Validity is the degree to which a measurement measures what it sets out to measure ie. Its accuracy- two types: Internal or External

External Validity/Generalisability:
The extent to which the results of a study can also apply to people who are not in the study

Internal Validity:
The degree to which the results of an observation are correct for a particular group of people being studied. Internal validity can be undermined by all types of systematic error and can be improved by good research design.

Types of Internal Validity:
Face validity- how much a method measures what it is intended to measure
Content validity-
Criterion validity- how much it agrees with the “gold standard”
Construct validity- how much it agrees with other tests

Examples of Threats to Validity:
• low statistical power
• assumptions of statistical tests violated
• reliability/variability of data
• prior involvement of some subjects with the intervention
• subjects respond differently on second test due to passage of time (eg older; healthier; etc)
• subjects leave study
• repeated tests of a variable changes the responses
• changes in calibration of instrument
• biased selection of subjects
• control group unintentionally exposed to intervention

Hawthorne Effect:
Attention paid to participants taking part in research alters their behaviour

Rosenthal Effect:
When expectations of the researcher are conveyed to the participants’  participants respond. Deal with by double blinding or withholding information from participant (? ethical)

Comments are closed.