FAIR USE NOTICE.  This document contains copyrighted material 
whose use has not been specifically authorized by the copyright owner.
The CHANCE project is making this material available as part of our
mission to promote critical thinking about statistical issues.
We believe that this constitutes a `fair use' of the copyrighted
material as provided for in section 107 of the US Copyright Law.
If you wish to use this copyrighted material for purposes of your own
that go beyond `fair use', you must obtain permission from the
copyright owner.

                 Copyright 1993 The San Diego Union-Tribune
                          The San Diego Union-Tribune

                                 April  27, 1993


LENGTH: 1040 words

HEADLINE: Coke and Pepsi put fizz into statistics class

SOURCE:  Staff Writer


   At UCSD last week, a gaggle of professors and students tackled one of the key
questions of our age.

   "OK, OK, who here can tell the difference between Pepsi and Coke?" asked 
math professor Peter  Doyle,  preparing to stage a taste test among his students.

   The frivolity of the question was not lost on the Harvard-trained teacher nor
his students, but the intent of the University of California San Diego course is

   The experimental class, called "Chance," is part of a budding reform movement
aimed at invigorating math education on university campuses nationwide.

   Team-taught by  Doyle  and a visiting professor from Dartmouth College, 
J. Laurie  Snell,  the course puts a topical spin on statistics -- a subject that
often elicits yawns from undergraduates.

   "The basic statistics course is a notoriously boring course,"  Snell  said.

   The UCSD class combines lectures with lively group discussions and
experiments to probe the role that statistics and probability -- or chance --
play in issues from AIDS and cancer to lightweight concerns like soda rivalries.

   "It seems a lot more practical for a math class," said sophomore Amy
Wasserman.  "It relates to life."

   On Thursday, Wasserman sat in a circle of desks with five classmates as they
designed and executed a test to see if they could tell the difference between
Coca-Cola and Pepsi.

   Should they rinse their mouths with water in between tastes, someone asked.
Or use crackers?  And how could they be certain a student was not just guessing
the answer?

   Student Dave Yao sipped from a paper cup.  "I'm pretty sure it's Coke," he
said.  "But it might be Pepsi."

   Yao, a junior, slid his tongue along his palate and took the plunge as others
carefully documented his answer.  "It's Coke," he said.

   It was Pepsi.  students -- comparing one soda-filled cup to another --
successfully distinguished between the two over repeated trials.  Some even
wagered on their tasting skills.

   "The bet is even money that he can't tell the difference," said  Doyle,
pointing to student Michael Stavros.

   Stavros nailed it.  Coca-Cola it was.   Doyle  slipped his wallet out,
handing Stavros a buck.

    Doyle and Snell  used the experiment as a springboard to weightier issues,
emphasizing that the statistical issues raised by it are not that different from
what medical researchers or other scientists face in running clinical trials.

   In fact, a similar study conducted several decades ago comparing English teas
represented the birth of modern statistics,  Snell  said.

    Snell  said the course, which he and  Doyle  first taught at Princeton
University in 1991, often has a profound impact on students' faith in scientific

   An understanding of probability and statistics,  Doyle  said, allows students
to make their own decisions about whether they believe something or not.  In
recent weeks, the class has discussed the accuracy of AIDS tests and cancer

   "What I've been able to learn is that I can't assume as much as I used to,"
Wasserman said.

   Final examinations in the course often take an unconventional bent.  One
student in a Chance class on another campus gauged the predictability of the
order of songs played at a Grateful Dead concert.

   Versions of Chance are being offered at several East Coast and Midwest
universities with the help of a $300,000 grant from the National Science
Foundation (NSF).

    Snell,  who has taught math since 1954, said the grant's intent is to
encourage educators to experiment with other models of teaching besides 
extended lectures.

   "Colleges are just realizing that lecturing is not the only way to do
things,"  Snell  said.

   The course is being taught at UCSD for the first time, but is not part of the
NSF grant.

    Doyle  said he is unsure when it might be offered here again, but he and
 Snell  hope that professors nationwide will incorporate some of its lessons
into introductory statistics courses.

   "We're trying to develop something that will be useful across the country,"
 Doyle  said.

   And perhaps settle the Pepsi vs.  Coca-Cola issue at the same time.

    Doyle  bet someone that Stavros couldn't again tell the difference. Stavros
swished both colas in his mouth, one at a time, correctly naming each.

   "That was a dumb bet,"  Doyle  said.