There is great interest in magnesium and magnesium alloys for use as structural materials due to their high strength-to-weight ratio compared to aluminum and steel. However, the relationship between the macroscopic failure behavior of magnesium and the micro- and mesoscopic deformation mechanisms in magnesium are relatively poorly understood. Basal slip is ubiquitous in magnesium deformation, but the factors that determine the secondary deformation mechanisms, non-basal slip and twinning, are less well studied. For this reason, the aim of this study was to identify the relationship between twin occurrence and grain size. Towards this goal, a combination of electron backscatter diffraction (EBSD), scanning electron microscope digital image correlation (SEM-DIC) were used to capture full-field strain maps of a ~1mm x 1mm field of view on a pure magnesium dogbone sample under monotonic compression. It was found that the likelihood that a grain exhibits twinning increases as the size of the grain increases. K-means clustering, filtered cross-correlation, k-means clustering, and an AlexNet convolutional neural network (CNN) were also used to attempt to automate the twin identification process. Though the results of this work exhibited the same trend (increased twin likelihood with increased grain size), the precision, sensitivity, and accuracy of these methods was analyzed and it was found that more work needs to be done before significant conclusions can be drawn from their results.