The Wisdom of the Crowd Effect: Theory, Experiment, Examples and Explanation

1,694 views Wait, is this logic right? • Jun 06, 2023

Slog Reference: Wisdom of the Crowds

Description

What is the wisdom of the crowd? The wisdom of the crowds is a popular theory that gives you the option to find the right numeric answer to problems. It works by asking the same question to multiple people and averaging their answers to get to the right one. Let's understand what exactly is the wisdom of the crowds, whether is it actually reliable, how you can test it yourself, and how can it help you, with some real-life examples.

Hope you enjoyed FutureIQ by Navin Kabra and Shrikant Joshi. Do hit us up on Twitter:
@ngkabra http://twitter.com/ngkabra
@shrikant https://twitter.com/shrikant

Listen it on the podcast provider of your choice: https://tapthe.link/FutureIQRSS
Watch other episodes of The FutureIQ podcast: https://www.youtube.com/playlist?list=PLAppTB0r5_TaYueZ0adD42Wiw5X-wTE4v

#futureiq

Related Slog Matches

Wisdom of the Crowds

Fuzzy Text

100.00

Transcript

- so follow print which is supposed to be you ask an expert right is right only about 65 percent of the time do two wrongs make a right the surprising answer srikanth is that if you average together enough wrong answers you actually get a pretty good correct answer that is actually true let me we tested it out yeah so this is a concept called uh the wisdom of the crowds and let me start with an example that we did just now okay like literally a few minutes ago so we wanted to check if anybody knows the distance from Pune to Paris yeah so we went around asking in the office what is the distance between Pune and Paris like as the crow

0m 42s flies unless you are a weirdo this is not a distance you are going to know and thankfully this office does not have any weirdos we asked 10 people and we got all kinds of answers hold on this office has been a dose but different kinds of beards okay so uh I mean at the low end we got one four thousand seven hundred one five thousand uh there was a 50 000 also but let's that's an outlier okay then we got rid of two outliers like you know like really outliers can mess up things so you get rid of outliers so two outliers we got rid of the remaining range from 4700 to 13 000 okay

1m 21s which is well within the circumference of the earth so 13 000 is somewhere on the earth and then I took an average actually the geometric mean of all these numbers okay and it came out to be seven two six four seven two six four kilometers was the average guess geometric mean guess of what is the distance between Pune and Paris and the actual answer is seven one two one error only Okay so Nobody Knows the correct answer in fact most people know like a ridiculously wrong answer and yet the average is just two percent off okay yeah before doing this experiment I had no idea if this is going to work or not I just did it because I believed I have

2m 11s heard so much about this concept this is the first time I'm trying it out and even I am surprised by the result okay yeah two persons two percent is pretty close where an error margin it is mind blowing out for something that people really don't know yeah who wouldn't want who would actually keep the number of Pune to Paris in their head if you are one of them comment yeah so um this concept uh was discovered in 1908 right Sir Francis Galton a statistician uh he had gone to a country fair and there there was a contest okay there was a big dead Ox hung there and you have to guess the weight of the ox just by

2m 52s looking at it okay whoever comes closest is going to win a prize a typical contest yeah yeah 800 people participated Sir Francis Galton took down all 800 answers and then he took an average of that in this case he took a median and by the way not a single person got the weight correct okay of course but the median weight was within one percent of the exact answer in fact you might also want to try this Pune to Paris experiment that we did with your friend circular in your class or in your office go ask about 10 15 people and you'll find the same thing that Sir Francis Galton found with the ox uh people might

3m 34s not individually know the distance between Pune to Paris but together geometric mean of their answers after discarding the outliers it will come really really close this finding has been replicated all over the world in many different contexts right not contests many different contradictions um okay the idea is that if you take enough guesses from different people all of them are wrong but they are wrong in different ways and their errors cancel out right because two wrongs do make a right well not two wrongs you need around 10 wrongs to for it to cancel out that doesn't mean you can go around doing wrong things and saying all of my wrong Financial each other that's not how it works please

4m 19s don't do that that's how it was democracy is pretty much a whole bunch of wrong but not one person doing ten different wrong things it is 10 different people doing 10 different wrong things yes one person doing 10 different things sends you in jail yeah don't forget that and then don't come back to us saying but you said wisdom of the crowd City wisdom of the crowds not one the wisdom of the crowd still works right I mean if you are in jail uh nine people are not in jail the average person is still not in jail so that's good so this this actually works in surprising places okay like word like so uh when the challenge of space

4m 59s shuttle blew up uh right within minutes okay the stock price of four companies which had uh components in the shuttle correct their stock prices crashed oh um by the end of the day right three of them had recovered back they were just three percent down from their High the fourth one had not recovered oh okay it was down significantly okay this is what one was a company called Martin theocol okay okay six months later when the presidential commission the committee which did like a fact-finding mission to see what was the real problem and it's a famous one yeah you might remember the O-ring that Feynman talked about yeah you should actually look it up if you

5m 44s don't know about the O-ring and The Challengers that same day before any investigation had happened the crowds had pretty much figured out that modern vehicle is going to get most of the blame right does this give me a little more uh belief little more trust in the stock market because absolutely yeah I mean that's a little more not too much because the stock market has its own other Shenanigans and manipulations and whatnot right the thing is that a stock market it goes every day for thousands of companies right once in a while something weird happens stocks get manipulated once in a while there is a bubble yeah but on most days for most companies for most of

