PSY 3-2 Flashcards
(20 cards)
When does a statistically significant result have little practical significance?
When external validity is poor, the effect size is very small, the treatment is too costly, or the effect is comparable to an existing treatment.
In multiple testing, how does Type I error probability change?
Increases: the probability of committing at least one Type I error across multiple tests is higher than for a single test.
What is the probability of at least one Type I error across two tests at α = .05?
About .0975
What is the probability of committing Type I errors in both tests at α = .05?
.0025
Why is replication important in hypothesis testing?
Repeated significant results make it very unlikely that findings are due to Type I errors.
What does it mean if only one out of twenty experiment replications is significant?
It is likely due to chance and not very impressive.
What does it mean if all twenty replications of an experiment are significant?
It is very unlikely to occur by chance and is very impressive.
What defines a quasi-experiment?
t lacks some features of true experiments, involves manipulation of an IV or treatment, and often lacks random assignment.
What are two basic types of one-group designs in quasi-experiments?
One-group posttest-only design and one-group pretest-posttest design.
threats to internal validity in quasi-experiments.
History, Maturation, Testing, Instrument Decay.
History:
Events that occur during participation that affect behavior
Maturation:
Changes due to the passage of time
that affect behavior
Testing
Taking a test can affect subsequent testing
Instrument Decay
Changes in measurement instruments (including observers) over time
Threats to internal validity controlled by true experiments (and that often exist in quasi-experiments)
Regression toward the mean, Subject attrition (mortality), Selection
Regression toward the mean:
Extreme scores are likely to be followed by more moderate scores
Subject attrition (mortality):
Participants selectively drop out of the experiment.
Selection:
When control and experimental groups
are chosen in such a way that they are not equivalent
Nonequivalent control group design
Uses an experimental group and control group, but they are not equivalent (e.g., natural groups)
One lab does exercises, other lab doe not
Groups are “Self-Selecting”
Nonequivalent control group pretest-posttest design
Nonequivalent groups (use pretest to show equivalence; or use pretest to show differential change for the two groups)
Possible additional problems: Different Section Leaders (a confound), observer bias, contamination
