ABSTRACT
Measurement of psychological attributes, such as intelligence or personality traits, is paramount to relating attributes to one another to test hypotheses about human activity and possibly find support for or reject theories. Psychological measurement is problematic, because the theories about the attributes usually are not developed well enough to allow the derivation of unambiguous guidelines for designing a valid measurement procedure. Together with other error sources beyond the researcher’s control, incomplete attribute theory causes data collected with candidate measurement instruments to be rather messy and lacking full information about the attribute. Chapter 1 discusses the basics of measurement in the human sciences and the role that psychometric measurement models play in constructing an instrument—a test or a questionnaire—for measuring people on a scale. For the attribute of transitive reasoning, based on the literature, we explain how theory is a precursor of a scale for the attribute. We discuss a four-step cycle of instrument construction showing how both weak and strong theory are used to operationalize the attribute, collect data with the preliminary instrument, analyze the data using a measurement model, and feedback the results to improve the theory and therefore the measurement.
