Page 285 - Statistics for Dummies

Chapter 17: Experiments: Medical Breakthroughs or Misleading Results?
new laser surgery technique for nearsightedness, and you’ve been selected at
random to participate in our study. When can you come in for the surgery?”
Something tells me that this approach wouldn’t go over very well with many
people receiving the call (although some would probably jump at the chance,
especially if they didn’t have to pay for the procedure).
The point is that getting a truly random sample of people to participate in an
experiment is generally more difficult than getting a random sample of folks
to participate in a survey. However, statisticians can build techniques into the
design of an experiment to help minimize the potential bias that can occur.
Making random assignments
One way to minimize bias in an experiment is to introduce some randomness.
After the sample has been decided on, the subjects are randomly divided
into treatment and control groups. The treatment groups receive the vari-
ous treatments being studied, and the control group receives the current (or 269
standard) treatment, no treatment, or a placebo. (See the section “Designing
the experiment to make comparisons” earlier in this chapter.)
Making random assignments of subjects to treatments is an extremely critical
step toward minimizing bias in an experiment. Suppose a researcher wants to
determine the effects of exercise on heart rate. The subjects in his treatment
group run 5 miles and have their heart rates measured before and after the
run. The subjects in his control group sit on the couch the whole time and
watch reruns of old TV shows. Which group would you rather be in? Some
health nuts out there would no doubt volunteer for the treatment group. If
you’re not crazy about the idea of running five miles, you may opt for the
easy way out and volunteer to be a couch potato. (Or maybe you hate to
watch old reruns so much that you’d run five miles to avoid that.)
Finding volunteers
To find subjects for their experiments, research- doctors and patients depend on these studies
ers often advertise for volunteers and offer them being representative of the general population.
incentives such as money, free treatments, or In order to recruit such representative subjects,
follow-up care for their participation. Medical researchers have to do a broad advertisement
research on humans is complicated and diffi- campaign and select enough participants with
cult, but it’s necessary in order to really know enough different characteristics to represent a
whether a treatment works, how well it works, cross section of the populations of folks who
what the dosage should be, and what the side will be prescribed these treatments in the
effects are. In order to prescribe the right treat- future.
ments in the right amounts in real-life situations,

3/25/11 8:13 PM
25_9780470911082-ch17.indd 269
25_9780470911082-ch17.indd 269 3/25/11 8:13 PM