The uncertainty is expressed as range that indicates the precision of individual measurements. When an analog measuring device is used to collect raw data, the uncertainty is half of the minimum increment shown in the measuring device. When a digital device is used to measure raw data, the uncertainty is equal to the minimum measurement that the device can record.
Examples: The minimum increment in a ruler represents 1 mm, so the uncertainty of a ruler is ±0.5 mm or ±0.05 cm. A pH probe that measures pH with one decimal place (for example 7.2), has an uncertainty of ±0.1
Imagine that you measure something, for example the length of a plant shoot. According to you and your ruler it measures 5,3 cm. But is that really true? Does it measure exactly 5.3 cm? Can you be 100% sure that if you measure it again next week, you will record exactly the same value? And... if another person uses the same ruler to measure the same plant, are you sure that the measurement will be exactly the same? What if a different ruler is used to measure the same plant shoot, will the result be exactly 5.3 cm?
In science, every time we measure something there is some degree of uncertainty. This mainly depends on the measuring tool that we use. Uncertainty must be considered for every measurement taken. Different methods are used to calculate the uncertainty in analog and digital measuring devices.
Analog devices use a continuous range of values to represent the information measured. When using an apparatus with regular increments (divisions), the uncertainty is half of the smallest division/increment. Why? Because if a measurement falls in between two increments, you can only measure it accurately assuming ±0.5 of the smallest increment. How long is the grey bar in the images below?
Image source: Gross. “MEASUREMENT AND ERROR CHARACTERISTICS OF MEASURING INSTRUMENTS ANALYSIS OF MEASURED DATA UNCERTAINTY ANALYSIS.” Slideplayer.Com, 2015, slideplayer.com/slide/10386227/. Accessed 19 Aug. 2019.
The calculation of the measurement uncertainty for each measuring device should be explained and the possible effect of each of these uncertainties in the data collected should be discussed.
Often, the uncertainty of a device is provided by the manufacturer of the device. If that is the case, the manufacturer's site must be cited/referenced. If not, when an analog measuring devices is used to collect raw data, the uncertainty is half of the minimum increment shown in the measuring device. When a digital device is used to measure raw data, the uncertainty is equal to the minimum measurement that the device can record. The method used to calculate each measurement uncertainty must be stated.
State the method used to calculate the uncertainty, given the type of device or provide a reference to the source in which the uncertainty is published
Analog devices use discrete or discontinuous values to represent the information measured. When using a digital apparatus, the uncertainty should be ±1 for the last measurable digit. In the picture below, what is the mass of the avocado?
Image source: Gasparini, John, and We’ll. “The Data Collection and Statistical Analysis in IB Biology John Gasparini The Munich International School Part I – Introduction and Uncertainty in Measurement.” Slideplayer.Com, 2019, slideplayer.com/slide/10177049/. Accessed 19 Aug. 2019.
The measurement uncertainty is a range and it’s always expressed as ± a value. Differences within the uncertainty range are not actual measured differences, but differences due to the device’s “ability to measure”. Differences larger than the uncertainty range (outside the uncertainty range) are measurements that the device has detected as different . That means that the device, with its “possibilities” (its uncertainty), has been able to detect differences between two trials or between treatments. However if two trials have different numbers within the uncertainty value, those values are not different measurements. For example if the uncertainty of a device is ±0.1(whatever unit the device measures) the values 34.1 and 34.2 are NOT different values, because the device can’t actually measure differences between them. If the uncertainty of a device is even larger, like 0.5, then the range is aso larger and in that case, 34,5 and 35.9 are NOT different measurements.
The possible impact of the measurement uncertainty in an experiment must be considered for ALL variables. For example, the uncertainty in the measurement of temperature may explain some variability in the data. In controlled variables, a large uncertainty may have an effect in the data and may affect the reliability and repeatability of the experiment.