Validity vs. Reliability

 

Validity is the evaluation of the adequacy and appropriateness of the interpretations and uses of assessment results. (p. 73, Linn & Gronlund).  Validity refers to the appropriateness of the interpretation of the results, is a matter of degree, is always specific to some particular use or interpretation, is a unitary concept, and involves an overall evaluative judgment.  Validity includes considerations that contribute to the meaning of a set of results which includes evidence of content, construct, assessment-criterion relationships, and consequences. (p. 103, Linn & Gronlund)

 

Reliability is the consistency of assessment results.  We cannot have validity without reliability.  If something is reliable we should expect similar scores from the same students when scores are taken on the same assessment during two occasions. (p. 74 & 107, Linn & Gronlund).

 

Validity refers to the whether the assessment interpretations really measure the objective being tested.  To be valid, the assessment should contain a relevant and representative sample of the content or tasks being assessed.  One should clearly identify what is being measured, what factors affect the measurement and if it can, indeed, be measured by the assessment. One should be sure that there are logical analysis and correlational studies that support the assessment interpretation.  Evidence must be present to support the validity of an assessment.

 

While validity refers to interpretation, reliability refers to whether the interpretation holds up or is similar on different occasions with the same group of students.