Topic

By checking whether it can actually be researched before committing to it — which is the step many candidates skip.

Bear in mind that unlike the most senior bricks and mortar universities, most university doctoral programs …and nearly all online programs … have no provision for major research efforts. Indeed, most online universities actively discourage such research – and the associated need to get research grants and so forth. Generally, online universities will expect to see either quantitative research (that is research that results in hard numbers susceptible of statistical analysis) or qualitative research where a small number of subjects (located by the doctoral candidate) undergo interviews, followed by qualitative analysis. In practice, for many doctoral candidates this means research that is online survey-based if quantitative. If qualitative, generally a very small group of individuals is interviewed and those results analyzed by one of various qualitative approaches (phenomenological, etc.). Mixed methods partake of both qualitative and quantitative approaches and are not often seen in the online academic space and are best avoided even at most bricks and mortar universities unless your Chair is supportive.

Quantitative Research is concrete and, if the survey questions are in alignment with the research questions subject to standard statistical analysis (whether done by the candidate or with help from a statistician). Quantitative research provides clear results from statistical analysis that can be performed in days and allows you to get on with reporting your results and writing your conclusions. Thus, Quant based studies are often easier to bring to fruition than qualitative studies as it is easy for doctoral candidates to become confused and lost in doing qualitative analysis if they do not have solid guidance.

Qualitative Research can be less costly (not necessarily but can be) if the Candidate is confident in their ability to do their own analysis. The very small group of subjects to be interviewed, in interviews conducted by the doctoral candidate helps with this. However, developing good questions and doing good interviews is harder than it might appear.

Note,for a qual topic that can be effectively researched and analyzed you need to ensure, 1. That you can get interviewees with the right knowledge. 2. That you thoroughly understand your qualitative analytical framework.

Finally, qualitative analysis, especially phenomenological qualitative analysis requires not only a great deal of work but can be a trap where candidates get lost in the coding and theme-generation processes.

Qualitative studies should not be chosen by candidates because they feel intimidated by statistics (lots of help is available for that) or because they appear 'easier' somehow. They are not.

If you have a topic, or topic area, in mind and your Chair agrees it is a good idea jump ahead to the Scope section. If you do not – or even if you do – you may find your Chair pushing a specific topic or range of topics. The reasons for this vary and can be anything from the Chair being familiar with that area of study, personal interest, or just following academic trends. Suggestions by Chairs can be extremely helpful or irksome if they are pushing a hobbyhorse a candidate is not interested in. Candidates doing topics that only their Chair wants do lead to less Candidate satisfaction with their dissertation and sometimes, theses not suited to a candidate's career or life goals. If a topic is totally uninteresting to you and you are being pushed? Don't do it. You are paying a very great deal of money for your doctorate, and it is easier to change Chair's than get a second doctorate…or to drop out of the program with a lot of student loan debt.

No idea? Chair has no viable suggestions? Time to look at academic journals in fields you are interested in and see what scholarship is taking place. If you find areas you are interested in, the research papers you read will already highlight areas fruitful of future research. Now is the time to ask what you are really interested in and where you think your life is headed in future. Try to find topics related to your interests. Avoid things that are so cutting-edge that there is little scholarship.

The literature review is where a thin topic becomes obvious.

In our experience — and it matches Capella's chapter guides, which set a twenty-five-page minimum — the review section needs at least twenty-five pages reporting on previous scholarship. And relevant is the load-bearing word: relevant to your specific research question, not to your field in general.

So a topic in an area with little recent work is a trap. Recent usually means the last five years, sometimes the last seven, seminal work aside — check what your own program expects. Twenty-five pages of reporting on research that does not exist is not a writing problem you can work harder at.

And it comes back in chapter 5. Your concluding chapter has to show how your work builds on what came before.

This is not an institutional requirement you might get around. It is what doctoral scholarship is. A contribution is a contribution to something — and if the literature directly supporting your question is thin, there is nothing for your findings to be a contribution to. No program anywhere can accept that, and the ones that seem to are the ones you should worry about.

You find out at chapter 5, which is the worst possible moment.

If you are facing some of these problems and not finding the right topic, give us a call, from ideation to scope, we can help.

Summary of a good topic

• A topic you are interested in.
• Lots of recent research, generally less than five years old.
• A topic that advances the art — new research, addressing a question no previous scholar has answered.
• Survey instruments easily available.
• Survey instruments that fit your chosen theoretical framework and analytical approach.
• For a qualitative topic, interviewees you can actually reach who have the right knowledge.
• For a qualitative topic, an analytical framework you thoroughly understand.

A model will give you twenty dissertation topics in thirty seconds. Every one will sound plausible. That is the problem, not the benefit — alas.

Two things it cannot do, and both are the whole job at this stage.

It cannot tell you the gap is real. A topic is viable because nobody has answered that question yet and enough directly supporting research exists to build the twenty-five pages of literature review that we, and Capella's own chapter guides, treat as a floor. Both halves have to be true, and only reading the literature establishes either.

And if you find yourself leaning on one source for everything that matters, that is the same signal arriving late. A review that rests on one body of work usually means the adjacent literature is thinner than it looked — which is a finding about the topic, not a fault in your reading.

"Directly supporting" is where candidates most often go wrong, and it is worth being precise about. It does not mean research in your field. It means research your study builds on — comparable constructs, comparable populations, comparable designs — the work you will cite when you explain what your findings mean. Research that is merely about the same subject is adjacent, and adjacent research cannot carry a literature review.

A generated topic sounds like a gap. Whether it is one is a question about what has actually been published, and the model is guessing.

It cannot tell you an instrument exists. As this page explains, many programs — and most online programs — expect a validated survey instrument you did not build. A model does not know which instruments exist, which are validated, or which your institution will accept — so a generated topic frequently cannot be researched at all under the rules you are working to.

Where it genuinely helps

• Narrowing a topic you already have. Give it your broad area and ask for progressively narrower versions. You are asking it to rephrase, not to know things.
• Generating search terms for the databases where the real work happens.
• Sorting scholarship you have already found and read into themes.

One tool class is worth knowing about specifically.

Tools such as NotebookLM let you load your own PDFs — papers you have found, downloaded, and read — and then ask questions across only those documents. Nothing else. No open internet, no general model knowledge, no invented sources.

That closes the single largest risk at this stage. A model asked about the literature in general may produce papers that do not exist. A tool restricted to twenty PDFs sitting on your own disk cannot, because it has nothing else to draw on.

Use it to interrogate a corpus you have already vetted: what do these twenty papers say about my population? Which of them used the instrument I am considering? Where do they disagree? Those are questions with answers inside the documents, and the answers can be checked in seconds because you have the documents.

It still cannot tell you a gap is real, because a gap is a claim about everything that has not been published, and your twenty — or two hundred — PDFs are not everything. But for working with what you have found, it is the safest way to use these tools at this stage.

The rule is the same one that runs through this whole site: use it on your text, not on your research. A topic is a claim about what the literature does and does not contain, and that claim has to be checked in the literature.