The importance of using multiple outcome measures in infant research

Abstract

Collecting data with infants is notoriously difficult. As a result, many of our studies consist of small samples, with only a single measure, in a single age group, at a single time point. With renewed calls for greater academic rigor in data collection practices, using multiple outcome measures in infant research is one way to increase rigor, and, at the same time, enable us to more accurately interpret our data. Here, we illustrate the importance of using multiple measures in psychological research with examples from our own work on rapid threat detection and from the broader infancy literature. First, we describe our initial studies using a single outcome measure, and how this strategy caused us to nearly miss a rich and complex story about attention biases for threat and their development. We demonstrate how using converging measures can help researchers make inferences about infant behavior, and how using additional measures allows us to more deeply examine the mechanisms that drive developmental change. Finally, we provide practical and statistical recommendations for how researchers can use multiple measures in future work.

Figure 1.

A one-level (left) and two-level (right) conceptualization of the relationship between a psychological construct (in this case emotion) and outcome measures