For each experiment described below, identify any confounds that may be present. (Be careful not to identify things as confounds that are not.)

Then, redesign each study to eliminate any confounds that you find. Write a short paragraph for each case below, identifying the confound and indicating how you would eliminate it.

a. A pharmaceutical company developed a new drug to relieve depression and hired a research organization to investigate the potential effectiveness of the drug. The researchers contacted a group of psychiatric patients who were experiencing chronic depression and randomly assigned half of the patients to the drug group and half of the patients to the placebo group. To avoid any possible confusion in administering the drug or placebo to the patients, one psychiatric nurse always administered the drug and another nurse always administered the placebo.

However, to control experimenter expectancy effects, the nurses did not know which drug they were administering. One month later the drug group had dramatically improved compared to the placebo group, and the pharmaceutical company concluded that the new antidepressant was effective.

b. An investigator hypothesized that people in a fearful situation desire to be with other individuals. To test her hypothesis, the experimenter randomly assigned 50 participants to either a high or low fear group. Participants in the low fear group were told that they would be shocked but that they would experience only a small tingle that would not hurt. Participants in the high fear group were told that the shock would be quite painful and might burn the skin, but would not cause any permanent damage.

After being told this, eight participants in the high fear group declined to participate in the study. The experimenter released them (as she was ethically bound to do) and conducted the experiment. Each group of participants was then told to wait while the shock equipment was being prepared and that they could wait either in a room by themselves or with other people. No difference was found in the extent to which the high and low fear groups wanted to wait with others.

c. A study was conducted to investigate the hypothesis that watching televised violence increases aggression in children. Fifty kindergarten children were randomly assigned to watch either a violent or a nonviolent television program. After watching the television program, the children were allowed to engage in an hour of free play while trained observers watched for aggressive behavior and recorded the frequency with which aggressive acts took place.

To avoid the possibility of fatigue setting in, two observers observed the children for the first 30 minutes, and two other observers observed the children for the second 30 minutes. Results showed that children who watched the violent program behaved more aggressively than those who watched the nonviolent show. (Be careful with this one!)