Psychology2016

The Science of Psychology 31

Experimental Hazards and Controlling for Effects

1.9 Recall two common sources of problems in an experiment and some ways to
control for these effects.

There are a few other problems that might arise in any experiment, even with the use
of control groups and random assignment. These problems are especially likely when
studying people instead of animals, because people are often influenced by their own
thoughts or biases about what’s going on in an experiment.

THE PLACEBO EFFECT AND THE EXPERIMENTER EFFECT For example, say there is
a new drug that is supposed to improve memory in people who are in the very early
stages of Alzheimer’s disease (a form of mental deterioration that occurs in some people
as they grow old). to Learning Objective 6.13. Researchers would want to test
the drug to see if it really is effective in helping improve memory, so they would get
a sample of people who are in the early stages of the disease, divide them into two
groups, give one group the drug, and then test for improvement. They would probably
have to do a test of memory both before and after the administration of the drug to be
able to measure improvement.

Let me see if I’ve got this straight. The group that gets the drug

would be the experimental group, and the one that doesn’t is the

control group, right?

Right, and getting or not getting the drug is the independent variable,
whereas the measure of memory improvement is the dependent variable. But
there’s still a problem with doing it this way. What if the researchers do find that
the drug group had greater memory improvement than the group that received
nothing? Can they really say that the drug itself caused the improvement? Or is
it possible that the participants who received the drug knew that they were sup-
posed to improve in memory and, therefore, made a major effort to do so? The
improvement may have had more to do with participants’ belief in the drug than
the drug itself, a phenomenon* known as the placebo effect: The expectations
and biases of the participants in a study can influence their behavior. In medical
research, the control group is often given a harmless substitute for the real drug,
such as a sugar pill or an injection of salt water, and this substitute (which has no
medical effect) is called the placebo. If there is a placebo effect, the control group
will show changes in the dependent variable even though the participants in that
group received only a placebo.
Another way that expectations about the outcome of the experiment can
influence the results, even when the participants are animals rather than people,
is called the experimenter effect. It has to do with the expectations of the experi-
menter, not the participants. As discussed earlier in the section about naturalistic
observations, sometimes observers are biased—they see what they expect to see.
Observer bias can also happen in an experiment. When the researcher is measuring the
dependent variable, it’s possible that he or she could give the participants clues about how
they are supposed to respond—through the use of body language, tone of voice, or even
eye contact. Although not deliberate, it does happen. It could go something like this in the
memory drug example mentioned earlier: You, the Alzheimer’s patient, are in the exper-
imenter’s office to take your second memory test after trying the drug. The experimenter
seems to pay a lot of attention to you and to every answer that you give in the test, so
you get the feeling that you are supposed to have improved a lot. So you try harder, and

This elderly woman has Alzheimer’s disease, which

causes a severe loss of recent memory. If she were given

a new drug in the very early stages of her disease, in the

attempt to improve her memory, the researcher could not

be certain that any improvement shown was caused by

the drug rather than by the elderly woman’s belief that

the drug would work. The expectations of any person

in an experimental study can affect the outcome of the

study, a phenomenon known as the placebo effect.

placebo effect

the phenomenon in which the expec-

tations of the participants in a study

can influence their behavior.

experimenter effect
tendency of the experimenter’s expec-
tations for a study to unintentionally
*phenomenon: an observable fact or event. influence the results of the study.