How unwise is it to change fields/advisors late in my PhD?

Question

I’m sorry for asking a dumb question I already know the answer to, but I’m asking it because I feel desperate and need someone to talk sense into me.

Some background: I’m a fifth-year PhD student, with no real graduation in sight. I’m in one of the most packed and competitive subfields I know of (let’s call it field A), and I haven’t had much success, only a handful of small publications. To graduate my advisor tells me I need at least one or two major publications, but field A is so competitive at this point that many of these journals/conferences have nearly 10% acceptance rates.

But that’s not even the real issue. The real issue is I hate the research area I’m in. I’ve hated it since at least my third year. I thought maybe I would just push through it, but I’ve only hated it more and more and it’s come to the point where my hatred for it has become so visceral that I just don’t know how to keep pushing on. Not to mention a host of other problems - I feel like a complete outsider in my lab, and my advisor couldn’t care less if I come or go; he’s never provided any help and I’ve basically been left to fend for myself for the last 5 years.

Which brings me to the main reason for my question. This might sound strange, but since around my third year I found another research area (let’s say field B) that I’ve actually been very interested in, and on and off I’ve been reading a lot of results from this area - there were some stretches where I spent significantly more time reading results in B than in A. I know it didn’t get me closer to graduation, but field B was just so much more interesting to me. I’d have loved to do research in field B, but at the time there were good reasons why I didn’t really pursue trying to change or even contact the professor at our university who works in that area:

1. I had a decent career lined up in field A. I’ve had several internships in major industry labs, and I have an offer waiting for me if ever I graduate.
2. I may have an avenue to research that is essentially in field B. During one of these internships I met a researcher who told me if I get hired I could possibly do work related to field B on the side and maybe transfer entirely if things go well.
3. Location matters to me. There are fewer jobs in field B, but field A is booming. I’d be much more likely to find a job close to my family, and in fact, as I said in (1), I already have.
4. Timing. It was already 3 years into my PhD, I didn’t want to start over. Now it's been even more.
5. Maybe I would actually hate field B just as much when I got into it? I don't think so, but who knows?

With all that said, here’s my question: Knowing that all of these things are still just as true, and seeing as it’s my fifth year, how stupid would it be for me to try to change areas now? What barriers would I face?

I can’t imagine it would be anything other than idiotic to attempt it, assuming that I’d even be allowed to, but as I spend yet another day staring at my computer, unable to muster any willpower to work on my research, instead procrastinating with yet more papers in field B, I have to wonder if it’s all worth it? If I did switch, I’d still start from scratch and need mentorship, but I think I would have at least as much background by now as maybe a third-year PhD who had been researching field B from the beginning, maybe more.

Answer 1

Your profile does not say, where you are from. I suppose there are still areas on the globe, where academic titles matter. However, this is certainly not the case where I live. So, in order to provide useful answers, we need to obtain detailed informations.

I still don't understand your motivation!

• If you don't like working in field A, why do you consider taking a job in this field?
• And if you do it to pay the bills, start doing it now. If you're doing a great job in industry, nobody will ever let you go. However, if you underperform, no phd will safe you.

So I reckon the question you should ask yourself is, why do you believe you need a phd? What do you believe does a phd proof?

Answer 2

Exhaust any opportunity/possibility of just getting done in A before moving to B. Once you have the Ph.D., you can use it on whatever (even chemists doing engineering or physics, if they can get it done...i.e. much bigger changes of field than within parts of CompSci.)

Don't just take the dictates of your advisor as the last word. Push, argue, fight, write up a draft thesis (without "permission"), bring in the department chair, etc. effing etc. to get your butt done and out of there. A job waiting for you is strong support and will be noticed by the department chairman, etc. (Much better than their grads they have hanging around having a hard time getting hired! Looks good.)

Only abort to the other field, if a STRONG effort to get the little tinpot dictator to cut you loose has been unsuccessful. You have zero idea how long that other effort will take. And even then, the department is more likely to feel a duty to help you get out of that new one, fast, if you've already raised a ruckus about the old one (versus being perceived as just a lost soul flitting around).

Ignore any comments about how you need to suck up to the old prof for better letters and the like. That time is over. Throw down and get your ass graduated.