Page 355 - Statistics for Dummies

Chapter 21: Ten Surefire Exam Score Boosters
Using these three columns, fill in your if-then-how chart with each different
type of problem you’ve covered in class. Don’t write down every little exam-
ple; look for patterns in the problems and boil down the number of scenarios
to a doable list.
If-then-how charts should be customized to your needs, so the only way it’s
going to work is if you make it yourself. No two people think alike; what works
for your friend may not work for you. However, it might be helpful to compare
your chart with a friend’s once you are both finished, to see if you’ve left any-
thing out.

If you’re allowed to bring a review sheet to exams, I suggest putting your if-
then-how chart on one side. On the other side, write down those little nuggets
of information your professor gave you in lecture but didn’t write down. If you
aren’t allowed to have a review sheet during the exam, call me crazy, but I’ll
argue that you should still make one to study from. Making one really helps
you sort out all the ideas so when you take the exam you’ll be much more
clear about what to look for and how to set up and solve problems. Lots of 339
students come out of an exam saying they didn’t even use their review sheet,
and that’s when you know you’ve done a good job putting one together: When
it went on the sheet, it went into your mind!
Figure Out What the Question Is Asking
Students often tell me that they don’t understand what a problem is asking
for. That’s the million dollar question, isn’t it? And it’s not a trivial matter.
Oftentimes the actual question is embedded somewhere in the language of
the problem; it isn’t usually as clear as: “Find the mean of this data set.”
For example, a question may ask you to “interpret” a statistical result. What
does “interpret” really mean? To most professors the word “interpret” means
to explain in words that a nonstatistician would understand.
Suppose you are given some computer output analyzing number of crimes
and number of police officers, and you are asked to interpret the correlation
between them. First you pick off the number from the output that represents
the correlation (say it’s –0.85); then you talk about its important features in
language that is easy for others to understand. The answer I would like to
see on an exam goes something like this: “The correlation between number
of police officers and number of crimes is –0.85; they have a strong negative
linear relationship. As the number of police officers increases, number of
crimes decreases.”

3/25/11 8:12 PM
30_9780470911082-ch21.indd 339 3/25/11 8:12 PM
30_9780470911082-ch21.indd 339