What your supervisor really means when they say, "It's up to you"

You've prepared for the meeting. You have two possible research questions, three candidate methods, and a genuine, well-reasoned uncertainty about which combination is best. You lay them out. Your supervisor listens, nods, asks a couple of sharp questions - and then leans back and says the sentence every dissertation student comes to dread:

"Well, they'd all work. It's really up to you."

You leave the room with exactly the problem you arrived with, plus a new one: the nagging suspicion that a person who knows far more than you has just declined to help. If you have been a successful student for fifteen years - and most people who reach a dissertation have - this is a profoundly disorienting moment. Nothing in your education has prepared you for an expert who knows the answer and won't tell you.

Except, mostly, they don't know the answer. And that's the point. This article is about the transition that "It's up to you" marks: from the world of set questions and detailed briefs to the world of consequential, self-directed decisions. It looks at why the shift unsettles even very capable students, what supervisors actually mean by their maddening ambiguity, where their boundaries genuinely lie, why fear of the wrong choice is so paralysing - and what legitimate support exists for someone managing their first major independent project.

The transition nobody announces

Consider what assessment looked like for the whole of your education until now. Someone else chose the question. Someone else decided the scope, the word count, the reading list, the deadline, and - crucially - what a good answer would look like. Your job was to execute against a brief. You became extremely good at it: reading a rubric, inferring what a marker wanted, hitting the criteria. That skill got you here.

The dissertation quietly removes almost every one of those supports at once. You choose the question. You decide whether it's answerable. You select the method and defend the selection. You judge when the literature review is "enough." You decide what to do when the data disappoint you. And there is no rubric that can tell you in advance whether your specific project is a good idea, because your project has never existed before.

The American higher-education researcher Barbara Lovitts named this shift precisely in a 2005 paper whose title is worth memorising: "Being a good course-taker is not enough." Studying the transition from coursework to independent research, she found that the qualities that make students excel at taught assessment - diligence, responsiveness to instruction, accuracy against a known standard - are necessary but nowhere near sufficient for research, which additionally demands tolerance of ambiguity, judgement under uncertainty, and the willingness to commit to a direction without external validation. Students weren't failing because they had become less able. They were being asked to do a fundamentally different kind of thing, without anyone telling them the rules had changed.

This isn't just a doctoral phenomenon. Malcolm Todd, Phil Bannister and Sue Clegg's influential 2004 study of final-year undergraduate dissertation students found exactly the same double edge: students valued the autonomy, authenticity and ownership of the dissertation, and simultaneously found the independence genuinely hard - with time management and uncertainty about expectations the dominant struggles. Their conclusion was blunt: "autonomous work does not come naturally to students," and the supervisor's first task is to support the move towards it rather than assume it. Two decades later, follow-up research keeps finding the same thing: in one recent study, only 40% of students felt they had managed to build a genuine working partnership with their supervisor, even though 75% believed that partnership strongly affected their final result.

Why capable students suddenly feel like beginners

There's a specific kind of shame in being a first-class student who suddenly can't decide anything. It helps enormously to understand that this feeling isn't a personal failing - it's predicted by several well-established accounts of how people develop intellectually.

Perry's scheme: the moment authorities stop having the answer

In the 1960s, the Harvard psychologist William Perry tracked students through their degrees and mapped a developmental sequence in how they understood knowledge itself. Early on, students tend towards dualism: questions have right answers, authorities know them, and learning means acquiring them. Later comes multiplicity: the realisation that experts disagree and many questions have no settled answer - often experienced as "so it's all just opinion." Then relativism: understanding that not all positions are equal, that some are better supported than others, and that judgement is possible. And finally commitment: the willingness to take a reasoned position and act on it while knowing it could be wrong.

Taught assessment can be passed almost entirely from within dualism and multiplicity. The dissertation is the first task that requires commitment: you must choose, justify, and own a direction with no authority to confirm you're right. "It's up to you" is your supervisor speaking from the stage of commitment to a student still hoping, quite reasonably, that somebody knows. The vertigo you feel is developmental, not deficient.

