![](/img/g/5/g5qsfh6m.png)
Technical Guidance for Calculating Scope 3 Emissions [154]
APPENDIX A Sampling
Sampling methods
Sampling methods available to companies include, but are not limited to:
• Simple random sampling
• Systematic sampling
• Stratied sampling
Each approach is summarized below. Alternative methods for sampling may also be used.
Simple Random Sampling
Simple random sampling involves randomly selecting activities (i.e., a sample) from a larger set of activities (i.e., the
entire population).
If the total number of activities from which a sample is selected is small, simple random sampling may be performed at
its most basic level by selecting activities at random. If the total number of activities is large, for example with hundreds
or thousands of activities, then random sampling is better performed by computer.
Advantages of simple random sampling include:
• With an appropriate sample size, simple random sampling creates a representative view of the entire population.
(For example, if a company has fty employees located within a close geographical area and wants to determine
the average commuting distance, it may choose to collect data from ten randomly selected employees as a
representative sample.)
• It is relatively straightforward to construct the sample.
Disadvantages of simple random sampling include:
• The sample size needed to generate appropriately representative results may be prohibitively large and
cumbersome to sample. (For example, if a retail organization has thousands of stores in many countries, randomly
selecting individual stores may result in a dicult and time-consuming data collection process.)
• It may not be possible to obtain a complete list of all activities from the sample size, which is a prerequisite for
simple random sampling. (For example, if a distribution company wants to determine the average backhaul capacity
of its trucks, it would have to list every journey before a random sample could be selected.)
Systematic Sampling
Systematic sampling involves randomly selecting the rst item to sample and then selecting subsequent activities at
regular intervals.
An appropriate sampling interval should be chosen so that the company achieves the desired sample size. For example,
if a company sourced agricultural products from 100 farms but only wanted to sample 20 farms, an appropriate
sampling interval would be every 5 farms. If the rst farm to be sampled was picked as Farm 3, the company would
subsequently sample from Farms 8, 13, 18, 23,…, 93, 98.