• Pretest­Posttest Design : research design where behavior is measured both before and after a treatment or condition is implemented o Quasi­experiment because there is no random assignment of participants to the treatment o All of the participants get the treatment
  • History Effect : events that occur during the course of a study to all or individual participants that can result in bias o During the time the depression study was being conducted, the economy took a turn for the worse and many people were losing their jobs and having difficulty paying bills
    • This can affect the depression levels of participants in the study
    • Makes it difficult to detect positive effects of the treatments being tested
  • Nonequivalent groups : participants are not randomly assigned
  • Solomon Four Group Design : pretest­posttest design with two sets of nonequivalent groups, one set that takes the pretest and posttest and one set that only takes the posttest o Allows comparison of the two sets of groups
  • Time series design : where patterns of scores over time are compared from before a treatment is implemented and after a treatment is implemented o Behaviors or attitudes may change week to week or month by month, so a pretest and posttest are unlikely to capture the true nature of the behavior
    • Used to determine if patterns of scores before and after treatment has a difference over time
  • Interrupted Time Series Design : where the “treatment” is an independent event, such as a historical event o War, passage of a new law, or other historical events are considered “treatments” in an interrupted time series design in that patterns of scores are compared before and after the event occurs
  • Non­interrupted Time Series Design : where the “treatment” is implemented by the researcher o A researcher interested in the effect of new therapy on depression. The researcher can choose to conduct an interrupted time series design study because depressive symptoms may change throughout the year
  • Maturation : natural changes that occur to the participants during the course of a study that can result in bias o Can affect a study conducted over a period of time
    • Suppose there are 2 groups of depressed participants: one group receives drug treatment for 12 weeks, whereas the other group receives a talk therapy for 12 weeks. Supposed the same amount of improvement occurs in both groups. One interpretation is that both treatments are equally effective in improving symptoms of depression. However, another possible interpretation is that symptoms of depression improve on their own without treatment and that neither treatment is actually affective.
    • Maturation affected the data such that the participants improved as a natural consequences of factors that occurred over the 12 week period
  • Attrition /Mortality: occurs when participants choose not to complete a study o It is possible that the participants who drop out of a study may have difference characteristics from the participants who remain in the study, thus biasing the full set of data toward the participants with the characteristics required for completing the study

Chapter Summary:

  • A quasi­experiment can be similar in design to an experiment, but no true IV is manipulated in the study
  • To test causal relationships between variables, an IV must be manipulated such that the participants are assigned to levels of the IV by the researcher
  • Pretest­posttest design is where the DV is measured both before and after treatment is implemented o It is a quasi­experiment because other factors may affect a change in scores besides the treatment (extraneous variables)
    • However, if a control group that does not receive the treatment is included with the participants randomly assigned to the treatment or control group, the design becomes an experiment
  • A time series design is similar to a pretest­posttest design in that the DV is measured more than once. o However, in a time series design, the DV is measured many times over a period of time, which may or may not include a treatment
    • The pattern of scores examined in a time series design may be affected by factors other than the treatment