Precision accuracy and uncertainty
In all experimental measurements, there is a degree of uncertainty. This is usually dependent on:
- The limitation of the measuring instrument.
- The skill of the person making the measurement.
The instruments' built-in or inherent errors are called systematic errors. If a scale is not calibrated corrected, it would yield a reading that is consistently too high or too low. When errors are introduced by the skill or ability to read the scientific instruments, this will lead to results that may be either to high or too low. These are called random errors.
Although all experimental measurements are subject to error, we can still trust our measurements in terms of their precision and accuracy.
Precision indicates the reproducibility of a measurement. That is, the closeness in agreement among the values when the same quantity is measured several times. If the series of measurements is reproducible, then good precision is obtained as careful inspection of each measurements deviates only by a small amount from the average of the series. On the other hand, if there is a wide deviation among the series of measurements the precision is poor. A measurement is said to be accurate if it is close to the known "accepted" or "most probable" value.
For example, the boiling point of pure water at sea level is 100°C. Using the same thermometer in four trials of measurements, the data collected is as follows:
Boiling point of pure water at sea level:
- 96.9°C
- 96.8°C
- 97.1°C
- 97.0°C
Since these figures show a high reproducibility, the measurements are precise. However, the values are considerably off from the accepted value of 100°C. So, the measurements are not accurate. In this set of measurements, we probably would suspect that the inaccuracy arises from a mis-calibrated thermometer.
Content suitability
BCIT courses: CHEM 0011
