Repeatability

If we collect a set of measurements in an experiment, one of the things we need to consider is how repeatable our measurements are.

Repeatability is essentially how close our measurements are when we repeat the experiment under the same conditions. That’s not repeating measurements in the same experiment, but rather repeating the entire experiment multiple times to see if we get consistent results.

Example of repeatability

For example, let’s say we want to measure the boiling point of water. We set up our apparatus and measure the temperature at which the water boils. We repeat the entire experiment three times, and we get the following results:

• Experiment 1: 100.2°C
• Experiment 2: 100.5°C
• Experiment 3: 100.3°C

Our experiments are very repeatable because the results are all very close to each other.

If instead we got these results, our experiments would not be very repeatable:

• Experiment 1: 100.2°C
• Experiment 2: 107.5°C
• Experiment 3: 94.3°C

Improving repeatability

To make sure our repeatability is as good as possible, we can:

• Use precise measuring apparatus
• Control the environment (aka control variables)
• Do exactly the same method each time
• Take multiple measurements in each experiment and average them

Repeatability vs reproducibility

• Repeatability measures: if we do the same experiment multiple times under the same conditions, do we get similar results?
• Reproducibility measures: if someone else does the same experiment under different conditions (e.g. different lab, different equipment), do they get similar results?

See more about reproducibility here.