Validity
Argue why the data reported really represents what it purports to represent
Good data should be valid, i. e. measure what it is supposed to measure, e. g. foot size is a reliable measure but it is not a valid measure say of intelligence. Validity is that a measurement indeed represents the construct/variable it purports to represent. It can be established in three ways:
Content validity, where the construct content is generally agreed to be self evident in the measure, e. g. asking "Did you enjoy the web site?" as a measure of web site enjoyment.
Criterion validity, where one measure is validated against another already accepted as valid, e. g. validating a web clicks "interest" measure against purchase data.
Construct validity, that the measure is a single construct that "hangs together" (convergent validity), and also separates itself from other measures (discriminant validity), e. g. a variable like "Circumstances of the Economy" lacks construct validity as it is not a unitary variable, i. e. one thing. Statistics like factor analysis can define the factors of a multi-dimensional construct.
One way to ensure valid measurement is to use a tool someone else has validated.
Example(s)
(Use a descriptive name, e. g. "ITExample". Or click on an existing collection and edit it.)