Resumen
Researchers? ability to draw inferences from their empirical work hinges on the degree of measurement error. The literature in Information Systems and other behavioural disciplines describes a plethora of sources of error. While it helps researchers deal with them when taking specific steps in the measurement process, like modelling constructs, developing instruments, collecting data, and analysing data, it does not provide an overall guide to help them prevent and deal with measurement error. This paper presents a synthesis of the insights in the literature through a decomposition of the logic of measurement. It shows how researchers can classify sources of error, evaluate their impact, and refine their measurement plans, in terms of specific steps or overall measurement approaches. We hope this will aid researchers in anticipating, avoiding, and alleviating error in measurement, and in drawing valid research conclusions.