Test validity is widely understood as the degree to which a test measures what it should measure (cf. Cattell, 1946). We argue that this conceptualization does not refer to a psychometric problem but to the correspondence between scientific language and everyday language.
Following Steven’s (1946), test results give an operational definition of attributes, qualifying any test as valid by definition. Following the representational theory of measurement (Krantz, Luce, Suppes & Tversky, 1971), an attribute is defined by an empirical relational structure and a corresponding measurement model. Since measurement depends on the specified empirical structure, if a test measures anything, it must be valid.
However, the question of validity can be asked in a meaningful way, if one interprets test results in the context of everyday language. We conclude that validity can be understood as the degree to which the variable measured by a test corresponds to concepts of everyday language.