What differentiates a passing PhD dissertation from good research below PhD standard?

Ross Woods, 2024

This question came up because so many dissertations are good; there's nothing actually wrong with them and their writers made no mistakes. The topic is original in that it reflects a gap in the literature. However, even the method for finding that gap is proceduralized. Then everything else in the dissertation is little more than following a series of well-established procedures.

A range of answers

Most of the answers I recieved mentioned significance in some way:

Lacking originality and significance.
A gap in the literature is not adequately compelling to justify the need for study. The student needs to demonstrate its significance.
It might be possible to replicate an existing study, but students need good reasons for doing so, for example, better methodology and compelling possibility of different results. Otherwise the dissertation risks a lack of significance. Similarly, a replication with only a minor adaptation might not be significant enough.

Some answers mentioned weaknesses in other parts of the dissertation:

Used and reported an effective methodology but did not give the reasons for methodological decisions.
Imported tools rather than creating one’s own.
Overlooked gaps in the literature when writing the literature review.
Weak analysis, indicating lack of understanding.
Failure to identify implications.

Several people commented that brevity is a problem in some fields, regardless of the quality of the research. They expect 100,000 words.

Prof. Βrian Hill said the standard varies. In some institutions, it is enough to “have a really good look.” In others, “you actually have to find something.”

Philips and Pugh* describe a PhD graduate as someone who has acquired the skills to be a “full professional” and a peer in the research community. A student who overly depends on a supervisor cannot “fly solo” as a researcher (Philips & Pugh, pp. 18, 20).

Drew Rae

This article was originally posted on Quora.com and reposted without change and with permission granted on 30 January 2024.

A PhD dissertation is required to provide a “a significant and original contribution to knowledge.” That may seem like a fairly vague standard, but it gives clues to how a dissertation might fall short. Either the contribution is not significant enough (i.e. there’s just not enough volume of work) or the contribution is not original enough (i.e. the work is just minimal increments over things that have already been done).

There’s a third criterion which is not always explicitly stated, which is that a dissertation is supposed to show that the candidate has a strong grasp of the literature, theory and methods within their field.

From my own examinations, as well as reasons my supervisees have been asked for revisions, here are some of the common differences between a pass standard and a request for revisions, ranked roughly from most common to least common.

The length, depth and quality of discussion around methodology is insufficient. Typically, the candidate has applied the methods competently, but they haven’t shown a good understanding of why they are doing what they are doing. Maybe the thesis is a collection of several papers, each with their own short method section, but there is no overarching methodological approach to the work. Maybe the methods chapter is just like a recipe, saying what was done but not why.
The engagement with the literature is insufficient. Depending on the format of the thesis, this may be specifically in a literature review chapter, in a discussion chapter, or throughout the thesis. Maybe there are big gaps where there is relevant work that the examiner expected the candidate to discuss. Maybe there are contested concepts where the candidate has just picked one side of the debate, without acknowledging that there is a strong existing critique. Maybe there are recent developments in the field that the candidate has ignored.
Too much of the work is below standard. A PhD isn’t meant to be perfect, so long as the candidate recognises and acknowledges the weaknesses in their own work. But there has to be enough good work to make a significant contribution. My personal rule of thumb is that the minimum contribution is equivalent to 3 good published papers. So I ask my own candidates to aim for 4 papers, just to give some room for disagreement. In other fields this benchmark might be a little higher or lower, depending on the typical size of published papers.
Imagine the examiner, after doing their detailed review, flicking through the overall thesis. “Literature review, ok, they’ve done the bare minimum, but they couldn’t publish this as a separate paper. Methods, ok, but there’s nothing original here. Study 1, that’s already been published in a good journal, that counts as one. Study 2, that’s good enough to publish, that counts as paper two. Study 3, no, if I was a peer reviewer, I’d be recommending rejection. So overall, that’s only two papers’ worth. That’s not enough yet.”
It’s just not original or important. This is the one that almost every PhD candidate worries about, but is fairly rare in practice. It’s more of a supervisor failure than a problem with the candidate. Getting to this point really requires not understanding your own field, and basing the whole project on outdated concepts or methods. This can happen, and is terrible for everyone when it does. But even then the examiners will usually try to offer a path forward by improving the literature review, methodology and discussion to elevate the overall contribution of the work.

... I think the advice of “finding a gap” is part of the problem. Not all gaps are meaningful. “This has been examined in the US, but not in Australia”. That doesn’t mean that there is a “gap”, unless there is some good reason to believe that this sort of things works differently in Australia. “People have studied ‘bananas’ and ‘oranges’, but no one has studied bananas and oranges at the same time”. That doesn’t mean there is a “gap” combining the two things. What’s the interesting question at the intersection of bananas and oranges?

This is the sort of thing that I’d consider to be as much the fault of the supervisor as the candidate. The absolute minimum a candidate should expect out of a supervisor is reliable feedback on whether the project, if completed appropriately, is capable of making a PhD-worthy contribution.

With thanks to Drew Rae, Griffith University

Philips and Pugh

Philips (Univ. London) and Pugh (Open University) also give several faults in research strategy that cause students to drop out of the PhD. These are better classified as deficiencies and not directly relevant to the original question, but are neverthless very informative:

Unable to frame a research question (p. 35)
Thinking of practical issues only and not theory (pp. 30f, 34)
Writes for the general reader, not researchers (pp. 31, 33f)
The conclusion does not add to the theory of the topic (pp. 32, 38f)
Not arguing for a particular conclusion (pp. 18, 38)
Unable to evaluate research (p. 20)

To their list, several more are relevant:

Avoids difficult theoretical issues that arise, or does not seem to even notice them.
Unable to challenge old explanations.
Gives opinions rather than supportable facts.

__________
*Estelle M. Philips and Derek S. Pugh. 1987. How to get a PhD: A Handbook for Students and their Supervisors (Milton Keynes & Philadelphia: Open University Press).