I’d like to dispel a common misconception about (social science) PhD research questions: they are not unchangeable. The research question(s) that appear in your final manuscript may not be the one(s) you proposed to your supervisor way-back-when or that took pride of place in your confirmation/upgrade document.
Let me be perfectly clear: I am not suggesting that you change your research question completely two-and-a-half years in. Down that road lies madness. What I am saying is that your research question(s) will evolve and shift as you get a better handle on the literature, as you enter the field…in short, as you do your research.
With that in mind, I suspect that the lifecycle of a (social science) PhD research question generally looks something like this…
Most PhD research doesn’t start with a question at all, but rather a broad topic of interest (e.g. an identified gap in the existing literature, a particularly interesting case that warrants investigation, a theory or concept that caught your interest during previous coursework). Having settled on a topic, the first few (or several…) months of research have two principle objectives: (1) becoming well versed in the relevant literature; (2) whittling down said topic of interest into a handful of researchable questions.
The first metamorphosis of your research question(s) occurs when your attention shifts from thinking about what you would like to research to what you will actually be able to research. It is important to distinguish between possible research questions and viable research questions. Not every question you come up with will be doable; or, at least not doable with the limited time and resources available to most PhD students. As you think more carefully about what you can realistically achieve, the nuance of your question(s) will probably shift. The focus may narrow, for instance, as you reign in your ambitions.
A second transformation might occur as you settle on a specific case study, particularly if the case you choose wasn’t implicit in your broad topic of interest. You might need to re-word one or several of your research questions to reflect the context that accompanies your choice of case study. Indeed, it is possible that the case itself might raise new questions that you didn’t consider when you were thinking in broad strokes about your general topic.
Having settled on your topic, chosen a case, and articulated your questions, all that remains to be done is the fieldwork (or original research in whatever form). Maybe. If you’re lucky, your fieldwork will confirm (or conform to) your expectations and it will be smooth sailing from here on in. Some of us (myself included) aren’t so lucky. The circumstances you encounter ‘on the ground’ can precipitate a dramatic re-think of your research questions. The worst-case scenario is that some questions may no longer be viable and you have to throw them out. Odds are, however, that most questions can survive the cull with some minor tweaking.
Data collection accomplished, you set in on your analysis. If you’re taking a deductive approach, your questions are probably safe at this point. If you’re adopting a more inductive analytic approach, however, you might decide that the data answers slightly different questions to the ones you had in mind. Or, for that matter, the data might raise new questions you hadn’t previously considered. At this point you have to decide whether you have the time to follow where the data leads. It might be more pragmatic to save these emergent questions for a ‘suggested-avenues-for-additional-research’ section in your conclusion chapter.
In short, there are no fewer (and quite possibly more) than six possible junctures where you might opt to re-jig your research questions. As you progress through your research journey, your questions serve as guideposts keeping you on the right track – or at least on track to completion of a cohesive thesis. At the end of the day, your final research question should be the question that your research actually answers. The stage of the research process at which you come up with the precise wording of that question is indeterminate.