Imagine hiring a senior manager, admitting students to a study program, making medical diagnoses, or deciding to release a prisoner on parole. In these situations, decision makers use multiple pieces of information (e.g., behavioral observations, test scores) to reach a decision. Most often, decision makers integrate information using their judgment. Yet, decades of research have produced the robust finding that more accurate decisions are made when information is integrated using simple algorithms. Unfortunately, decision makers rarely use algorithms in practice. Many people still strongly believe in the value of expert judgment. How can we increase algorithm use in practice and hence improve decision accuracy without losing the acceptance of key stakeholders? Do experts make more accurate decisions than novices? In this talk, I will discuss how decision making can be improved in practice and introduce the lens model as an overarching framework for studying (interventions to improve) decision making. Furthermore, I will present some preliminary data from an ongoing project and discuss future research ideas.

Marvin Neumann is an assistant professor in the Organizational Psychology section at the Vrije Universiteit Amsterdam. His research focuses on improving test use and decision making in selection, algorithm aversion, and the scientist-practitioner gap.