ABSTRACT
In the context of measurement, validity refers to the extent an instrument actually measures what it is supposed to measure. The word instrument is used in its broad sense here. It includes not just the physical measurement tool but also how the tool is used. An imaging technology may be supposed to measure the body surface area of a person. If the technology fails in distinguishing the person from the background, it ends up measuring the surface of a mixture of both the person and the background objects. School examinations may be supposed to measure knowledge and competency. But a concern is that such examinations may end up measuring short-term memory. The measurement of human growth tends to be more straightforward and involves less concern about validity than the measurement of development, which is relatively subjective and difficult to define to begin with. Reliability refers to the degree of consistency between multiple measurements of the same target. In the presence of systematic or random measurement errors, they disagree. Technically speaking, reliability is the ratio of the variance of the true values to the variance of the observed values. The observed values consist of the true values plus errors.
