pre-registration

When performing academic research that involves some form of formal experiment or study, it is good practice to preregister this in advance of running it. This is a way to avoid the file drawer effect by recording a statement before starting an experiment to specify the hypothesis being tested and which statistical technique will be used. This forces justification of the reasons for any post-hoc hypothesis. It is standard practice in many areas of science, and becoming so in HCI. Preregistration usually involves some form of public register of experiments, typically through a web interface, which enables you to record what is being planned, the proposed methods, and analysis techniques.