Chapter 7 introduces models as concise quantitative accounts of patterns and relationships between different components of natural ‘systems’. Model development is broken down into the four stages of observation, measurement, experimentation and theory development. Models uncover statistical relationships between coastal structure (response) variables, and function (driving) variables. Structure refers to the observable features of coastal landscapes e.g. the sizes, shapes and configurations of components. Function refers to the interactions between these features, e.g. flows of materials or organisms between them. Explanatory models construct an explanation for an observed pattern using empirical data, while predictive models generate a new observation by applying a statistical model to data. Statistical inference enables statements to be made about large populations of data samples on the basis smaller samples. Regression is a statistical technique for measuring the relationship between two or more variables. Where each data point is a location, a geographically weighted regression can utilise the spatial structure of the process of interest to improve the relationship modelled.