Confirmatory factor analysis is a way to check how well a set of items or questions relate to each other. This could be used to help figure out how different groups of people answer questions, or to see if a survey has been designed to measure specific topics.
Let's say we design a survey with questions about kindness and generosity. We want to make sure that the questions we ask actually measure those topics. We could do a confirmatory factor analysis to see how closely the answers to each question are related to each other. If questions about kindness and generosity are connected and the answers to those questions are similar, then it means that our questions are doing a good job of measuring those topics.