Threshold concepts and the liminal space

Jan Meyer and Ray Land's theory of threshold concepts describes certain ideas in every discipline that are transformative and irreversible - once you've grasped them, you see the field differently and can't unsee it - but which are also "troublesome" to acquire. Crucially, learners pass through a liminal space on the way: a period of being stuck between old and new understandings, oscillating, mimicking the language of mastery before genuinely possessing it.

Margaret Kiley and Gina Wisker extended this specifically to research students, identifying threshold concepts such as argument (understanding that a dissertation is a case, not a report), theory, framework, analysis and knowledge creation - the shift from consuming knowledge to producing it. Their work suggests that the miserable mid-dissertation months of feeling like a fraud are not a sign of failure but the liminal space itself: the precise period during which the transformation is happening. It doesn't make it pleasant. It does make it normal.

The impostor phenomenon meets its perfect trigger

Pauline Clance and Suzanne Imes first described the impostor phenomenon in 1978 among high-achieving people who attributed their success to luck or effort rather than ability and lived in fear of exposure. Reviews since have found it strikingly common among university students - with many studies reporting that a majority experience it at clinically meaningful levels at some point. The dissertation is almost engineered to trigger it: a high-stakes, individually owned, unfamiliar task, with ambiguous feedback and no external confirmation that you're doing it right. Students who have never previously felt like impostors frequently do here - again, not because they've become worse, but because the task has removed every mechanism that used to reassure them.

Self-authorship: the developmental task the dissertation demands

Marcia Baxter Magolda's decades-long longitudinal study of young adults traced the emergence of what she called self-authorship: the capacity to define one's own beliefs, identity and relationships internally rather than by following external formulas. Her participants mostly reached it in their late twenties or thirties - after university. The dissertation typically arrives at the "crossroads," her term for the uncomfortable stage of recognising that external formulas no longer suffice but not yet trusting one's own judgement. Read in this light, the whole dissertation is a self-authorship exercise slightly ahead of schedule, and "It's up to you" is an invitation to a stage of development you may not have reached yet. That's a lot to ask. It's also, arguably, the most valuable thing a degree can ask.

Decoding the supervisor: what the ambiguous phrases actually mean

Supervisors are not (usually) being evasive for sport. Most of their apparently non-committal phrases are doing real work, and they become far less maddening once translated.

• "It's up to you." Most often means: these options are all defensible, and the thing I am assessing is your ability to choose and justify - so a choice made by me would sabotage the assessment. Occasionally it means: I don't know your data or your interests well enough to decide, and it's your project. It almost never means I don't care.
• "Have you thought about…?" Frequently means: you have a gap here that I'm too polite to name directly. Treat every "have you thought about" as a flag, not a suggestion.
• "That's interesting." Depends heavily on intonation. Often a genuine spark; sometimes a placeholder while they decide whether it's a problem. Follow up: "Interesting good, or interesting worrying?"
• "You could go either way." Means the decision is genuinely underdetermined by the evidence, and the mark will attach to how well you argue for whichever you choose - not to which you choose.
• "Be careful with that." A warning delivered at low volume. Ask what specifically they'd want to see to be reassured.
• "Let's see what the data say." Means: stop deciding things in the abstract and go and do the work; the decision will be clearer with evidence in hand.
• "Is that really your question?" The most important one. It means your research question has drifted, ballooned, or was never sharp enough - and fixing it is the highest-leverage thing you can do this week.

Underneath these lies a skill the feedback researchers David Carless and David Boud call feedback literacy: the capacity to appreciate feedback, make judgements from it, manage the emotions it stirs, and act on it. Feedback on a dissertation is structurally different from feedback on an essay. It arrives mid-process rather than after, about work still in motion; it is often oral and unrecorded; and much of it is Socratic - questions rather than corrections - because a question preserves your ownership while a correction removes it. Students who struggle most are often those still waiting for the essay model of feedback (here is what was wrong, here is the fix) and not recognising that a supervisor's pointed question is the feedback.

Supervisor boundaries: what they will and won't do

Much dissertation misery stems from mismatched expectations about the supervisor's role - and research consistently finds that supervisors and students often hold quite different pictures of it. The higher-education researcher Anne Lee's widely used 2008 framework identifies five approaches supervisors take, usually in combination: functional (project management, deadlines, structure), enculturation (inducting you into the discipline's community and norms), critical thinking (questioning and challenging your reasoning), emancipation (fostering your independent development), and relationship development (support, encouragement, mentorship). A supervisor whose natural mode is critical thinking will feel brutal to a student wanting relationship development; one who leans emancipatory will say "It's up to you" constantly, on principle. Neither is wrong. But knowing which you have, and asking for what you need, transforms the dynamic.

