Peter Donnelly: How stats fool juries
wie Sie tun, machen, Film, Beispiel
22m 7sLänge
http://www.ted.com Oxford mathematician Peter Donnelly reveals the common mistakes humans make in interpreting statistics -- and the devastating impact these errors can have on the outcome of criminal trials. TEDTalks is a daily video podcast of the best talks and performances from the TED Conference, where the world's leading thinkers and doers are invited to give the talk of their lives in 18 minutes -- including speakers such as Jill Bolte Taylor, Sir Ken Robinson, Hans Rosling, Al Gore and Arthur Benjamin. TED stands for Technology, Entertainment, and Design, and TEDTalks cover these topics as well as science, business, politics and the arts. Watch the Top 10 TEDTalks on TED.com, at http://www.ted.com/index.php/talks/top10
Kommentare
-
this has honestly got to be one of the most boring videos ...like ever...I kept thinking he purposfully was this boring until the end so he could say that the way to fool a jury is to keep jumping from convo to convo until they didn't even know the original topic and they simply agree with you Just because you've talked so Damn much...but nope, I was wrong...this was honestly boring.
-
5:32 Heads Tails question/thought experiment begins
-
Yes
-
This was my statistics homework. Good talk.
-
This was my statistics homework. Good talk.
-
This is literally my math homework (to watch this video)
-
The first example is a word-game for a particular context: genetic dna strings, intertwined results. In real life, while we should look more often at how previous results influence future ones, it is almost always more useful to group events into distinct equally cardinal groups [HTT]vs[HTH] instead of [HTTHTH]vs[HTT]. The other example with the 99% faithful test results could very easily be summarized as: 1% error is a large number over a large population - 1 in 100 vs 10,000 in 1,000,000 and then factor in the rarity of the condition to begin with. I have a very rare condition, which makes it difficult for doctors to believe, but I point out that although rare, someone must have this condition, for it to be also true that so many others not have it.
-
This peter donnelly chap is a smart fella.
-
Another TED talk which shows useful math that should be taught in schools. Thank goodness I became an engineer. Otherwise my math education would have been predominantly pointless.
-
You may say That this is common sense. As he points out very well that is not common sense. Common sense works only if you have common values. If I said work all the time and that senior executives one very simple answers and they want the creation of the universe explained in three bullets and then we wonder why they cannot understand anything.What happened with healthcare.gov is that everyone was too busy debating the law without taking into account that someone in Oklahoma would have to click the button and it would have to work. As you know, people inside the Beltway are full of ideas but they don't know how to buy paper for the copier.Watching these videos is a way to extend knowledge, assuming that you are able to filter through some of the nonsense that is actually here that's why it is always good to have contradictory sources of information because the truth is usually somewhere in between. I subscribe to these talks because I like the ideas that some of these people are promoting and some of them are quite good and very relevant in a way of fighting personal ignorance.
-
This needs to be in the "Everything you know is wrong" playlist.
-
Great TED talk on stats.
-
Fully agreed. I think that shows two things : First, people do not have any basic knowledge in statistics, which is a pity, and then, a few people are real experts with statistics, finally the gap between them is just.. huge.
-
Don't scroll down, it's a warzone...
-
Jesus christ, what a boring talk. No wonder people run away when they hear the word statistician.
-
Genocide does not require violence per the UN definition of genocide, educate yourself. Anti-whites say White people don't exist when people who oppose White genocide confront them. But when we're not around, anti-Whites blame all White people everywhere and everywhen for slavery, the Holocaust, discrimination, and also talk about how "evil" we are. Anti-racist is a codeword for anti-White.
-
Why are you justifying White Genocide?
-
And still these mistakes are made everyday. I, for one, am happy that he shed light on the subject.
-
yes, it is something so simple yet something most people do not take into account and simply assume
-
I don't understand why everyone is praising the talk like it is something so profound. Everything he said in this video is basic Bayesian statistics. Not saying that the professor isn't brilliant but he is really talking about something really simple in statistics.