Threats Flashcards
One group, pretest/posttest design
Recruiting one group of participants, measuring them on a pretest, expose to treatment and measure posttest.
Maturation threat
Change in behaviour that emerges spontaneously over time. Prevented by control group
History threat
Historical or external factor that systematically affects most members. Prevented by control group
Regression threat (regression to the mean)
Unusually good or bad outcome, the next time it’s measured it’ll likely be less extreme. Prevented by control group
Attrition threat
When participants systematically drop out (only one kind of participant, extreme scores leave). Prevented by taking them out or checking pretest scores.
Testing threat
Change in the participants as a result of taking a test. Becoming better or worse because of fatigue or repetition. Prevent by posttest only or alternative measurements.
Instrumentation threat
Measuring instrument changes over time. Observers might change standards or two forms that are not equivalent. Prevented by posttest only or ensuring they are equivalent. Counterbalancing; switching who gets the test when.
Combined: Selection-history threat/selection attrition threat
Outside event or factor effects only those at one level of the independent variable. / Only one group suffers attrition.
Observer bias
Researchers expectations influence their interpretation of results. Prevented by double blind or masked design(Participants know observers don’t).
Demand characteristics
Participants guess what the study is about and change their behaviour in the expected direction. Prevented with double blind
Placebo effect
People receive treatment and improve, but only because the recipients believe they are receiving valid treatment. Prevented by double-blind placebo control study. Neither know, but if the actual drug group improved more you know it worked/no treatment control group
Non-specific effect
Effects not due to the treatment but due to the expectation that you will be treated. Prevented with waiting list control group. Giving the treatment after the posttest (not concerned with the results of treatment).
