The Composite Method Produces High False Positive Rates

Using a single pair or a small number of pairs of composite stimuli to assess the relationship between morphology and other traits is a common method in face research. Here, I use data simulation to demonstrate how this method inevitably leads to a high false positive rate, and how this problem is made worse by using a larger number of raters. I conclude by suggesting alter- native methods for assessing the relationship between face morphology and individual traits.