Internal/External Validity Flashcards
Counterbalancing is most useful for controlling which of the following threats to external validity?
A. reactivity
B. selection-treatment interaction
C. multiple treatment interference
D. pretest sensitization
Answer C is correct. Counterbalancing is used to control multiple treatment interference, which may occur whenever each participant receives multiple levels of the independent variable. It involves administering the levels in a different order to different groups of subjects.
The Solomon four-group design is used to:
A. evaluate the effects of pretesting.
B. reduce the effects of multiple treatment interference.
C. ensure that groups are similar at the beginning of the study.
D. reduce experimenter expectancy effects.
Answer A is correct. The Solomon four-group design is used to identify the effects of pretesting on a study’s internal and external validity.
The best way to reduce the effects of differential selection on the results of a research study is to:
A. randomly select subjects from the population.
B. randomly assign subjects to the treatment groups.
C. use the single-blind technique.
D. use the double-blind technique.
Answer B is correct. Differential selection occurs when subjects in different groups differ at the beginning of a study in a way that affects the results of the study, and this difference is due to the way that subjects were assigned to the groups. The best way to control differential selection is to randomly assign subjects to groups, which helps ensure that subjects in the different groups are similar at the beginning of the study.
A researcher conducted a study to assess the effects of cognitive behavior therapy for treating depression in adolescents by comparing it to nondirective supportive therapy. Fifty adolescents who recently received a diagnosis of major depressive disorder were randomly assigned to one of the two therapies. Each participant’s level of depression was assessed at the beginning of treatment and one month after treatment ended. The results indicated that adolescents in both groups improved to the same degree in terms of depressive symptoms. Which of the following is the biggest threat to this study’s internal validity?
A. differential selection
B. reactivity
C. maturation
D. instrumentation
Answer C is correct. The lack of difference in the outcomes for the two groups in this study could be due to the fact that the treatments were equally effective or to maturation since symptoms of major depressive disorder tend to decline in severity or remit over time with or without treatment.
To evaluate a brief intervention for improving the social skills of adolescents with autism spectrum disorder, a researcher administers a measure of social skills to a sample of adolescents with this diagnosis, chooses the adolescents with the lowest scores on this measure to participate in the intervention, and then re-administers the measure of social skills one week after the adolescents have participated in the intervention. The biggest threat to the internal validity of this study is:
A. instrumentation.
B. reactivity.
C. pretest sensitization.
D. statistical regression.
Answer D is correct. Statistical regression threatens a study’s internal validity whenever subjects are chosen because of their extreme scores on a pretest. Because this study included only adolescents with the lowest scores on the measure of social skills, their scores on the posttest measure of social skills are likely to improve somewhat just because of statistical regression.
