13 terms

Statistics (Internal Validity)

internal validity and threats to internal validity
Internal Validity
The extent to which the intervention rather than extraneous factors account for the results of the experiment
Threats to Internal Validity
1) History
2) Maturation
3) Testing (Repeated Testing)
4) Instrumentation
5) Statistical Regression
6) Selection Biases
7) Attrition
8) Reaction of Controls & Special Treatment
9) Diffusion of Treatment & Imitation of Treatment
10) Combination of Selection and Other Threats
Events occurring within or outside experiment that may account for the results
Internal changes within the participant that occurs over time and may account for the results
(Repeated) Testing
Practice or familiarity with a test may influence performance at the second testing not the intervention
Changes in the instruments or procedures over time may account for the results
Statistical Regression
The tendency for extreme scores to move toward the mean after repeated testing may account for the results
Selection Biases
Systematic differences in the group due to the way they were selected or assigned to groups may account for the results
Loss of participants in systematic direction may account for the results
Combination of Selection and Other Threats
threats of internal validity vary across groups making the threat selective for one group and not the other
Selection + Another Threat to Internal Validity
Diffusion of Treatment
The treatment accidentally is given to some or all control group participants
Special Treatment
Control group is not given the treatment but is given special attention
Reaction of Control
Control group realizes that it is not receiving the treatment and act in ways that obscures the results