Join 3,557 readers in helping fund MetaFilter (Hide)


The Mid-PhD Crisis
February 16, 2014 10:17 AM   Subscribe

Hi Hivemind, I'm currently enrolled in a PhD program for statistics and operations research, and in two more years I can grab that PhD. Alternatively, I can jump ship now with a MS, headed for the (inviting/inevitable) waters of industry. Knowing that I have no interest in staying in academia, give me some motivation to finish. Or, tell me to quit because actual work experience is more valuable! What roles should I be looking at other than Data Scientist? Would it be feasible to get a position doing private research, and would that be awesome? Bonus points if you can tell me what skills I should be cultivating to be hire-able (software engineering)? P.S. I will also do some more chatting with profs and former students to figure out how I should be directing myself, but I hope the HiveMind can provide some complementary ideas.
posted by zscore to Work & Money (12 answers total) 3 users marked this as a favorite
 
In my experience , data science careers are split between the architecture, engineering and maintenance of the data systems, and the guys who make the models. The guys who make the models usually are the ones who have PhDs. As far as other careers, you could maybe look at Rand's career openings to get some ideas.
posted by BabeTheBlueOX at 10:26 AM on February 16 [1 favorite]


Ok, can't help with the career advice, but I can try with the motivation to finish. What's your project about? Do you enjoy it? Will you be excited to complete it, or alternatively sad to see it collapse if you leave? Do you like your team at the moment, is the day to day work ok? Two years is no time at all, realistically six months of that will be writing up anyway. Can you push through for another eighteen months?

It's that sort of thing that helped me over that awful "nothing's working and my project isn't really progressing" bump that everybody hits in the middle. It gets much better once you start to see a bit of progress, get a few chapters finished or papers out.
posted by tinkletown at 11:14 AM on February 16


I jumped ship on the middle of a PhD program in math. Well, truthfully, I had to walk the plank. Whatever, I found myself with a MS in math and an interest in OR. I got a job with a company that did work for the Navy, then went to Pepsico to do management science. That worked until the invention of the PC and spreadsheets. At that point, the MBAs didn't feel they needed help, and the dept was disbanded. So, it was a long time ago, but I have feeling for where you are.

My view is that, outside the academy, OR is delivered in the form of software. You might find a job working with an oil company on optimizing flows, but more likely you will be building a tool that is more clever than the users realize. I did a tool that searched all the possible mortgage variants of a bank to see which would give the customer the biggest loan -- a bounded, discrete optimization program. Another example, not one I ever worked on, is the turn-by-turn driving direction programs.

If you want to work on these things at Google, you need the PhD, but elsewhere, maybe not. But you will need good programming skills.

THERE IS A TRAP. The market for mudane business programming is much bigger than the market for technical programming and it pays well, so it is easy to get sucked in. That's what happened to me.
posted by SemiSalt at 11:28 AM on February 16 [4 favorites]


I have a MS in OR, and now work in health economics. Basically it involves building cost benefit models of drugs or policy ideas. So in a way, it's like doing research, but it's a little less fun because somebody else defines your research questions. The flip side is that your results get put directly in front of decision makers. If that's something you're interested in learning more about, feel free to me-mail.
posted by tinymegalo at 11:49 AM on February 16


Have you considered actuarial science or logistics?
posted by carmicha at 12:56 PM on February 16


Data science or "Big Data" sure is the hottest thing right now. In some areas of industry like anything biotech, the degree will go a long way.
posted by sammyo at 5:40 PM on February 16


Is the PhD funded? Because that makes a difference. The pay differential for PhDs is shrinking, so acquiring significant additional debt to complete if you are not funded may not make financial sense. If it is funded, however, I would likely stick it out for the flexibility

If you want motivation, I offer up Bill Buxton, Microsoft's principal researcher. He utilises his many, many credentials in private industry and is lovingly known as Dr. Dr. Dr. Bill. And whatever about MS, he's an amazing man with a fierce and accessible intelligence.
posted by DarlingBri at 6:10 PM on February 16


Hey everyone, thanks for the input. I think my biggest question was if I would have a significantly different job and prospects with or without a PhD, and it seems like the majority think the answer is yes.

Also, thanks for the ideas on jobs! It's always good to know what people are out there doing!
posted by zscore at 8:55 PM on February 16


Also, DarlingBri, the program is funded, so it's just the pay-difference I'd be losing.
posted by zscore at 9:05 PM on February 16


I would definitely get the Ph.D.. I work in industry and Ph.D.s are provided many more opportunities for advancement than people with a Masters. There are people Masters degrees here with much more experience at the company than I, but I am provided more management training than they are. They seem to be stuck as technicians and such forever here. The Ph.D. automatically put me ahead even though I have little experience in the work I am doing here, which is FDA regulated medical devices. My Ph.D. is in immunology using a parasite model. Nothing in any way related to the medical devices we sell but the company is willing to train me due to my proven ability to finish a project. ABD's are often unable to sell their reasons for leaving the PhD program other than explaining that their adviser sucked (most do), family obligations (who doesn't have those, although serious illness would be a very good excuse, or something akin to that). If it's only two more years, seriously go for it and get it over with. I would tell myself that I have to work for the two years anyway, so I may as well get the Ph.D.

A Masters degree is a great accomplishment, but a Ph.D. could really help you out in the long run. You don't have to go into academia, industry is calling too.
posted by waving at 5:27 AM on February 17


If it's funded, I'd complete the PhD. You'll have more job options through your life with it, and industry isn't going anywhere.
posted by DarlingBri at 3:45 PM on February 17


Update: I dropped out and am working as a data scientist. No regrets
posted by zscore at 9:25 AM on September 5


« Older Where do you recommend to go s...   |  Text messaging consistently do... Newer »

You are not logged in, either login or create an account to post comments