Page 48 - Statistics II for Dummies

32 Part I: Tackling Data Analysis and Model-Building Basics

✓ Designing the data-collection instrument: Poorly designed instruments,
including surveys and their questions, can result in inconsistent or even
incorrect data. A survey question’s wording plays a large role in whether
or not results are biased. A leading question can make people feel like
they should answer a certain way. For example, “Don’t you think that
the president should be allowed to have a line-item veto to prevent gov-
ernment spending waste?” Who would feel they should say no to that?

✓ Collecting the data: In this case, bias can infiltrate the results if some-
one makes errors in recording the data or if interviewers deviate from
the script.
✓ Deciding how and when the data is collected: The time and place you
collect data can affect whether your results are biased. For example, if
you conduct a telephone survey during the middle of the day, people
who work from 9 to 5 aren’t able to participate. Depending on the issue,
the timing of this survey could lead to biased results.
The best way to deal with bias is to avoid it in the first place, but you also
can try to minimize it by

✓ Using a random process to select the sample from the population.
The only way a sample is truly random is if every single member of the
population has an equal chance of being selected. Self-selected samples
aren’t random.
✓ Making sure the data is collected in a fair and consistent way. Be sure
to use neutral question wording and time the survey properly.

Don’t put all your data in one basket!

An animal science researcher came to me one But after looking at his data for a few minutes I
time with a data set he was so proud of. He made a terrible realization — all his data came
was studying cows and the variables involved from exactly one cow. With no other cows to
in helping determine their longevity. His super- compare with and a sample size of just one, he
mega data set contained over 100,000 observa- had no way to even measure how much those
tions. He was thinking, “Wow, this is gonna be results would vary if he wanted to apply them to
great! I’ve been collecting this data for years another cow. His results were so biased toward
and years, and I can finally have it analyzed. that one animal that I couldn’t do anything with
There’s got to be loads of information I can get the data. After I summoned the courage to tell
out of this. The papers I’ll write, the talks I’ll be him so, it took a while to peel him off the floor.
invited to give . . . the raise I’ll get!” He turned The moral of the story, I suppose, is to run your
his precious data over to me with an expectant big plans by a statistician before you go down a
smile and sparkling eyes. cow path like this guy did.

06_466469-ch02.indd 32 7/24/09 9:31:39 AM