Abstract: Human mobility plays a major role in the spread of communicable diseases. Remote Indigenous populations in Australia tend to have a higher risk of communicable diseases, and although high levels of mobility at the household, community and regional scale are prevalent in these communities, limited data are available on the movement patterns of these populations. This project is motivated by the need to understand and model human mobility in rural regions when hampered by imperfect data, and aims to develop a robust methodology with techniques incorporated to address missing data and integrate multiple sources of data. In this study, I introduce numerous innovations to shed light on imperfect movement data. I applied Generalised Linear Mixed Effects Modelling (GLMM) for longitudinal data to understand the long- and medium-term mobility in Indigenous communities. I used Multiple Imputation by Chained Equations (MICE) to address the missing data and Two Sample Two Stage Least Squares (TSTSLS) estimation to combine two data sets with different variables. The significance of this project is that it creates new knowledge for modelling human mobility with longitudinal data and making inferences based on imperfect data from multiple sources. This knowledge could be effectively used to understand human mobility which could be applied in real world situations such as infectious disease surveillance and control.