6m 32s History stock market does an excellent job of guessing the correct value for a company and this takes into account everything how they're going to do in the future right as in in uh the early 2000s Amazon had a stock price which just made no sense all experts were saying this is ridiculous right why does a Bookseller have such a huge uh market capitalization look at the world today I mean in those days the wisdom of the crowd said that this is going to be a big company right most of the time it works it's just that our world is just looks at when it doesn't work that's when it there's big news headlines most of the days when it is

7m 15s working there are no headlines that oh stock market did a good job of finding the correct prices of stocks yeah I'd like to see that headline if you know what I mean but true that makes sense that makes sense uh but okay the stock market yes balenger shuttle yes but there are not many occasions in Life or in whatever where we go around asking crowds questions like not everybody is going to have a ready-made crowd in front of them right so when you do have a crowd make the most of it right so for example you end up on Who Wants to Be a Millionaire ask the audience that's why it's a life that makes sense

7m 55s that is why it's a Lifeline in fact an analysis shows so phone a friend which is supposed to be you ask an expert right is right only about 65 of the time whereas ask the audience is right about 91 of the time wow that's wisdom of the crowds wow that that puts the whole Kon banega crorepati uh utility of helpline scale in my head on a very different level yes because I was I always thought that uh if you don't know the answer to something it is always better to ask the expert to ask a friend to look it up and what not depends on the kind of question sometimes an expert huh is the right

8m 33s person to ask the question but there are sometimes especially when the question is vague and fuzzy and all of that wisdom of the crowds can be much better yeah 91 is a great strike rate to be very very honest but then uh okay what if I don't have an audience at all what if I'm thinking only about myself and only for myself whatever single person thinking about whatever funnily enough there has been research on this it's called The Crowd within okay you are not one shrikanti or many srikanths that is true so the way not that I have dissociative identity disorder but I have many shrikans I contain solitudes um the way this can be used will give

9m 12s two ways that have actually been researched right are you serious yes I thought wow if I ask you right uh what is the distance from Pune to Paris in our experiment you had it said 8 000 right correct if I had told you that you know that's the wrong answer can you come up with a better answer I would have actually gone lower correct in fact there is research showing that when you say this to people and they come up with a second answer of course both answers are wrong but the average is better than either answer usually right a different way of doing the same thing is that you give an answer then use a

9m 51s different method to come up with a different answer and then average them a third method is that you come up with an answer today try to forget everything about it three weeks later try to ask yourself the same question and come up with an answer and again research shows that average of the two answers is better than just this one answer right that's what he meant when he said you are multiple SRI cards like within technical term this is a technical term it's called The Crowd within and people do research on the crowd within you'll find lots of articles on the web about the crowd within anybody doing research about the crowd within you have multitudes of

10m 28s shrikans available for research I can give you a crowd within for your research let me know ah fascinating yeah but then uh can I use wisdom of the crowds for everything in life not really okay what are the examples where I can't I mean usually when the answer is numerical right then averaging out the answer gives you a good answer right but many questions in life are not numerical right after finishing after graduating should I do an MBA or should I go for a job it's not a numerical question so wisdom of the crowds isn't really going to help there because I was also thinking there is another saying that says listen to the people and then do

11m 09s what your mind tells you I covered so that is when it's not a numerical thing right another uh place where wisdom of the Crouch doesn't work is that you know there has to be an actual answer then the Crowds Are All spread out around the actual answer if there is a question which is like about the future or something like that then wisdom of the crowds doesn't work you just get a random answer because everyone is just randomly guessing and since I'm assuming this is based on statistical sampling the rules of statistic Supply which is basically that the crowd has to be of all kinds like yeah of course this is a very important point right if all of them come from the

11m 50s same background and they are all making the same kind of error the wisdom of the crowds isn't going to work right because they will have their own inherent biases right if all of them have been influenced by the same Baba right then all of them are going to have a wrong answer in a certain direction I wonder which Baba has an opinion on Pune to Paris but I get your point so uh what you want is the crowd to be diverse right in fact this is also a reason I mean more generally that in companies having teams that are more diverse have been shown again and again and again to have much better performance this is one of the

12m 27s contributory reasons for that right diversity is good and that explains why a lot of companies are now insisting on the entire Dei thing diversity quality and inclusion in there in their workplaces yes you know I'm actually wondering what happens uh now that we have this entire influencer culture where people on social media are telling you what to think and what to feel what stock to buy that's that's a problem right because influencers are influencing people in certain directions and that is reducing the diversity of the opinions in the crowd and making the wisdom of the crowds less uh useful we are not influencers are we are not influencing your opinion we are only telling you what to think all right

13m 10s please get influenced by 10 different influencers yeah diverse get influenced by 10 different influencers if you want names of those 10 influencers we'll tell them to you tell us which kind of influences you want but foreign there is one more important yes there is one more important thing that wisdom of the crowds works if the answers are coming independently ah okay because if you tell me an answer and Akash here's the answer that is going to influence our cash can you tell me why oh we've done an episode on this we've done an episode on this performance we are hardwired to sort of you know conform to what the group is saying so if all the group is sing and

13m 56s answers of a certain type you will without realizing it unconsciously shift your answer towards the group and then the answer will be biased so for wisdom of the crowds to work you have to get the answers independently sort of uh in secret right um thank you this is future IQ