Validity Flashcards
Internal Validity (definition)
In experiments: the extent to which extraneous variables are controlled by the researcher. Controlling for extraneous variables allows the researcher to draw conclusions from the results of the study as being attributed to the IV. In qualitative research: the extent to which the researcher has demonstrated a causal relationship between two phenomena by showing that other plausible factors could not have caused the relationship (POSITIVIST VIEW, am I right?).
History (definition)
When an experiment is extended over a period of time. An event or other extraneous variables which have an effect on the dependent variable. Example: Covid-19.
Maturation (definition)
When Participants grow over times of the experiment (getting older allows for natural cognitive development).
Testing (definition)
When a participant becomes “Test-wise” because they had a pre-test that was the same as the post-test and therefore they got a better score due to exposure or practice.
Instrumentation (definition)
When the experimenter applies the treatment incorrectly. This can be scoring differently in pre-post tests, acting differently to certain participants, having two experimenters with different personalities, or different instructions are given, etc).
Statistical regression (definition)
Regression toward the mean. In pre-posttest designs, those that perform on the extreme ends of the pre-test (score really low or really high) tend to regress toward the mean at post-test. Meaning that those who score 100 at pre-test tend to score more like the average on post-test.
Differential selection (definition)
When the selection of the participants into the treatment or control group is bias. For example: when you put the advanced students in the treatment group and you put the low performers in the control group to achieve a bigger difference.
Experimental Mortality (Attrition) (definition)
When participants drop out of the study, or they miss a pretest or posttest.
Selection Maturation interaction (definition)
when participants are Differentially selected. Example: If the treatment group is all 6 months older than the control. The treatment group will mature faster than the control. Meaning that you won’t know if it is the treatment or the maturation that is the reason for any significant effects.
Experimental treatment diffusion (definition)
When the participants in the control group start getting some of the treatment because it is desirable. Example: teachers borrowing lessons from the treatment teacher because it sounds fun or beneficial.
Compensatory rivalry by the control group (definition)
When participants perform better than their normal level because they believe they are in competition with the treatment group.
Compensatory equalization of treatments (definition)
When the control group is given goods or services that are of equal desirability as the treatment. The control group may become another treatment group.
Resentful demoralization of the control group (definition)
When the control group thinks that the treatment group is better and that they are suffering because they do not have the treatment they may be less motivated to perform at their actual ability making the difference in posttest scores larger.
History (Combatting the Threat)
Have a control group that is the same as the treatment (can be done by random assignment). The control group will allow any event that happens to happen to both groups and therefore any difference found can be attributed to the IV.
Maturation (Combatting the Threat)
A randomly assigned control/comparison group can combat maturation. Since the two groups will have aged (matured) similarly across the groups. This will allow any differences found to be attributed to the IV and not maturation.
