Page 40 - Statistics for Dummies

24
Part I: Vital Statistics about Statistics
to which statistics occur in the media is mind-boggling. You may not even be
aware of how many times you’re hit with numbers nowadays.
This section looks at just a few examples from one Sunday paper’s worth of
news that I read the other day. When you see how frequently statistics are
reported in the news without providing all the information you need, you may
find yourself getting nervous, wondering what you can and can’t believe any-
more. Relax! That’s what this book is for — to help you sort out the good infor-
mation from the bad (the chapters in Part II give you a great start on that).
Probing popcorn problems
The first article I came across that dealt with numbers was “Popcorn plant
faces health probe,” with the subheading: “Sick workers say flavoring chemi-
cals caused lung problems.” The article describes how the Centers for Disease
Control (CDC) expressed concern about a possible link between exposure to
chemicals in microwave popcorn flavorings and some cases of fixed obstruc-
tive lung disease. Eight people from one popcorn factory alone contracted this
lung disease, and four of them were awaiting lung transplants.
According to the article, similar cases were reported at other popcorn facto-
ries. Now, you may be wondering, what about the folks who eat microwave
popcorn? According to the article, the CDC finds “no reason to believe that
people who eat microwave popcorn have anything to fear.” (Stay tuned.)
The next step is to evaluate employees more in-depth, including conducting
surveys to determine health and possible exposures to the said chemicals,
checks of lung capacity, and detailed air samples. The question here is: How
many cases of this lung disease constitute a real pattern, compared to mere
chance or a statistical anomaly? (You find out more about this in Chapter 14.)
Venturing into viruses
The second article discussed a recent cyber attack: A wormlike virus made
its way through the Internet, slowing down Web browsing and e-mail delivery
across the world. How many computers were affected? The experts quoted in
the article said that 39,000 computers were infected, and they in turn affected
hundreds of thousands of other systems.
Questions: How did the experts get that number? Did they check each com-
puter out there to see whether it was affected? The fact that the article was
written less than 24 hours after the attack suggests the number is a guess.
Then why say 39,000 and not 40,000 — to make it seem less like a guess? To
find out more on how to guesstimate with confidence (and how to evaluate
someone else’s numbers), see Chapter 13.

3/25/11 8:18 PM
06_9780470911082-ch02.indd 24 3/25/11 8:18 PM
06_9780470911082-ch02.indd 24