Part 2: So you don't think your graduate student is working out, now what?
***DISCLAIMER: For the record, I am not a faculty member, so this post is an opinion based on my years working in industry.***
Not every person you hire is going to be a great fit with your company or your brand. It happens. Some people look great on paper when they aren't. Some people are really nice and would be a great fit in some one else's company, but they just don't fit with yours. The reasons people get fired run the gambit from personal to professional reasons depending on federal, state, and local laws. The real question is: what do you do when things aren't working out?
The first thing you need to do is stay calm. The graduate student-advisor relationship can be stressful on both sides. You've invested in this person and things aren't going as planned. It can be every bit as stressful as being a parent. This is why calm has to be the first order of the day. You need to separate your feelings about the person from his or her performance. Sometimes writing things down helps clarify things. If nothing else, it will help you clarify whether the issues can be resolved.
Secondly, let your student know early on in the process where they stand. Most graduate students want to be successful. Letting them know where they are doing well and where they need to improve can be beneficial. My current advisors are really amazing in giving us yearly evaluations and usually we have an end of semester sit-down where we talk about what worked, what didn't, and benchmark goals for the next semester. When things go poorly, we all talk about it right away and get things back on track. When they go really well, we celebrate and I walk away feeling like I can conquer the next step we've laid out. Under no circumstances should you just pretend your student doesn't exist until you fire them. It's a waste of your money and both parties time.
This leads into my third point: Keep it professional. At some level, graduate school is an apprenticeship and the graduate student does need to be evaluated. This is an evaluation of his/her performance, not who they are as a human being. Personal attacks and discussion of personal life (except where it bridges into professional life) are off-limits. It doesn't matter if you think your graduate student is a horrible excuse for a human being or whom your graduate student is dating/sleeping with, unless it is relevant to their position as a graduate student or TA.
If you have a problem with your student, talk to your student about it. I knew a faculty member who had huge divisions in his lab group because he would tell his favorite students all about the students he didn't like as well. These favorite students would then tell other people and it would eventually get back to the student who was the subject of the gossip. I told my current advisors when they hired me that I expected that if they had a problem with me, I would hear about it from them directly and not anyone else. I respect them because they have always come to me directly when I've made a mistake.
Most importantly, providing feedback early and often is really positive for you and for your graduate student. When I failed my comps the first time due to anxiety issues, my advisors told me to write them a one page corrective action plan that I would follow over the next semester to address the issue. It was enormously helpful, because I had an opportunity to figure out what went wrong and how to fix things. I laid out goals, including mini-comps with my advisors, that helped me get used to speaking in front of people. I got help from people on campus as well. I still was nervous the second time around, but I was able to hold it together enough to pass and I knew that I had done everything I could to fix my problem. If I hadn't, well, they could have shown me my plan, how I didn't follow it, and I wouldn't have been able to blame them if they booted me. I think this is the way things should work.
I am stressed out like most graduate students, but I admit that having advisors who run the personnel end of the lab like a company has really been a blessing. I feel more confident because I know where I stand. When I got booted from my previous PhD attempt, it was a bolt out of the blue. I never knew my advisor was unhappy with me until it was too late to do anything about it. I really don't think I have become more brilliant in the years between being kicked out of school and coming back. I think having great advisors who understand the mentoring process was what I needed to be successful. They aren't perfect and there are things I would do differently, but they have been just great. I also have a great committee who has been there every step of the way with me too.
Of course there are exceptions to these guidelines, like academic misconduct. That is a whole other ball of wax. Most universities have defined rules on how such incidents are handled. This is why I did not touch on such behaviors. I feel like the guidelines I have written are more for instances that are more routine such as when your student is slacking, or has an issue like anxiety, etc.