The Math Forum

Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Math Forum » Discussions » sci.math.* » sci.stat.math

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: Bayes vs Hypothesis--false positive-negative and hypothesis nomenclature
Replies: 12   Last Post: Aug 4, 2012 2:13 PM

Advanced Search

Back to Topic List Back to Topic List Jump to Tree View Jump to Tree View   Messages: [ Previous | Next ]
W. eWatson

Posts: 132
Registered: 8/27/09
Bayes vs Hypothesis--false positive-negative and hypothesis nomenclature
Posted: May 17, 2012 2:19 PM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply

I'm confused about two seemingly different approaches to errors. I'm
looking at a non-stat book that uses the false pos/neg approach to
discuss type 1 and 2 errors. The author is discussing cognitive errors
(You misjudge a rustle of grass in the field as not a snake, but it is a
snake. It seems this (false pos/neg) has more to do with the Bayesian
approach than the usual testing of a null hypothesis.

Another book, The Cartoon Guide to Statistics (actually a good read to
introducing people to statistics) talks to both ideas. The "false
pos/neg approach is decidedly Bayesian. He really only addresses the
false positive in an example. He does not mention error. Much later he
gets into hypothesis testing with mention of type 1 and 2 errors.

Years ago I worked with medical people and they liked to use the false
pos/neg concept. It seems that's the typical approach in medical
articles I've read. Maybe the idea of "false" appeals to them more than
the idea of type. Quite possible it's used in psychology studies.

I would gather that we do have two ways of playing with errors. One is
based on the idea of conditional probabilities and the other on
probability distributions. Comments?

Point your RSS reader here for a feed of the latest messages in this topic.

[Privacy Policy] [Terms of Use]

© The Math Forum at NCTM 1994-2018. All Rights Reserved.