Some boundaries are near-universal, and it helps to know them in advance rather than discover them by bumping into them:

• They will not choose your topic, question or method for you - because those choices are what's being assessed, and because it's your project, not theirs.
• They will not read unlimited drafts. Most departments specify a number of full-draft readings (often one or two) and a turnaround time (often two weeks or more). Plan around this rather than resenting it.
• They will not proofread or copy-edit, and cannot fix your referencing or your prose. Their comments address argument, structure, evidence and method.
• They will not tell you your mark. They frequently don't mark it alone, and they genuinely cannot predict how a second marker or external examiner will read it.
• They will not chase you. The initiative - booking meetings, sending agendas, circulating work in advance - sits with you. Silence from a student is usually read as "fine," not "struggling."

Against those limits, notice what supervisors do reliably provide, and how highly students rate it when they do. The 2025 Postgraduate Research Experience Survey - over 35,000 research students across 93 institutions - found supervision the strongest-performing area of the entire research-degree experience: 89% satisfied overall, 93% agreeing their supervisors had the skills and knowledge to support their research, and 91% agreeing their supervisors' feedback helped direct their research. Tellingly, the weakest supervision item - at 80% - was "my supervisors help me identify my development needs as a researcher." Supervisors are excellent at reacting to what you bring them; they are less good at diagnosing what you don't yet know you lack. Which is a strong argument for bringing them more, and more specific, questions.

Getting more from meetings: practical moves

1. Never ask an open question when a closed one will do. "What do you think I should do?" invites "It's up to you." "I'm leaning towards interviews rather than a survey because X; what's the strongest objection to that?" invites real engagement - and still leaves the decision yours.
2. Send a one-page agenda 48 hours before each meeting: what you've done, what you've decided, what you're stuck on. Supervisors give sharper feedback to work they've had time to read.
3. Write minutes and email them afterwards - "my understanding of what we agreed is…" This catches misreadings of ambiguous feedback before they cost a month.
4. Ask directly about their style and limits early: how many drafts, what turnaround, how they prefer to be contacted, whether they'd rather see rough work often or polished work rarely. Several studies recommend a written student–supervisor agreement precisely because expectations diverge so often.
5. Ask "what would make you confident about this?" whenever you hear "be careful." It converts a vague warning into a specific standard.

The fear of the wrong choice - and why it's mostly misplaced

The paralysis of "It's up to you" is really the fear of making a consequential mistake that can't be undone. Two things help: understanding why choice feels so heavy, and understanding how dissertations are actually judged.

On the first: psychologists Sheena Iyengar and Mark Lepper's famous 2000 experiments showed that people offered 24 varieties of jam were far less likely to buy any (3%) than those offered six (30%), and reported less satisfaction with what they chose. Abundant options degrade decision-making and increase regret. The early dissertation is a 24-jam problem: infinite possible topics, methods and framings, with the additional burden that this jar matters. The practical response is deliberate constraint - impose your own limits (a time window, a population, a single dataset, one theoretical lens) and choose among a small set, quickly. A good research question is a small one you can actually answer.

On the second: students consistently imagine that dissertations are marked on whether the choice was right. They are overwhelmingly marked on whether it was reasoned. Examiners want to see that you understood the alternatives, chose with justification, executed with care, and reflected honestly on the limitations - including the ones that emerged because of the choice you made. A study with disappointing findings, honestly analysed and thoughtfully discussed, routinely outscores a "successful" study whose author cannot explain why they did what they did. Null results are results. The unexpected direction your data took is a discussion section, not a disaster.

Most decisions, moreover, are more reversible than they feel. A research question can be sharpened after a pilot. A method can be adjusted before data collection. A chapter structure can be reorganised in an afternoon. The genuinely irreversible decisions - ethics approval, data collected the wrong way - are exactly the ones where supervisors do step in with directive advice. If yours has said "It's up to you," that is itself information: the choice is one you're allowed to get slightly wrong.

