There are many different formats for establishing an evaluation committee for PhD students. Some include the presence of the mentor, while others explicitly exclude the mentors presence. What is considered best practice and why? Does the presence or absence of the mentor play any meaningful role?