Guidelines for co-authorship

These are our guidelines for decisions concerning co-authorship on scientific papers where statisticians have been involved in advice, planning or analysis. You will hardly find them unreasonable and we rarely have problems in this respect.

The basic rule is that if the statistician contributes in any substantial way to a publication, with advice, planning, or analysis, then he or she is entitled to be a co-author (if his or her material is being used), and in that case the statistician also has a responsibility for the publication as a whole.

The interpretation is a matter of judgement, but usually advice during a couple of meetings is below the limit for becoming a co-author, whereas the statistician should usually be a co-author if he or she performs analyses, for example.

Note that a first meeting with us never obligates you to share the authorship, unless that it agreed at the meeting.

Author: ims@kvl.dk (Ib Skovgaard) 1997-07-07