Where legitimate help actually lives

"Independent" is the most misunderstood word in the dissertation. It does not mean unassisted. Professional researchers - the people your dissertation is teaching you to resemble - work inside dense networks of support: methods specialists, statisticians, librarians, peer reviewers, editors, collaborators, mentors. Independence means you own the decisions. It has never meant you must reach them alone. Students who understand this use their institution's resources heavily; those who don't often struggle in silence, which is the one strategy the research says never works. A map of the legitimate support around you:

• Your supervisor - for argument, direction, disciplinary judgement, and the sanity check on your reasoning. Use them for what they're for, and don't waste your limited meetings on things others can do better.
• Subject librarians - the most underused people in any university. They can turn a floundering literature search into a systematic one in a single appointment, teach you database strategies, and set up alerts so you don't miss a key paper published in month five.
• Methods advisers and statistics clinics - many universities run drop-in services staffed by methodologists. Notice how often "help with methodology" surfaces in PRES comments: even doctoral students with excellent supervisors want a second, specialist opinion on design and analysis. Go before you collect data, not after.
• Writing centres and academic skills services - for structure, argumentation, and the genre conventions of a dissertation, which differ from anything you've written before.
• Ethics committees - not merely a hurdle. A good ethics application forces you to specify your design precisely, and reviewers' questions frequently improve the project.
• Peer writing groups - the Todd, Bannister and Clegg study found peer support surprisingly important. Weekly accountability with others in the same liminal space normalises the wobble and keeps you moving.
• Specialist dissertation support - model chapters, methodology examples, and expert feedback on your drafts. For most students the dissertation is the first time they've had to write a literature review that argues rather than summarises, a methodology chapter that justifies rather than describes, or a discussion that genuinely integrates findings with theory. As we've argued throughout this series, the fastest way into an unfamiliar genre is to study a well-constructed example before writing your own - and a dissertation has several unfamiliar genres stacked on top of each other. Services such as UKEssays' dissertation writing services provide model chapters written by academics who understand what examiners look for, and feedback on your own work while there's still time to act on it. Used as intended - study the model to see how a methodology defends its choices or how a discussion handles limitations, then write your own chapters from your own data and reasoning - it's the closest thing available to the personal tutor your fees never quite bought. The rule is the same as for every other scaffold in this series: learn from it; never submit it. The independence being assessed is yours, and it's worth having.

The common thread: none of these people will make your decisions for you, and all of them will make your decisions better informed. That's the whole distinction between support and substitution - and it's the same distinction your supervisor is drawing when they hand a choice back to you.

Reframing the sentence

Here, finally, is what "It's up to you" really means, once the panic subsides. It means your supervisor believes the options in front of you are all defensible. It means the thing being assessed is your judgement, and they're refusing to rob you of the chance to demonstrate it. It means they are treating you - perhaps for the first time in your education - as a colleague-in-training rather than a student to be instructed. And it means the decision is one you're permitted to get imperfectly right, because a well-reasoned imperfect choice is precisely what research looks like.

Todd and colleagues noted that supervisors' involvement naturally recedes across the year, from directive to background, as students grow into the work. Later studies describe the same arc as a movement from tutor to mentor to peer. The uncomfortable truth is that the receding is the teaching. A supervisor who decided everything for you would produce a competent dissertation and an incompetent researcher. One who hands the decision back - while staying available for the questions, the warnings, and the occasional firm "no, not that" - is doing the harder and more generous thing.

The bottom line

The dissertation is unsettling because it is the first assessment that tests a capacity your previous education didn't require: the ability to make consequential decisions under uncertainty and own them. Perry called the destination commitment; Baxter Magolda called it self-authorship; Lovitts observed that being a good course-taker was never going to be enough for it. Feeling like a beginner is not evidence that you are one - it's evidence that you have reached the edge of a kind of learning you've mastered and stepped into one you haven't yet. Read your supervisor's ambiguity as a form of respect, ask closed questions that keep the decision yours, constrain your choices so they're makeable, and surround yourself with every legitimate form of help your institution and the wider world offer. Then decide. Justify it. Do the work. Reflect honestly on what happened. That sequence - not any particular choice within it - is what "It's up to you" was asking for all along.