Abstract
Differences between controlled laboratory experimentation and measurement of personnel performance in functioning man-machine systems are contrasted. Types of non-laboratory performance measurement are described.