Threats to Validity

Validity = a measure of the truth or accuracy of a claim; the extent to which the data collection instrument actually reflects the abstract construct being examined.

Internal validity is the degree to which it can be inferred that the experimental treatment, rather than an uncontrolled condition, resulted in the observed effects (outcomes)

History - an unplanned event occurs during the study

Selection bias - subjects in study not randomly selected or assigned to treatment groups

Maturation - subjects changing during the study

Testing - outcomes due to number of times the subjects' responses were tested

Instrumentation - outcomes due to changes in measurement instruments between pretest and post-test rather than manipulated treatment (IV)

Mortality - subjects who drop out of study are different from subjects remaining in the study

External validity - external validity is the degree to which findings of a study can be generalized to other populations or environments

Hawthorne effect (reactive) - subjects responses influenced by their awareness of the study

Novelty - newness of tx. affect subject’s and researcher’s behavior, so when the newness

Interaction of history & treatment - external event coinciding with tx. may effect results, so that if study is replicated without the external event the result are not produced again with a new group of subjects

Experimenter - characteristics of researcher & his/her biases influence subject responses or influence researcher's observations of subjects