This chapter discusses the principled distinction of whether the test score is valid or whether the test is valid; that is, the measurement value or the measurement instrument. Both conceptions date back to the 1920s but through the years the test-score interpretation of validity gained the upper hand. Modern validity theorists, although still a minority, have criticized the prevailing test-score validity concept and emphasized the need to found the construction of a measurement instrument on substantive theory of the attribute for which the instrument is constructed. A sound theoretical foundation often is absent or only immature and in need of further development. I argue that test construction that is not based on a sound theoretical foundation is unlikely to contribute to the theory development for the attribute the test intends to measure. Hence, a sound theoretical basis for test construction is necessary for the test to contribute to the advancement of psychology as an empirical science. This chapter was based on a paper presented in the Invited Symposium “Metaphors and Measurement: An Invited Symposium on Validity”, at the International Meeting of the Psychometric Society 2012, July 9–12, 2012, in Lincoln, NE, USA.
CITATION STYLE
Sijtsma, K. (2013). Theory development as a precursor for test validity. In Springer Proceedings in Mathematics and Statistics (Vol. 66, pp. 267–274). Springer New York LLC. https://doi.org/10.1007/978-1-4614-9348-8_17
Mendeley helps you to discover research relevant for your work.