Page 344 - Statistics for Dummies

328
Part VI: The Part of Tens
the results from the two groups, attributing any significant differences to the
treatment (and to nothing else, in an ideal world).
This seaweed study wasn’t a designed experiment; it was an observational
study. In observational studies, no control for any variables exists; people
are merely observed, and information is recorded. Observational studies are
great for surveys and polls, but not for showing cause-and-effect relationships,
because they don’t control for confounding variables. A well-designed experi-
ment provides much stronger evidence.
If doing an experiment is unethical (for example, showing smoking causes
lung cancer by forcing half of the subjects in the experiment to smoke ten
packs a day for 20 years while the other half of the subjects smoke nothing),
then you must rely on mounting evidence from many observational studies
over many different situations, all leading to the same result. (See Chapter 17
for all the details on designing experiments.)
Inspect the Numbers
Just because a statistic appears in the media doesn’t mean it’s correct. In
fact, errors appear all the time (by mistake or by design), so stay on the look-
out for them. Here are some tips for spotting botched numbers:
✓ Make sure everything adds up to what it’s reported to. With pie charts,
be sure all the percentages add up to 100 percent (subject to a small
amount of rounding error).
✓ Double-check even the most basic of calculations. For example, a pie
chart shows that about 83.33 percent of Americans are in favor of an
issue, but the accompanying article reports “7 out of every 8” Americans
are in favor of the issue. Are these statements saying the same thing?
No; 7 divided by 8 is 87.5 percent — if you want 83.33 percent, it’s 5 out
of 6.
✓ Look for the response rate of a survey; don’t just be happy with the
number of participants. (The response rate is the number of people
who responded divided by the total number of people surveyed times
100 percent.) If the response rate is much lower than 50 percent, the
results may be biased, because you don’t know what the non-respon-
dents would have said. (See Chapter 16 for the full scoop on surveys and
their response rates.)
✓ Question the type of statistic used, to determine whether it’s appropri-
ate. For example, suppose the number of crimes went up, but so did the
population size. Instead of reporting the number of crimes, the media
need to report the crime rate (number of crimes per capita).

3/25/11 8:12 PM
29_9780470911082-ch20.indd 328 3/25/11 8:12 PM
29_9780470911082-ch20.indd 328