Abstract
Training programs for statisticians and data scientists in healthcare should give greater importance to fostering inductive reasoning toward developing a mindset for optimizing Big Data. This can complement the current predominant focus on the hypothetico-deductive reasoning model, and is theoretically supported by the constructivist philosophy and Gestalt theory. Big-Data analytics is primarily exploratory in nature, aimed at discovery and innovation, and this requires fluid or inductive reasoning, which can be facilitated by epidemiological concepts (taxonomic and causal) as intuitive theories. Pedagogical strategies such as problem-based learning (PBL) and cooperative learning can be effective in this regard. Empirical research is required to ascertain instructors’ and practitioners’ perceptions about the role of inductive reasoning in Big-Data analytics, what constitutes effective pedagogy, and how core epidemiological concepts interact with the evidence from Big Data to produce outcomes. Together these can support the development of guidelines for an effective integrated curriculum for the training of statisticians and data scientists First published February 2020 at Statistics Education Research Journal Archives