Real-Life Decision Making
It's the weekend. You rarely watch television, but you have some free
time and so you ask yourself, "Why not?" However, not long after you turn
on the tube you remember why you don't watch television very often: too
many commercials! Use this laundry detergent, fly on this airline, run in
these shoes!
You become frustrated and are about to turn it off and do some work on
your award-winning garden when an advertisement comes on that appeals to your
professional interest. The commercial is for FastGrow, a new brand of tomato
seeds. The announcer claims that FastGrow seeds produce larger tomatoes than
the tomatoes grown from HeartyGrow seeds, the current number one selling brand
on the market. Even though you are a plant physiologist and often do research
on the effectiveness of tomato seed hybrids, you have never heard of FastGrow.
It strikes you that it would be interesting to conduct research to see
if FastGrow's advertisement is true. "It's not uncommon for plant
physiologists to conduct experiments testing the products of two different
companies," says plant physiologist Norm Hopper. "First, you try to determine
if one product is better than the other, and then you try to figure out why
you got the results you did."
To conduct your research, you find two acres of land that are as much alike
as possible. The soil composition and the amount of sunshine and water that
each acre gets should be equal. You plant FastGrow seeds in the first acre,
and HeartyGrow seeds in the second.
During the growing season, you monitor and record the growth of both acres
of tomatoes. At the end of the season, after the yields have been calculated,
your data shows that HeartyGrow grew 20 percent better than FastGrow. However,
before you publish the results of your study, you know you must repeat the
experiment at least one more time.
"Plant physiologists conduct experiments once, record the results, and
then conduct a second experiment exactly the same as the first one in order
to see if the results are consistent. No conclusions should be made until
the experiment has been conducted at least twice," says Hopper.
You repeat the experiment over a second growing season, but this time your
data shows that FastGrow outperformed HeartyGrow by 20 percent. The FastGrow
managers learn the results of your second experiment, and ask you to appear
in one of their television commercials to tell the world that your most recent
research showed that FastGrow seeds performed 20 percent better than HeartyGrow
seeds. What do you say?
You could say yes to the FastGrow managers because they are simply asking
you to say that FastGrow outgrew HeartyGrow in your most recent experiment
-- a claim that is absolutely true.
Then again, you might want to say no to the FastGrow people because you
know your results are inconclusive.