Return rate on 200 wallets dropped in 20 cities

[OldBroad44]

Here's a fun video from Mark Rober describing his experiment on discovering the return rates for 10 wallets dropped in each of 20 cities.

Experiment was 4 years ago. MR recruited social media friends to help with wallet drops. All wallets were identical. Each contained $6 US and $200 Filipino $$ (so it looked like an impressive wad of cash), a fake credit card with no photo, a dog photo, a couple of business cards, and a card that looked like it came with the wallet saying "If found call xxx" This phone number allowed MR to talk with the honest wallet returners and find out more about them. Some cities were small, most large. Half the wallets were dropped in a rich section of each city, half in a poor area. Most of the wallets were dropped on a sidewalk. For each city, two wallets were dropped in a men's room and two in a women's room. It was counted as an honest return when the wallet was returned with all contents .

As judged by return rates from rich vs poor areas and bathroons and conversations from returners, richer and poorer people were equally likely to return wallets intact. Likewise for men vs women. One city, a Disneyland, had to be discarded from the data because a couple of wallets were returned to security, and their identity gave away the show and security said they were not calling about any more wallets. Below are the results in number of the 10 wallets returned for each city, Including Portland and Seattle.

10 Chicago, Salt Lake City
9 Nashville (IA), Hill City (SD), Portland (OR)
8 Parma (ID), Las Vegas
7 San Franscisco, Winnepeg, Washington DC
6 Huntsville (AL), New London (CN)
5 Seattle, LA, Miami, Dallas, Edmonton
4 NYC
3 Detroit
2 --
1 --
Of the 10 Salt Lake City returners, btw, only a couple were particularly religious and only one attended church regularly. Chicago was the big surprise to me. I was happy to see Portland's high ranking. Surprised by Seattle's mediocre ranking. Unsurprised by Detroit and NYC being at the bottom.

My own interpretation: The low amt of money meant that keeping the wallets would make little difference to most finders. But the fake credit card would make presumed worry and hassle for the loser. I think you would see a different result if there was, say, $5,000 or more US $ in the wallets, enough to matter to most people. As it was, I think this was not a test of financial honesty so much as a response to the question "Do you care enough about others so that you are willing to spend ten minutes of your time to save a random stranger some worry and hassle (cancelling credit card)?"

Read More

 

[OldBroad44]

Anybody had a particularly interesting experience returning a wallet?

M favorite: About 6am. I was sitting on the front porch of my house on Western Ave in Corvallis drinking a cup of coffee. A guy zipped by on a bike heading toward downtown about eleven blocks away. I see something fall on the road as he zips by. I go and retrieve the wallet. It has no money or ID, but a serious amount of food stamps. I guess that at this time of morning he is going to the Beanery for coffee and breakfast. Nowhere else would be open. I might be able to intercept him. So I call the Beanery. Turns out, the biker was one of their regulars. He had just gone through the line and discovered his wallet was missing when he hit the cash register . They had let him have his food and coffee anyway, so he was still there eating. So they told him his wallet had been found and passed along my address. About half an hour later he had his wallet and food stamps back. Pretty good given he had no idea where he lost the wallet and there was no ID in it.

 

[Cavedweller]

If the intent is to compare return rates for different cities, it is too small a sample size to draw any conclusions. Results (city ranks) could be entirely due to chance.

Especially trying to stratify sampling with such a small number of wallets into rich vs poor areas, bathroom vs. sidewalk distribution, male vs. female restrooms. It's a very complex experimental design which would need much larger numbers to produce any meaningful results.

 

[OldBroad44]

If the intent is to compare return rates for different cities, it is too small a sample size to draw any conclusions. Results (city ranks) could be entirely due to chance.

Especially trying to stratify sampling with such a small number of wallets into rich vs poor areas, bathroom vs. sidewalk distribution, male vs. female restrooms. It's a very complex experimental design which would need much larger numbers to produce any meaningful results.

I was expecting that challenge. Yes of course it's a small sample size. But even small sample sizes can give useful information. My guess is that if you did the experiment with a thousand wallets per city those scoring 9 or 10 this time would still be in the top half with the much larger sampke and those scoring at bottom would still be in the bottom half. But you certainly couldn't expect the larger sample size to give the exact positions again. Some might, but most would probably shift by a rank or more. Especially those in the middle

Many times a sample size of one gives definative information. If you hear one raven speaking English it means at least some ravens can learn to speak English words. Its possible only one raven on the planet can do it, but that isn't very likely. And it probably means all ravens have the physical structures needed to make the sounds involved in English words.

 

[clanhanson]

Finders keepers

 

[Cavedweller]

But even small sample sizes can give useful information.

A sample such as this may be useful in determining what an adequate sample size is. There are statistical formulas for determining such things, based on the amount of variation in a preliminary sample and the degree of confidence one wishes to have in the final results. But 10 wallets dropped in a city the size of Chicago, Portland or Seattle, and drawing any conclusions about the total populations as a result? No. I'm not buying it.

