An introduction to the reliability of measurements, ranging from nominal classifications, through simple ordinal rating scales to data assumed to be measured on an interval scale. The main theme of the volume is that the measurement models and the statistical methods used in the assessment of the accuracy are common to virtually all fields of scientific endeavor. However there is little communication between the fields about development of those methods. The text draws them all together and integrates the design and analysis of this type of study, including new data sets and examples. Monte Carlo simulations and the bootstrap are included.