Experiment!: Planning, Implementing and Interpreting

11.3 Developing a Measurement System

Experiments are devised to answer unique questions. For this reason, experimenters must often develop unique setups and unique measurement systems to obtain relevant data. It is of course difficult to give general advice that is useful in all possible situations across several disciplines, but there are a number of general problems that everyone setting up a measurement system has to consider.

Firstly, it is important to note that the measurement system includes the measuring instrument, the measurement object and the operator, and so is very much part of the experiment itself. In many of the previous examples in this book we have seen that the measurement system is co-developed with the experiment. In Chapter 6 we saw how Millikan developed the balanced drop method by trying to avoid the uncertainties inherent in the charged cloud method. In Experiment 1 in the last chapter the wooden arena for the dung beetles was developed to avoid the beetles using the camera stand for orientation. In Chapter 4 we saw how central the details of the measurements were to the successful implementation of Galileo's experiment on the inclined plane. The method for measuring time did not only have to be accurate in itself but also practically suited to the experiment. Keeping an internal rhythm by singing a song actually provided better time measurements than using the swings of a “sophisticated” pendulum, which would require the coordination of two external cues. ...