How to Survive and Finish Your PhD in Social Science: An Honest Supervisor’s Guide
If you are reading this in the middle of your PhD, wondering whether you are cut out for it, let me tell you the first and most important thing I tell my own scholars: the doubt you are feeling is not a sign that you are failing. It is a sign that you are doing a PhD. Nearly everyone worth their doctorate has sat where you are sitting, quietly convinced they were the one who would not make it. I have supervised scholars through exactly that feeling, and I want to offer you the honest guidance I give them — not the glossy version, the real one.
A doctorate in the social sciences is a long, largely solitary undertaking, and the numbers are sobering: completion rates in behavioural and social science programmes often sit below half (American Academy of Arts & Sciences, 2026). But here is what those numbers hide — most scholars who do not finish are not the least capable. They are the ones who ran out of support, structure or belief. This guide is about protecting all three.
1. Understand the shape of the journey
A social science PhD commonly takes six to ten years in many systems — longer than most STEM fields, partly because so much of it is solitary writing (Wally Boston, 2026). Knowing this in advance matters, because unrealistic expectations are a quiet destroyer of morale. You are not slow; the work is genuinely long. Picture it in phases — coursework and proposal, fieldwork and data, then the long analysis-and-writing stretch — and expect each to feel different. The scholar who plans for a marathon paces themselves; the one who expects a sprint burns out in year two.
2. Protect yourself in the “valley” — the dissertation-writing phase
Ask any supervisor where scholars are most at risk, and most will say the same thing: the long middle-to-late stretch of writing, after the excitement of data collection has faded and before the finish line is in sight. It is isolating, unstructured, and easy to drift in. This is where “all but dissertation” happens — brilliant people who did everything except finish.
The antidote is not heroic willpower; it is structure and company. Write regularly in small, protected blocks rather than waiting for rare marathon days. Set tiny, visible milestones. And do not do it alone — a writing group, a peer check-in, a weekly coffee with another scholar can be the difference between finishing and fading. Isolation is the enemy; community is the cure.
3. Manage your supervisor — yes, manage
Difficulties with supervisors are among the most-cited reasons scholars struggle, and yet the relationship is far more within your influence than you think. Treat it as the professional working partnership it is:
- Agree expectations early — how often you will meet, how quickly they return drafts, what they want to see and when.
- Come prepared — send work in advance, bring an agenda and specific questions. A focused hour beats three vague ones.
- Be honest when you are stuck — supervisors cannot help with problems we do not know about, and “I’m struggling” is a strength, not a confession.
Remember: a good supervisor genuinely wants you to finish. Your success is our success. Make it easy for us to help you.
4. Do the craft well — it saves you time later
Much of the pain of a PhD comes from foundations that were rushed. A muddled research question, a thin literature review, a shaky methodology — each comes back to haunt you in the viva. Investing in the craft early is not perfectionism; it is efficiency. If you want the practical building blocks, I have written detailed guides on the pieces scholars most often struggle with:
- Writing sharp research aims and objectives — the backbone of the whole thesis.
- A literature review that argues, not just summarises.
- A methodology chapter that gets approved first time.
- A discussion chapter that interprets rather than repeats.
5. Guard your wellbeing — it is not optional
I want to be direct about this, because too many supervisors are not. The doctoral journey takes a real toll: burnout, anxiety and low mood are common companions, especially in the isolating phases. Your wellbeing is not a distraction from your PhD — it is the engine that will get you through it. Keep parts of your life that have nothing to do with your research. Protect your sleep, your relationships, your body. And if you find yourself persistently struggling, please treat that as seriously as you would a broken methodology: reach out to your university’s counselling services, your supervisor, or people you trust. Needing support is not weakness. Finishing a PhD is, among other things, an act of stamina — and you cannot sustain stamina on an empty tank.
6. Start publishing before you finish
One of the kindest things you can do for your future self is to publish along the way, not only at the end. Turning a chapter or a sub-study into a journal article builds your CV, sharpens your argument, and — crucially — gives you feedback and momentum when motivation is low. It also makes the final thesis easier, because parts of it are already tested. If that feels daunting, my guides on publishing in Q1 and Q2 journals and responding to reviewer comments walk you through it step by step.
7. Remember why you started
Somewhere under the deadlines and the revisions is the reason you began — a question that genuinely mattered to you, a problem you wanted to help solve, a community whose story you wanted told. In the hardest months, that “why” is not sentimental; it is fuel. Keep it visible. Write it on the first page of your notebook. A PhD in the social sciences is, ultimately, an act of service to knowledge and to the people your work touches. That purpose is what carries scholars across the line when technique alone runs out.
Frequently asked questions
Why do so many PhD students not finish?
Social science completion is often below half — usually because of isolation and burnout, difficult supervisor relationships, financial strain, and losing direction while writing, rather than lack of ability. Most of these are manageable with support and structure.
How long does a PhD in social science take?
Commonly six to ten years in many systems — longer than most STEM fields, largely due to the solitary writing phase. Planning for that length protects your motivation.
How do I manage my relationship with my PhD supervisor?
Agree expectations early, come to meetings prepared with an agenda and questions, send work ahead, and be honest when stuck. A good supervisor wants you to finish.
What is the hardest part of a PhD?
Usually the long, isolating dissertation-writing stretch after data collection. Steady routines, community and small milestones help more than heroic bursts.
Should I publish during my PhD?
Yes, where you can. Publishing chapters as articles builds your CV, sharpens your work, and provides momentum and feedback along the way.
A final word from a supervisor
If I could put one thing on a poster in every doctoral office, it would be this: the scholars who finish are rarely the most brilliant — they are the ones who kept going, and who let themselves be supported. Your PhD will test your intellect, but it will test your resilience more. Build your foundations well, guard your wellbeing fiercely, manage the relationships around you, and keep your purpose in sight. You do not have to be the smartest person in the room. You have to be the one who stays in it. And if no one has told you lately: the work you are doing matters, and so do you.
Need guidance on your proposal, methodology, analysis or publishing strategy? Supporting scholars through exactly these stages is part of what I do — see the services page.
References
American Academy of Arts & Sciences. (2026). Attrition in humanities doctorate programs. Humanities Indicators. https://www.amacad.org/humanities-indicators/higher-education/attrition-humanities-doctorate-programs
Boston, W. (2026). Exploring PhD completion rates: How many finish on time? https://wallyboston.com/phd-completion-rates/
Council of Graduate Schools. (2026). Ph.D. completion and attrition: Analysis of baseline program data. https://cgsnet.org/