If the goal of a study is to establish causality between sleep quality and academic performance, what kind of statistical analysis would be most appropriate?