Collective cell migration underpins a considerable range of processes, including embryo development, cancer invasion and wound healing. This is driven in large part by mechanical interactions such as cell-cell and cell-substrate adhesion. Interestingly, it has been observed experimentally that the coordinated nature of collective cell migration can give rise to strong correlations in cell positions and velocities and these can extend over multiple cell diameters. In the first part of the talk, I will introduce an individual cell based model (IBM), which can accurately capture the appearance of these correlations. However, an IBM-based approach is limited in that such models are computationally expensive and difficult to analyse mathematically. We therefore derive a continuum approximation of the IBM, which is both accurate and is amenable to rapid computational solution. In the second part of the talk, I will discuss the application of this model to real data obtained from our experimental partners in the DKFZ. In particular, we investigate the migratory properties of lung cancer cells when subjected to different hormone treatments and signalling inhibitors. Using particle image velocimetry (PIV) we obtain quantitative information, such as the velocity correlation distance or spatiotemporal changes in the underlying cell speed distributions. By fitting the model to the PIV data, we are trying to understand how the mechanical properties of individual cells are being affected by the treatments, which ultimately rise to the observed alterations in collective cell migration.