A student sits down to write a dissertation proposal and types “social media and mental health” into a blank document, then stares at it for twenty minutes. The subject is real, the interest is genuine, and yet nothing happens. That stuck feeling is not a lack of ideas. It is the gap between having an interest and having a question, and almost every struggling research proposal lives in that gap.
Communication as a field makes this gap wider than most. It touches psychology, technology, politics, culture and organisational behaviour, so almost anything can be framed as a communication issue. That breadth feels like an advantage until you try to write a research question from it. A subject that can mean anything ends up meaning nothing specific enough to investigate.
Interest Is Not the Same as a Question
Most students confuse a topic area with a research problem. “Influencer marketing” is an area you could spend a career inside it and never run out of things to say. A problem looks different: does an audience actually notice sponsored content when the only disclosure is a small tag buried in the caption, or do they read it as a genuine recommendation regardless? The area gives you a subject to be interested in. The problem gives you something you could actually go and check.
This matters because a broad area gives you nowhere to stand. You cannot design a survey, a discourse analysis or an interview schedule around “influencer marketing” without first deciding what specifically you are trying to find out about it. Students often skip that decision and go straight to method, which is why so many projects stall at the literature review stage they are reading everything because they have not yet decided what they need to know.
Why the Obvious Angle Usually Fails
The first angle that occurs to you is usually the one that occurred to thousands of other students too, which means the existing literature is already dense and the obvious contribution has already been made. This is not a reason to avoid the subject altogether. It is a reason to look for the specific corner of it that has been assumed rather than tested.
A good habit here is to read not for information but for gaps the authors admit to. Papers often end with a line such as “further research is needed to examine…” or quietly limit their claims to one platform, one demographic or one time period. Those limitations are not throwaway sentences. They are the field telling you, in its own cautious language, where the unfinished work sits.
Narrowing Without Losing the Point
Narrowing a subject is often taught as a mechanical exercise pick a population, pick a context, pick a variable and that advice is not wrong, but it misses why narrowing works. A narrow question is not smaller for the sake of being manageable; it is narrow because specificity is what makes a claim testable. “How does misinformation spread online” cannot be answered by any single study. “How do correction messages from health authorities affect the sharing behaviour of users who have already engaged with a false claim” can be, because it names an actor, a behaviour and a condition.
When you are close to a workable version of a question, you can usually restate it as a sentence with a clear subject and a clear outcome, without the word “and” doing all the heavy lifting. If your working title still needs three clauses joined by “and” to make sense, it is probably three studies pretending to be one.
It is around this point of confusion armed with a subject but not yet a question that most people quietly open a new tab. There’s nothing wrong with that instinct; wanting to see what other people are circling around is a perfectly sensible way to test whether your own half-formed idea is even in the right neighbourhood. If you’ve ever done this yourself, browsing pages built around communication research topics 2026 and similar searches, you’ll recognise the feeling that follows: a long scroll of subject headings, some of them genuinely interesting, none of them telling you whether you could actually pull the research off given your access, your timeline or your ethics board’s patience. That’s not a wasted step. It just means the list was only ever meant to start the thinking, not finish it.
Testing Whether an Idea Can Actually Be Researched
Before committing to a direction, it helps to ask a blunt question: what would count as evidence here, and can you actually get it? An idea about how corporate leaders should communicate during a crisis is interesting, but if you have no access to real organisational communications or leaders willing to be interviewed, you are left analysing press releases and hoping they reveal something they were written specifically not to reveal.
Feasibility is not a boring afterthought bolted onto an exciting idea. It is part of what makes an idea good in the first place. A brilliant question with no realistic way to answer it is not a research topic yet; it is a wish.
A second, quieter test is whether you can explain why the answer matters to someone outside your seminar room. If the honest answer is “because it’s interesting to me,” that is a fine starting point but not a finished justification. The stronger version connects the question to a decision someone makes, a policy someone writes, or a practice someone could change once they know the answer.
Letting the Topic Change Shape
The direction you start with rarely survives first contact with the literature, and that is not a failure. Reading twenty relevant sources should shift your question, sometimes subtly and sometimes completely, because you are learning what has already been settled and what genuinely has not. Treating your first draft of a topic as fixed is one of the more common ways students waste months defending an angle that the evidence quietly stopped supporting weeks earlier.
The students who end up with strong, defensible projects are rarely the ones who found the cleverest subject. They are the ones who kept asking what exactly they wanted to know, kept checking whether they could realistically find it out, and were willing to let the question sharpen as they learned more. That process is slower than picking something off a list, but it is the only version that produces a question you can actually stand behind when someone asks you why it matters.