Description:

Measurement

In both the theory and practice of statistics, measurement has often been viewed as an afterthought. We as statisticians often blindly accept that “y” is measured with precision and immediately concern ourselves with problems of prediction, interval estimation and hypothesis testing. More careful inspection often reveals that the quality of measurement is so poor that it limits more traditional statistical practice. Measurement is a fundamental piece of the statistical process and in many cases the statistical foundation of the measurements are poorly understood and limit statistical inference. In this talk, I review some of the principles of scientific measurement in the physical, biological and social sciences and discuss their statistical underpinnings. In particular, I compare and contrast measurement in analytical chemistry and mental health and highlight opportunities for statistical research and practice.