User Tag List

1. I've returned maybe 3 wallets/purses to people. And a couple of cell phones. I don't know if that's higher than average, but I think I might have an eye for it.

One girl was more freaked out that I could find her phone number on the internet using the address on her ID card than she was thankful that I went through the trouble to return it and didn't steal any cash. She was odd. It would have been weirder (and more work) to knock on her door...I think, at least. Another guy hardly even thanked me, just took it and walked away.

The rest were cool though, even giving me money in return.

Ditto on the statistics commentary. (So yeah, the above isn't statistically significant, but it is my experience.)

2. Originally Posted by Jaguar
Tell that to those who make claims about a group of people based on nothing more than their experience with one person.
that sort of person is WAY too stupid to understand the concept of numbers!

3. This looks good, and I'd still have 91 cents left for candy.

4. Originally Posted by whatever
why do they have such a small sample size?

everyone (who knows anything about stats) knows that a sample size that small is completely inconclusive!
This was just one guy doing it himself, it isn't really an 'official' study. Apart from sample size, there is an issue of representative sampling, controlling for wealth, etc. This is by no means a proper study whereby one can extrapolate the population parameter.

Let us assume a representative sample.

The 95% confidence interval for the population mean, based upon the sample, is:

Men
47.5 - 74.9%

Women
76.8 - 95.7%

Young
39.2 - 72.6%

Middle Age
69.1 - 92.8%

Old
74.3 - 100%

Black
36.3 - 76.8%

White
70.2 - 88.3%

*Ranges indicate percent of truthful people in the population*

As you can see, the sample size being small does not restrict the results as much as one might imagine. Yes, the 95% confidence interval will be larger, as you can see by looking at the different ranges for blacks (~40% with a sample of 23 people) and whites (~18% with a sample of 77 people), but it doesn't mean the results don't say anything. However, as I stated before, this is unlikely a representative sample, and there are other factors at play.

5. any less than 1,000 people of each demographic group is still a bullshit underrepresentation

not to mention, where are the other racial and ethnic groups in this survey?

6. It isn't a survey, it is an experiment. He placed a wallet on the ground and watched whether people stole it or returned it. I agree, it would be difficult to get a representative sample with so few people, but I caution the use of 1,000 people as an arbitrary measure of the validity of a sample. The sample size will reflect the precision with which one can make a claim at X level of confidence, but large sample sizes aren't always necessary... I'm serious.

7. but did he try it out in various types of neighborhoods or in a variety of different cities? judging a race based on a mere handful of people seems like a dangerous overgeneralization (or an age group or anything... don't judge me on the basis of how people in New York behave, for an example)

8. I'm guessing the location of where the money was found would influence the decision of the subject.

This is what I would do: Place the belongings out in the middle of asscrack nowhere and plant a phone number where they would think they're contacting the "owner", but they're actually calling me.

That's a true test of principle.

9. Originally Posted by whatever
but did he try it out in various types of neighborhoods or in a variety of different cities? judging a race based on a mere handful of people seems like a dangerous overgeneralization (or an age group or anything... don't judge me on the basis of how people in New York behave, for an example)
Oh no, I wasn't arguing that his sample was representative; it most assuredly isn't. I was just mentioning that sample size isn't as important as one might think. You are correct. He did it in one city (I believe) at several different locations. No other races but visually black and white were observed.

Originally Posted by Ginkgo
I'm guessing the location of where the money was found would influence the decision of the subject.

This is what I would do: Place the belongings out in the middle of asscrack nowhere and plant a phone number where they would think they're contacting the "owner", but they're actually calling me.

That's a true test of principle.
I agree with you, and was thinking the same thing. One of the people who found the wallet, for example, was a young man who worked at the gas station where it was dropped. I'm sure him being on duty at his place of employment influenced his decision.

10. Originally Posted by guesswho
LOL

WHO THE HELL STEALS 2 DOLLARS AND 10 CENTS
Heh. My first thought was, who'd go to the bother of returning two dollars and ten cents? That would be more of a test of laziness than anything else, I'd think.

According to the link though, there was also a (fake) fifty dollar gift certificate in there: Wallet Info

I have found a couple wallets in my university's library when I was a student, and I did return them. I don't know if I'd bother though, if there was nothing in it worth mentioning. *shrug*

Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
Single Sign On provided by vBSSO