Professor Eakin and All --

> I caught part of a news story on TV saying that an office 
> that had tested
> positive for anthrax was pronounced clean after further testing. Does
> anyone have any numbers on the probability of a false 
> positive when using
> the test for anthrax? This info would be useful since I am just now
> talking about Type I and Type II errors in class.

  I have spent some time in the library looking, but have not been very
successful.  My sense of things is that the actual test(s) is/are very much
a pattern-matching exercise (so the probabilities might not be simple), and
also that the prevalance might be so slight nobody has much of a clue about
the prevalance of false positives & false negatives.

  You will find some numbers in Li, L., & Kuo, H. P.  A study of
string-of-pearls test of B. anthracis.  Biological Abstracts, 49, entry #
40917. 1968.

  By the time I ran out of time, I had found the following, and would
suggest them as a decent place to start looking for more recent numbers with
much more recent techniques:

  Turnbull, P. C. B.  Definitive identification of Bacillus anthracis -- a
review.  J of Applied Microbiology, 1999, 87, 237-240.

  Henderson, et al.  Differentiation of Basillus anthracis from other
Bacillus cereus group bacteria with the PCR.  International Journal of
Systematic Bacteriology, 1994, 44, 99-105.

  Turnbull, P. C. B., et al.  Bacillus anthracis but not always anthrax.
Journal of Applied Bacteriology, 1992, 72, 21 - 28.

  Please let me know if you find anything interesting!

  -- Chris

Chris Olsen
George Washington High School
2205 Forest Drive SE
Cedar Rapids, IA

(319)-398-2161 

> -----Original Message-----
> From: EAKIN MARK E [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, October 18, 2001 6:03 PM
> To: edstat
> Subject: false positives when testing for anthrax
> 
> 

> 
> 
> Mark Eakin    
> Associate Professor
> Information Systems and Management Sciences Department
> University of Texas at Arlington
> [EMAIL PROTECTED] or
> [EMAIL PROTECTED]
> 
> 
> 
> =================================================================
> Instructions for joining and leaving this list and remarks about
> the problem of INAPPROPRIATE MESSAGES are available at
>                   http://jse.stat.ncsu.edu/
> =================================================================
> 


=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
                  http://jse.stat.ncsu.edu/
=================================================================

Reply via email to