Quiz 2
Instrumentation
Changes in reliability of instruments (rater fatigue, equipment failure)
true experimental designs
Designs where groups are randomly formed that control most sources of invalidity
Expectancy
Experimenters or testers anticipating that certain participants will perform better
9 Threats to Internal Validity
History, Maturation, Testing, Instrumentation, Statistical Regression, Selection Bias, Experimental Mortality, Selection Maturation Interaction, Expectancy
true experimental designs
Randomized groups design, pretest posttest randomized groups design
Statistical regression
Selection based on extreme score (in subsequent observations, a child who were very active will be less active and the very inactive child will be more active)
Testing
Sometimes called carry over effect/learning effect
Maturation
The physical, intellectual, or emotional changes that occur naturally in participants over time (aging, loss of motivation)
reactive effect of experimental arrangements
Treatments work in labs but not in the real world. Hawthorne effect.
History
Unexpected events occurred during the experiment that are not part of the treatment (participating in other program before/during intervention)
pre experimental designs
designs that control very few sources of invalidity; no random assignment of participants.
Selection maturation interaction
groups with different maturity characteristics (one group is 5 months younger than the others; school districts have different admission policies)
Selection bias
Problems in non random sampling and non random assignment, which lead to systematically different groups. Volunteers have the potential to not representative because of high motivation
Halo effect
Raters ratings are affected by previous experience or knowledge about the subject
What are the 4 threats to external validity?
Reactive or interactive effects on testing, interaction of selection bias and experimental tx, Reactive effect of experimental arrangements, multiple treatment interference.
John henry effect
Subjects try harder because they know they are in the treatment group, therefore should do better.
interaction of selection bias and experimental tx
Treatment may work only on participants selected on specific characteristic, not on the general population they represent. Avis Effect
pre experimental designs
one shot study, one group pretest and posttest design, static group comparison.
multiple treatment interference
one treatment may influence the next treatment findings may be confounded by the use of multiple treatments.
ExperimentalMortality
participants drop out, which changes the characteristics of the groups and may significantly affect the study's results.
reactive or interactive effects on testing
wearing a motion sensor may make people react in a way that they engage in more physical activity than they would normally. Performing a pre test can motivate people to work harder