slideshare ppt on research

Monday, 24 September 2012

Measurement Validity and Reliability

- Measurement validity isn’t only about whether respondents lie on the survey; responses may be inaccurate for lots of unintentional reasons.

- Measurement reliability is about the consistency of responses for specific individuals on
specific questions; re-tests are usually for a subset of the original sample and must be within a short time frame (e.g. 2 weeks) so that nothing changes in reality.

o Do not confuse test-retest reliability with a longitudinal study, in which surveys are
repeated after a long period of time and the expectation is that there will be changes
during this period.

o Do not confuse test-retest reliability with convergent validity, in which a survey
includes two or more different questions that measure the same variable and the
responses are compared for consistency