Understanding Psychology (10th Ed)

 
Module 6 
Critical Research Issues 
53
disorders in children early enough to prevent permanent damage, to communicating 
more effectively with severely retarded children, and to reducing chronic pain in 
people. Still, the use of research using nonhumans is controversial, involving com-
plex moral and philosophical concerns. Consequently, all research involving non-
humans must be carefully reviewed beforehand to ensure that it is conducted ethically 
(Hackam, 2007; Saucier & Cain, 2006; Shankar & Simmons, 2009).
Threats to Experimental Validity: 
Avoiding Experimental Bias
Even the best-laid experimental plans are susceptible to
experimental bias
—factors 
that distort the way the independent variable affects the dependent variable in an 
experiment. One of the most common forms of experimental bias is experimenter 
expectations: An experimenter unintentionally transmits cues to participants about the 
way they are expected to behave in a given experimental condition. The danger is 
that those expectations will bring about an “appropriate” behavior—one that other-
wise might not have occurred (Rosenthal, 2002, 2003). 
A related problem is participant expectations about appropriate behavior. If you 
have ever been a participant in an experiment, you know that you quickly develop 
guesses about what is expected of you. In fact, it is typical for people to develop 
their own hypotheses about what the experimenter hopes to learn from the study. 
If participants form their own hypotheses, it may be the participant’s expectations, 
rather than the experimental manipulation, that produce an effect (Rutherford et al., 
2009). 
To guard against participant expectations biasing the results of an experiment, 
the experimenter may try to disguise the true purpose of the experiment. Participants 
who do not know that helping behavior is being studied, for example, are more apt 
to act in a “natural” way than they would if they knew. 
Sometimes it is impossible to hide the actual purpose of research; when that is 
the case, other techniques are available to prevent bias. Suppose you were interested 
in testing the ability of a new drug to alleviate the symptoms of severe depression. 
If you simply gave the drug to half your participants and not to the other half, the