kvmdock.blogg.se

Logical validity does not require the test developers to carry out criterion-based validation studies ( Henderson et al.

2007), whereas a predictive selection test (PST) could appear unrelated to the job and almost inevitably includes inherent error, relying on predictive relationships ( Henderson et al. Logical validity is strongly favoured by the courts when determining a PES, it is perceived that DTS are “job-related” ( Berkman v. Logical validity, more commonly known as “face” validity, is achieved when a task analysis includes consultation with subject-matter experts (SMEs), experienced supervisors, and employees, and is most apparent when direct task simulations (DTS) are used as selection tests in developing PESs. As such, reliability and validity theory are interconnected research areas, and quantities derived in the former bound or limit the inferences in the latter, i.e., reliability is integral to validity in that a selection test or PES cannot be considered valid if it is not reliable. Specifically, reliability is a question of data quality, whereas validity is a question of inferential quality ( Zumbo and Rupp 2004). This issue is better cast as one of measurement precision so that one strives to have as little error as possible in measurement and inference.

There has been some debate in the statistical psychometric literature as to whether reliability is a necessary but not sufficient condition for validity ( Zumbo 2007). Likewise, contemporary validity theory takes an integrative view of validity and reliability evidence. Although it is a concept not widely used in the PES research literature, where possible, the terminology presented in the Test Standards will be used. For example, the Test Standards ( American Educational Research Association, American Psychological Association, and National Council on Measurement in Education (AERA)1999, 2014) applied within the educational system describe 5 sources of validity evidence based on ( i) test content, ( ii) response processes, ( iii) internal structure, ( iv) relations to other variables, and ( v) the consequences of testing.

Contemporary viewpoints on validity have emphasized “sources of validity evidence” rather than “types of validity”, thereby focusing on the evidence for validity rather than simply its classification.