Wednesday, March 14, 2007

Numb3rs

Peter Donelly at TED on statistics.

What are the average number of coin flips before you will get the sequence "heads-tails-tails" versus the sequence "heads-tails-heads"? Even odds, right? Wrong. Listen and find out why. It makes perfect sense once you hear it.

If there is a test for a disease that is 99% accurate, what are the chances that a person receiving a positive test has the disease? As you might guess it is not 99%. It depends on how common the disease is in the general population.

These are illustrations of how statistics can be used to mislead people.

"Humans are not good at reasoning with uncertainty."

Pastor Rod

"Helping You Become the Person God Created You to Be"

3 comments:

John said...

What are the odds that you will get the sequence "heads-tails-tails" versus the sequence "heads-tails-heads"? Even odds, right? Wrong. Listen and find out why. It makes perfect sense once you hear it.

Your question is different than his answer. The odd would be the same for just three tosses - he is talking about a series of sequential tosses until the pattern appears.

Pastor Rod said...

John,

Thanks. You're right I didn't say that very well. Sloppy logic on my part.

I'll make the correction.

Rod

jeff franczak said...

How to have your statistical cake and eat to too... [grin]

“Howard Littlefield...Doctor of Philosophy in economics of Yale. ... He could...prove, absolutely, with figures all in rows and with precedents from Poland and New Zealand, that the street-car company loved the Public and yearned over its employees; that all its stock was owned by Widows and Orphans; and that whatever it desired to do would benefit the property-owners by increasing rental values, and help the poor by lowering rents.

Sinclair Lewis in the Novel “Babbitt”, 1922
(emphasis mine)