Our second main contribution is to provide evidence from the first “repeated measures study” of competing models in repeated SSGs. Arepeated measures study is an observational research method in which data is gathered for the same sub-jects repeatedly over a period of time, sometimes spanning years or even decades. In our study, asuite of well-established models and SHARP are compared in human subjects experiments on the Amazon Mechanical Turk (AMT). In conducting these repeated measures experiments with AMT, we also provide a procedural contribution – specifically an empirically validated study of the process of conducting repeated measures experiments on AMT. We address challenges faced while conducting several 2.3 week long (on average) repeated measures studies, 46 weeks in total, with various behavioral models and show that our methods help maintain high retention rates of the participants throughout the course of the study and therefore enables valid comparison of the various models.