Page 356 - Statistics for Dummies

340
Part VI: The Part of Tens
If you know what the problem is asking for, you have a better chance of actu-
ally solving it. You’ll gain confidence when you know what you are supposed
to do. On the flip side, if you don’t know what the problem is asking, even
starting it will be very hard. Your anxiety will go up, which can affect your
ability to work other problems as well. So how do you boil down a problem to
figure out exactly what it’s asking for? Here are some tips to follow:
✓ Check the very last sentence of the problem — that’s usually where
the question is located. Rather than reading the entire problem a
second (and third and fourth) time and getting yourself all worked up,
just read it once and then focus on the end of the problem.
✓ Practice boiling down questions ahead of time. Look at all the exam-
ples from your lecture notes, your homework problems, and problems
in your textbook and try to figure out what each problem is asking for.
Eventually you’ll start to see patterns in the way problems are worded,
and you’ll get better at figuring out what they are really asking for.
✓ Ask your professor what clues you should look for, and bring example
problems with you. She will be impressed because you are trying to
figure out the big picture, and oh, how professors love those “big pic-
ture” questions! And after she helps you, you can add those to your if-
then-how chart (see “Make an ‘If-Then-How’ Chart”).
✓ Translate the wording of the problem into a statistical statement. This
involves labeling not only what you are given (as discussed in the next
section), but also what you want to find.
For example, Professor Barb wants to give 20 percent of her students an A
on her statistics exam; your job is to find the cutoff exam score for an A,
and this translates to “find the score representing the 80th percentile.”
Label What You’re Given
Many students try to work problems by pushing around numbers that are
given in the problem. This approach may work with easy problems, but
everyone hits the wall at some point and needs more support to solve harder
problems. You’ll benefit from getting into the habit of labeling everything
properly — labeling is the critical connection between the if column and the
then column in your if-then-how chart (described earlier in this chapter). You
may read a problem and know what you need to do, but without understand-
ing how to use what you’re given in the problem, you won’t be able to solve it
correctly. To really understand the numbers the problem gives you, take each
one and write down what it stands for.
Suppose you’re given the following problem to solve: “You want to use the
size of a house in a certain city (in square feet) to predict its price (in thou-
sands). You collect data on 100 randomly selected homes that have recently

3/25/11 8:12 PM
30_9780470911082-ch21.indd 340 3/25/11 8:12 PM
30_9780470911082-ch21.indd 340