Must all tests be valid?

Psychometric Properties of Two Islamic Measures among Young Adults in Kuwait: The Sahin-Francis Scale of Attitude toward Islam and the Sahin Index of Islamic Moral Values

“…The majority of the respondents were highly religious in terms of their commitment to prayer: 20% practised the obligatory prayers and additional prayers; 60% practised the obligatory prayers; 17% prayed sometimes; 1% prayed on Fridays only; and 2% never prayed. Table 1 While the formal statistics of reliability are relatively easy to calculate for an attitude scale (Livingston, 1988), the question of assessing validity is more problematic (Ebel, 1961;Zeller, 1988). Steps towards assessing the construct validity of these scales can be made by assessing the extent to which certain predictions about the theoretical variations in attitude scores are reflected empirically (Cronbach & Meehl, 1955;Orton, 1987).…”

Assessment of reliability, sensitivity, objectivity and validity of MCQ Pharmacology Exams as a potential output variable for predictive analysis

“…"), then the assessment is not focused on features we would like to evaluate. 33 There is no precise method to measure validity, only its indirect evaluation is possible and it usually consists in using one out of three ideas used for establishing measurement validity: content, 34 empirical, 35 and construct validity. 36,22 analysed the validity of PCAT and showed that a number of assessment strategies can be used: content, internal structure, concurrent, and predictive validity.…”

Validity

“…We will next examine such one-sided content validity claims in more detail because they have a long history in educational and psychological measurement, arguing that they are at best only one-sided and at worst perverse. Ebel (1961) has argued that "if the test we propose to use prOVides in itself the best available operational definition, the concept of validity does not apply" (p. 643). But Cronbach (1971) counters that "this language gives the game &MaY, for the 'best available' definition is presumably not the best conceivable, and How good is the operation?…”