Professor Haiyan Liu’s research centers broadly on statistical modeling of psychological and behavioral data. She is interested in developing and applying modern statistical approaches to address issues arising in the analysis of messy data such as high-dimensional data, longitudinal/time series data, and categorical data. Her recent research includes the structural equation modeling of social networks. This line of research bridges human behaviors and networks. She also conducts methodological research on the Bayesian inference of structural equation modeling. Some other effort in her research includes calibrating the impact of the dishonest responses in survey data, study designs, and power analysis. The core of Professor Liu’s research is to use statistics to make a more proper understanding of human behaviors.