Skip nav

A ruler with the word 'Outcome' next to it being measured.

For each outcome, at least one appropriate measure (or instrument) must be identified. The scientific usefulness of a measure is characterized in terms of its validity and its reliability. Validity is the degree to which an instrument measures what it purports to measure. Reliability refers to an instrument's consistency of measurement. Ideally, a measure will give consistent results even when it is administered in different places, at different times, or by different people (this latter aspect is known as inter-rater reliability). In addition, related items should produce similar results, showing consistency of results within the measure itself (this is known as inter-item reliability).

When you can document that your measures have demonstrated validity and reliability, you can be much more certain in the quality of the data they yield. Validity and reliability of measures are key aspects of research quality, and in fact they account for one third of the total Quality of Research score that NREPP assigns when reviewing a given outcome in a study.

Select "Next" to continue.