Having the right number of people for a fit test is incredibly important to the overall success of the test. With adequate numbers of people in each size group, we can not only verify the sizing and patterning itself, but we can also identify if some sizes are not needed, saving the company or government client a significant amount of money. Also, fit tests can determine how many of each size item the client or government agency should purchase. This too can save a significant amount of money because it ensures sizes won’t run out and replacement items won’t need to be ordered in a hurry, at a premium price.
In an earlier blog post, we detailed the three factors that go into determining the right sample size for general anthropometric data collection. Today, we want to talk specifically about determining the right sample size for an anthropometric fit test.
Fit tests are different from anthropometric surveys in a number of ways, namely, for a fit test, we test the interface of a product and the person we are measuring, whereas in anthropometric surveys, we’re only interested in the person, not a product.
Special Considerations for Determining Fit Test Sample Sizes
Most anthropometric dimensions are characterized by a normal distribution – the bell curve. In that distribution, there are typically many people represented in the middle, and a few people at each end.
Of course, a product with sizes almost always needs to fit people not only in the middle, but also at the ends. So, when conducting a fit test, if we recruited people for the fit test based on the anthropometric distribution, there may be only one person who needs the size XXSmall or one person who needs the XXLarge. But, it would be foolish to draw any conclusions about the fit of the XXSmall from only a single person being measured and tested. To adequately represent each size of the product we’re testing, we would need to include proportionately more of the very large and very small people, in order to ensure there are enough people trying on those sizes, too. The end result is a fit test that includes more people and is more accurate.
Another factor we consider for anthropometric surveys is ethnic representation. In basic body sizes survey data collection, it’s very important to represent multiple ethnic groups and make sure they are present in sufficient numbers to capture the variability of that ethnic group. Ethnic representation should be a consideration in the fit test sample, but it may be less of a priority than you may think. In fit tests, it’s more important to have the size distribution right. Naturally, you won’t want to sample only one ethnic group, but in a fit test, the ethnic or age distribution is less important than the distribution of sizes.
So, What’s the Right Size?
There is no one size fits all solution for the number of people that should be included in a fit test. At Anthrotech, we’ve completed fit tests with as few as 50 people and as many as 2,200, with most fit tests landing somewhere in the middle. If your business or agency is in need of a fit test, contact us to discuss what the size and scope of your project might be.