soc150 lec 6

stimulus/treatment

independent variable

pretest

posttest

A measurement of the dependent variable taken at the outset of an experiment before the stimulus or treatment has been introduced.

opposite

history threat

The possibility that an observed change in the dependent variable is caused by extraneous or past events rather than by the experimental treatment. The history threat (also called the history effect) is one kind of threat to the internal validity of an experiment.

maturation threat

The possibility that a change in the dependent variable observed within an experiment could be caused by natural changes in research participants rather than by the experimental treatment. The maturation threat (also called the maturation effect) is one kind of threat to the internal validity of an experiment.

testing threat

The possibility that a change in the dependent variable observed within an experiment could be caused by the participants’ pretest responses influencing their posttest responses. The testing threat (also called the testing effect) is one kind of threat to the internal validity of an experiment.

instrumentation effect

differences in difficulty in pretest and posttest

regression to the mean

The statistical tendency of a group’s overall performance on a measure during a posttest to move toward the mean of that measure rather than in the hypothesized direction. For example, if participants score very low on a pretest, they will have a tendency to score higher on the posttest (closer to the mean). Regression to the mean (also called regression, though note that this usage is distinct from “regression” in the sense of regression analysis) is one kind of threat to the internal validity of an experiment.

selection threat

certain kind of participant more likely to be picked

classical experimental design

An experimental design with a number of features designed to ensure the experiment’s internal validity. (Also called a true experimental design.) Study participants are randomly assigned to an experimental group or a control group, and pretests and posttests (measurements of the dependent variable) are conducted before and after the introduction of the stimulus or treatment (for the experimental group) or any placebo (for the control group).

social interaction threat

treatment diffusion and imitation

compensatory rivalry and resentful demoralization

compensatory equalization of treatment

experimenter expectancy