threats to internal validity

Invalid IVs or DVs

if either is invalid, the study doesn’t truly test what it claims to test (weak internal validity)

subject selection

groups differ before the study begins

history

outside events affect results

testing or practice

taking a test once affects later performance

sequencing

when the order in which participants experience conditions affects their performance

mortality

participants drop out unevenly between groups

subject bias

when participants change their behavior because they know they’re in a study. 

lack of random assignment to comparison

when participants aren’t randomly placed into groups, the groups may differ from the start

use appropriately sensitive DVs

if the dv is too crude or not precise, true effects may go unnoticed (weaker internal validity)

maturation

participants naturally change overtime

instrumentation

changes in measurement tools or observers

statistical regression

extreme scores move toward average on retesting

experimenter bias

researcher expectations influence results