In traditional scientific publishing, studies are designed, conducted and written-up, then submitted to a conference or journal for peer review. This means that any feedback from reviewers on the design of the study is too late to be acted upon. It also means that reviewers cannot detect questionable practices such as bending of the written-up study aims to match the found data, or modification of the analysis to produce a statistically significant result.

Registered Reports are a new kind of scientific publication process where the study design is written-up and peer reviewed before the study is conducted. This way, changes to data collection or procedure can still be made in response to reviewer comments, and the analysis has been stated on the record. The accept/reject decision is also made in principle at this stage, which avoids the decision being based on whether results are significant and can increase researcher confidence that the work will be published. At the time of writing, Registered Reports are available at the Computer Science Education journal and are being added at ACM Transactions on Computer Education. If not publishing at these venues, preregistration is a voluntary alternative where the design is published at a trusted third-party site like the Open Science Foundation (with an embargo), but without the peer review and early accept/reject.

This birds of a feather session will bring together researchers with an interest in registering their study, reviewing such studies, or who have questions or interest in this new process.