friends:-

'What's 3 x 2'? '6'... and so on, then 'Name a carrot'. .

Geoff

[helene_t]

The log-likelihood for the null hypotheis that the first 48 votes were random, was -43.52327. Compare to this histogram of 10,000 simulations of the poll:

This is a p-value of 0.49%. Clearly significant. So the result of this poll is now outside what could plausibly happen by chance.

[barmar]

I picked 6 because I figured most people would pick prime numbers and I wanted to go against the field.

On a slightly related note...

I'm trying to remember where I read it, but some book talked about the paradox that there are no "uninteresting" numbers. "Interesting" is defined as having some unique properties -- 1 is the multiplicative identity, 2 is the only even prime, 3 is the number of the trinity, etc. Eventually you encounter a number for which there are no known special properties; but it's interesting because it's the lowest such number.