In recent years, there have been multiple instances of misconduct in science, yet no coherent framework exists for characterizing this phenomenon. The thesis of this article is that economic analysis can provide such a framework. Economic analysis leads to two categories of misconduct: replication failure and fraud. Replication failure can be understood as the scientist making optimal use of time in a professional environment where innovation is emphasized rather than replication. Fraud can be depicted as a deliberate gamble under conditions of uncertainty: The scientist takes advantage of the complexity of science and undermines the integrity of science for personal gain or advancement.
