As your textbook authors note, experimental research carries with it the risk of situational limitations and constraints. Differentiate artificiality, demand characteristics, and generalizability as they relate to experimental research. Identify how they can be controlled in an experiment and generate an example of each constraint and the problem it poses.

Question

What will be an ideal response?

Answer 1

Artificiality occurs when an experiment must be conducted in an environment and under conditions that are not "ordinary" for the behavior being studied. Artificiality can be improved by creating as natural an environment for the study as possible. For instance, if you are researching a question about the willingness of people to help others, bringing them into a laboratory situation where an artificial interaction occurs between the participant and someone needing help may bias the data. Most people, knowing that they are being observed, will change their behavior; therefore, it would lead to false behavior. Exposing participants to a similar situation in a natural environment may yield more accurate results. Demand characteristics are the cues that let a participant know the true purpose of an experiment. Due to their knowledge, the participants may alter their behavior. Demand characteristics can be reduced by using placebo's and deception. For instance, if participants are told that the study they are participating in is to measure morality, it will most likely change their behavior. So, the researcher can use deception to "hide" the true purpose of the experiment in hopes of getting more accurate results. Generalizability refers to the degree to which a study's findings are true of a larger group of people beyond the participants themselves. It can be increased by increasing the size of the sample used in the study. A consideration is that the majority of studies done in psychology are done with college students, and yet they are generalized to the population of the United States. College students may or may not be a good predictor of behavior in the general public.