PS295 Chapter Notes - Chapter 13: Experiment, Interrupted Time Series, Construct Validity
Document Summary
Do not have the control features of a true experiment, may lack either: Quasi-independent variables used to indicate the the variable is not a true iv; manipulated by the researcher but rather is an event that participants experienced for other reasons. Example may examine effects of naturally occurring events. Example may examine people in groups that already exist. Generally not as high in internal validity as experiments. Include some design features to eliminate as many threats to internal validity as possible. The one group pretest/posttest design as an illustration. Also called a preexperimental design it lacks control, has not internal validity, and fails to meet requirements of a research design. Gives sample of new university students measures of depression. Selects those who are most depressed as sample. Participants take part is success-failure therapy over 8 week period. Measures depression via self-report and also using a behavioural measure.
