Coming up with the right sample size for anthropometric data gathering is a balancing act. You want to measure enough people to capture all the variability needed to design your product, but you don’t want to waste time and money measuring people you don’t need. Selecting the appropriate sample size is a fundamental step in determining the anthropometric variability in a population, so it is important that it is done right.
We have found that there are three key components that can help establish your target sample size.
1. Know how variable the population is that you want to measure.
People often incorrectly think that sample size is related to population size, so they assume that for a very large population you would need to measure many, many people. But think of it this way – if the population of China were all exactly the same, you’d need to measure only one person! Instead, the important issue for sample size is variability.
For a random sample, people will show up to be measured in approximately the same proportions that they are in the population. As a result, there will be more people with typical measurements showing up more often, and more unusual people (very short or very tall, for example) showing up less often. So, if the population is quite variable, you will need to measure more of them to make sure you capture the atypical as well as the typical person. For design purposes, we are often more interested in those ends of the distribution – the small end and the large end of the bell curve – because if we can accommodate those, we will also accommodate the folks in the middle. If you haven’t captured those more extreme individuals, you risk making your doorway too short, or your airplane seat too narrow.
2. Know how precise the population statistics need to be.
The reason we’re measuring a number of people is that we will calculate a set of statistics to characterize the population. Our clients then use those statistics in their designs. The product design could be anything from a chemical protective mask to a T-shirt to a jet engine cockpit. The level of precision needed for each type of design differs. Safety-critical products typically require greater precision in the statistics than, for example, a clothing item. We would want to estimate population statistics to the nearest millimeter for a gas mask, for example, since a poorly fitting design can have dire consequences. For that T-shirt, though, a half-inch might be close enough. Knowing how precise you need to be will ensure time and resources are spent wisely.
3. Know exactly how confident you must be in the results.
Just as you need to know how precise your measurements must be, it is also helpful to know how much confidence you can place in the results. Do you need to be 99 percent sure you have the right average sleeve length, or do you need to be 80 percent confident? For those working on projects relevant to safety, a higher vote of confidence is usually needed. On the other hand, projects where safety is not an issue, or where consumers have many marketplace options, may require less confidence.
Keeping these key factors in mind, we can successfully calculate sample sizes that are large enough to capture the required variability, with appropriate precision and confidence, but without wasteful oversampling. Interested in learning more? Let us know how we can help you!