My guess is that if you did the experiment with a thousand wallets per city those scoring 9 or 10 this time would still be in the top half with the much larger sampke and those scoring at bottom would still be in the bottom half.

And that's just a guess. Totally without foundation. 10 people simply cannot represent a total population of millions. The results are more likely to be due to chance than anything else.

Many times a sample size of one gives definative information. If you hear one raven speaking English it means at least some ravens can learn to speak English words. Its possible only one raven on the planet can do it, but that isn't very likely. And it probably means all ravens have the physical structures needed to make the sounds involved in English words.

First of all, ravens don't speak English, or any other language. Ravens can learn to mimic sounds. But to continue with your analogy, if you heard 1 raven mimicking the sound of an English word, it would not give you any information about the proportion of the raven population capable of doing so. Nor would it allow you to draw any conclusions regarding whether or not a greater proportion of ravens in Portland were capable of doing so than ravens from Chicago. You could design an experiment to find out, but it would require some preliminary sampling, and the application of appropriate statistics.

By the logic you have applied, if one person in Chicago returned a wallet, then all persons in the human population should return wallets. They have the physical structures needed to do so. But we know that they don't. Apples and oranges.

 

[kmk1012]

Twice in my life I have found a wallet containing over $1000 in them. First time I was eight years old and my late Step Father drove me to their address so I could return it. The last time was less than three years ago, returned it to the rightful owner with a little help from our local police, ID didn't have address. Each time no money was missing, I've found several more wallets/purses in my life but those two stand out as a hot like that to my pocketbook would hurt.

 

[OldBroad44]

A sample such as this may be useful in determining what an adequate sample size is. There are statistical formulas for determining such things, based on the amount of variation in a preliminary sample and the degree of confidence one wishes to have in the final results. But 10 wallets dropped in a city the size of Chicago, Portland or Seattle, and drawing any conclusions about the total populations as a result? No. I'm not buying it.

And that's just a guess. Totally without foundation. 10 people simply cannot represent a total population of millions. The results are more likely to be due to chance than anything else.


First of all, ravens don't speak English, or any other language. Ravens can learn to mimic sounds. But to continue with your analogy, if you heard 1 raven mimicking the sound of an English word, it would not give you any information about the proportion of the raven population capable of doing so. Nor would it allow you to draw any conclusions regarding whether or not a greater proportion of ravens in Portland were capable of doing so than ravens from Chicago. You could design an experiment to find out, but it would require some preliminary sampling, and the application of appropriate statistics.

By the logic you have applied, if one person in Chicago returned a wallet, then all persons in the human population should return wallets. They have the physical structures needed to do so. But we know that they don't. Apples and oranges.

There is no clean line between sample sizes large enough to give definitive conclusions and those too small to give any information at all. And many an important behavior has first been reported with a sample size of one observation of one creature. Tool use in nearly every species that uses tools, for example. The beginning grad student is likely to underestimate the importance of small sample sizes. A little later when he does an experiment with mistakes in design and tat end up with insufficient numbers to draw definitive conclusions e tends to reject the entire experiment as ysekess. But usually some results are meaningful, even if not all those he hoped for. So it's a matter of teaching him to get as much as he can out of the imperfect inadequate data has.

A few decades ago it used to be the typical party line among ethologists to discount every sophisticated animal behavior that resembled something sophisticated in humans as something else. Those of us who have watched a lot of animals deliberately deceiving each other turned out to be right. Many animals do deliberately deceive each other. They of course disbelieved Jane Goodall's claim that male chimps organize war parties and kill off males from adjacent tribes, but this was ultimately shown to be normal behavior of chimps in the wild. Just one case of such things reported would be the way it starts. One good case observed is meaningful as the mechanisms of biological evolution dictate against the behavior of an animal being totally unique in most cases. It takes lots more observations to tell how widespread or frequent the behavior is. Below is one of my favorites. Its a first experiment by de Waal with one monkey only to address the question as to whether capuchin monkeys have a sense of fairness. Its been done many times since with many species. But just this first experiment with one monkey was meaningful.

 

[Certaindeaf]

I found a purse in Monterey where the parking lot garage arm comes down. I snagged it, parked, found the DL, studied it and left the purse in my PU.. keeping my peepers peeled while checking out the town.
Sure enough, found her. She was very happy. She insisted on a reward so hey, lunch for two paid for.

 

[Alexx1401]

Twice in my life I have found a wallet containing over $1000 in them. First time I was eight years old and my late Step Father drove me to their address so I could return it. The last time was less than three years ago, returned it to the rightful owner with a little help from our local police, ID didn't have address. Each time no money was missing, I've found several more wallets/purses in my life but those two stand out as a hot like that to my pocketbook would hurt.

