Validity, History of

The Concept of Validity Prior to the 1950s

Validity theory has primarily been a concern among theorists in education and other social sciences. The advent of validity theory within these disciplines coincides with early efforts to legitimize the practice of educational and psychological measurement as a scientific activity. As educational and psychological testing became more prominent in the early 1900s, theorists sought to provide an account of validity to justify the consequential actions derived from test scores. Two developments influenced conceptions of validity before the publication of the first technical recommendations in 1952: The formalization of the correlation coefficient and concerns raised by educational researchers about the connection between test content and validity. Both developments are consistent with the view that tests themselves are valid or invalid.

Correlations, which reflect the direction and magnitude of a relationship between two variables, are crucial for making empirical predictions. The formalization of the correlation coefficient led to the concept of criterion-related validity. Criterion-related validity was a concern when researchers were interested in using a test as an indication of a criterion of ultimate interest. The criterion of interest may reflect the future or current status on a variable. Tests were therefore viewed as valid for anything with which it correlated. However, educational researchers indicated that the criterion-related view failed to address the role of test content. These researchers argued that the content of a test, such as an achievement test, should be evaluated for its relevance to, and representation of, a specified domain. In sum, prior to the 1950s, validity was largely viewed as two distinct, though in some sense connected, types which included criterion-related validity and content validity.

...