Madhu Viswanathan, Measurement Error and Research Design

At the core of the scientific method is the need for well-designed research studies that enable the researcher to make reasonable and valid conclusions about how the world works. Any researcher is likely to admit that no study is perfect: one must always contend with error. This error can arise from a number of sources, including but not limited to, incorrect or incomplete definitions of the constructs being measured, methodological flaws that threaten internal validity, and poorly constructed methods of assessment. Error compromises the researcher’s ability to draw clean and accurate inferences from his or her data, and therefore it is critical that the researcher is able to recognize, evaluate, and remediate potential sources of error. As such, measurement error and research design are two inextricably linked concepts that are vital to conducting sound research. …