That shows REAL integrity that seems to be in short supply these days. I have never had that happen but have often returned critters who were lost because I was sure as hell glad when someone did this for one of mine. Got a package from FedEx one day and along with mine is something for an address a good ways away. Since it was a nice day I took one dog for a walk to the house and handed them their stuff. They did look at me funny as I guess they did not recognize me as being a neighbor.

 

[kmk1012]

That shows REAL integrity that seems to be in short supply these days. I have never had that happen but have often returned critters who were lost because I was sure as hell glad when someone did this for one of mine. Got a package from FedEx one day and along with mine is something for an address a good ways away. Since it was a nice day I took one dog for a walk to the house and handed them their stuff. They did look at me funny as I guess they did not recognize me as being a neighbor.

The best part is that with the second one I found my 11 year old (at the time) was with me. He REALLY wanted a dirt bike and had been working and saving for one. I mentioned that it was enough money that he could get one now, I was never going to keep the money. He simply said "Dad, it's taken me this long to save $500, how long did it take this person to save this $1600?" I knew then that I'm raising him with a great personality. So, we located the proper owner and returned it for nothing more than a thank you. He still remembers doing that and feels great about his decision.

 

[TTSX]

I've found 3 wallets. One was at a college which I turned into security anonymously. The other two I put right back where I found them. The reason is, there was zero money in any of them and I'm not going to be the guy who drives all the way across town just to be accused of stealing an unknown amount of cash. Likely someone else got there first, took the cash and ran.

 

[Andy54Hawken]

Twice I have found a wallet.
Once when I was kid....found it on our street.
I called the owner...and was rewarded with $10.

Once as an adult...
I was at gun show and found a wallet on a table.
Not gonna lie...was tempted as the wallet was full of cash.
However....
I turned it in and over the PA an announcement was made.
Ended up with the owner buying me lunch and paying for the ammo I was going to get.

I would try and find or contact the owner...and make sure what was in the wallet , stayed in the wallet.
Not for any hope of a reward...although that is nice....but just 'cause I would want someone to give me back my lost wallet and items.
Andy

 

[Unicykle]

about 3 1/2 decades ago When I was a kid probably eight or 10 years old I found a wallet in the mall right about this time of year before Christmas and it was loaded with cash. My dad drove us to a police station where they contacted the lady and the police let me wait there and give her back her wallet myself. it was an older lady probably in her 70s or 80s and that was her Christmas money for her grandkids she was beyond ecstatic and crying.She open the wallet and gave me $20 for doing the right thing my dad made me give it back to her because he said doing the right thing is reward enough. I didn't understand that as a kid, but it's something that has shaped who I am.

 

[MTpockets]

A couple of months ago... I was out for a walk when I noticed that a briefcase was laying in my path.
I crossed the street to avoid it.

True Story. Stay safe out there, and enjoy your Holidays.

 

[OldBroad44]

The only other time I found a wallet was also a case were I saw it being lost. A large purse actually. A small foreign woman, from India possibly, From her exotic dress, the wife of a foreign student or visiting scholar at OSU was my guess. She and I were both spooning contents out of bulk bins in a large grocery store. The area was crowded. I noticed the woman because she had put her large purse down on top of an adjacent bin. My situational awareness had gone up because conditions were perfect for someone to snatch that purse. Instead the owner of the purse put her bag of bulk food in her cart and left, leaving her purse behind. It happen so fast she was just suddenly gone and the crowd of people and carts closed around her. I grabbed the purse and headed for the closest cashier. Cashier said she would give purse to their Lost and Found department. I said no, the woman was in this section of the store now. Lets get the purse back to her right now so she didn't have to worry and fret. (And I did not necessarily trust cashier. I wanted to be watching and make sure nothing was removed from purse and see it reclaimed by the right person.) I said open purse now and use ID to call women over intercom and tell her her purse had been found. (I had not open purse myself.) Cashier tried to argue but a supervisor appeared instantly and did exactly what I said. A few minutes later the woman appeared and got her purse back. She had not even realized it was gone when she heard it had been found. She was grateful and amazed. She said in her country nobody would have returned a purse. I said it was no big deal, glad to help. And in America most people would return a wallet or purse.

 

[monkeybutler]

There is no clean line between sample sizes large enough to give definitive conclusions and those too small to give any information at all.

Yeah, but, in this case, the sample is simply too small. There are plenty of applications where small sample sizes can at least be solidly directional. This is clearly not one of them.

 

[OldBroad44]

A couple of months ago... I was out for a walk when I noticed that a briefcase was laying in my path.
I crossed the street to avoid it.

True Story. Stay safe out there, and enjoy your Holidays.

Yeah. I think these days my first reaction to an abandoned briefcase or backpack would indeed be to get the he!! out of blast range. Second would be to ask a nearby store keeper to call cops and get out of there too. That is, report it but essentially anonymously. Since I would have nothing useful to contribute other than existence and location of object, no reason to get involved with police and lots of reasons not to.