Page 43 - Statistics for Dummies

Studying surveys of all shapes and sizes
Surveys and polls are among the most visible
actually be a good number of subjects in some
mechanisms used by today’s media to grab your
situations. The issues of what sample size is
large enough and what percentage of respon-
attention. It seems that everyone wants to do a
dents is big enough are addressed in full detail
survey, including market managers, insurance
in Chapter 16.)
companies, TV stations, community groups, and
even students in high school classes. Here are
Some surveys are based on current interests
just a few examples of survey results that are
and trends. For example, a recent Harris-
part of today’s news:
Interactive survey found that nearly half (47%)
of U.S. teens say their social lives would end
With the aging of the American workforce,
companies are planning for their future lead-
or be worsened without their cellphones, and
57% go as far as to say that their cellphones
ership. (How do they know that the American
workforce is aging, and if it is, by how much is
are the key to their social life. The study also
found that 42% of teens say that they can text
it aging?) A recent survey shows that nearly
67% of human-resources managers polled said Chapter 2: The Statistics of Everyday Life 27
while blindfolded (how do you really test this?).
that planning for succession had become more Keep in perspective, though, that the study did
important in the past five years than it had been not tell you what percentage of teens actually
in the past. The survey also says that 88% of the have cellphones or what demographic charac-
210 respondents said they usually or often fill teristics those teens have compared to teens
senior positions with internal candidates. But who do not have cellphones. And remember
how many managers did not respond, and is that data collected on topics like this aren’t
210 respondents really enough people to war- always accurate, because the individuals who
rant a story on the front page of the business are surveyed may tend to give biased answers
section? Believe it or not, when you start look- (who wouldn’t want to say they can text blind-
ing for them, you’ll find numerous examples in folded?). For more information on how to inter-
the news of surveys based on far fewer par- pret and evaluate the results of surveys, see
ticipants than 210. (To be fair, however, 210 can Chapter 16.
Studies like this appear all the time, and the only way to know what to believe
is to understand what questions to ask and to be able to critique the quality
of the study. That’s all part of statistics! The good news is, with a few clarify-
ing questions, you can quickly critique statistical studies and their results.
Chapter 17 helps you do just that.
Studying sports
The sports section is probably the most numerically jampacked section of the
newspaper. Beginning with game scores, the win/loss percentages for each
team, and the relative standing for each team, the specialized statistics reported
in the sports world are so deep they require wading boots to get through.
For example, basketball statistics are broken down by team, by quarter, and
3/25/11 8:18 PM
06_9780470911082-ch02.indd 27 3/25/11 8:18 PM
06_9780470911082-ch02.indd 27