SINCE AN EVALUATION DESIGN INVOLVES A PLAN FOR A SET OF SYSTEMATIC MEASUREMENTS, CONSIDERATION MUST BE GIVEN TO WHAT MEASUREMENTS ARE TO BE MADE AND TO WHAT ANALYSES AND COMPARISONS ARE TO BE PERFORMED UPON THE DATA OBTAINED FROM THOSE MEASUREMENTS. IN ORDER TO ANSWER THESE QUESTIONS ABOUT A REAL PROCESS, IT IS FREQUENTLY NECESSARY TO KEEP TRACK OF INTERRELATIONSHIPS BETWEEN MEASUREMENTS (TO EACH OTHER OVER TIME, IN RELATION TO OTHER PARTS OF THE PROCESS, IN RELATION TO FACTORS OUTSIDE THE PROCESS, ETC.). BOTH IN ORDER TO DECIDE HOW TO DO THIS AND IN ORDER TO EXPLAIN TO OTHERS WHAT IS TO BE DONE, A MEASUREMENT MODEL THAT REPRESENTS THE IMPORTANT, INTERRELATED CHARACTERISTICS OF REALITY-THOSE THAT WILL BE CONSIDERED--IS FREQUENTLY NECESSARY. THE MEASUREMENT MODEL IS AN ATTEMPT TO COMPACTLY DISPLAY INTERRELATIONSHIP AND MEASUREMENT POINTS. THIS PAPER ILLUSTRATES THESE MEASUREMENT MODELS USING AN EXAMPLE DRAWN FROM THE HOME HEATING SYSTEM; ADDITIONAL EXAMPLES DRAWN FROM GARBAGE TRANSFER AND KNOWLEDGE TRANSFER ARE ALSO PROVIDED. THE 'LOCATION' AND PERSPECTIVE OF THE OBSERVER CONSTRUCTING THE MODEL IS FOUND TO HAVE AN IMPORTANT EFFECT; THIS INFLUENCE IS BRIEFLY EXAMINED.