Michael Rogers' Star Talk Weblog
Astronomy, technology, small business, politics
 

 








 
 

 Wednesday, October 09, 2002


News -- Report by National Research Council says lie-detectors tests don't work very well: http://www.nytimes.com/2002/10/08/science/08CND-LIE.html

Comment -- The problem is that the scientific evidence on the accuracy of polygraph testing is very weak. And, the test create a lot of false-positives -- that is, innocent people who are nervous or anxious can fail the tests and come across as liars. On the other hand, an calm, cool and collected liar can pass the test.

I guess the lesson is don't be impressed when a criminal suspect passes a lie detector test, or alarmed when it's reported that someone failed a test -- that person could still be innocent.


Click here to see current comments and add your own.

12:45:25 PM    


Click here to visit the Radio UserLand website. Click to see the XML version of this web page. © Copyright 2002 Michael Rogers.
Last update: 11/1/02; 1:57:53 PM.
Click here to send an email to the editor of this weblog.


 

October 2002
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Sep   Nov