12

With regards to PhD thesis that spans several disciplines, at what point does one assume that an article doesn't require explanation and it is standard knowledge?

I have had this discussion with several academics and there doesn't appear to be consensus. One suggested a middle ground that I quite liked: Anything considered standard knowledge, should be placed into an appendix so as to not disturb the flow of the thesis but still somewhere handy enough for the reader to bush up on or explore.

Is this a common technique? I've read so many theses that contain waffle and filler explaining (what I consider to be) the basic requirements of the field, however my option of "basic" is just that; An option. People have different degrees, life experiences, interests and so their options will no doubt differ regarding what is considered standard knowledge.

So at what point is something consider "standard knowledge"? If a technique/paper/method is widely written about in master/Bachelor's course texts, journal papers and has a wiki page - does that qualify? For example; consider the most basis statistical techniques, such as chi squared test etc - does one really need to explain, in detail, what it is? Would a reference suffice or would a thesis begin to look ridiculously short? Perhaps placing it into an appendix?

Is there a general accepted rule of thumb that I am unaware of? Or is it truly a black art? I don't want to look lazy by not including material, but at the same time I don't want to waffle on.

  • Don't the theses you've already read, and your discussions with academics already answer your question? There's no consensus and you should include as much information as you think should be there. – David Richerby Sep 5 '15 at 14:54
10

I'm struggling with this problem myself at the moment. I think there's no one-size-fits all situation. It's your book, you have to consider who your audience is, and how you're going to tailor the thing to them. Start with your committee, make sure that they have the tools to understand what you've written. After them, the most important people to think of are probably the other PhD students entering your field, possibly building on your work. Think back to what you struggled with when you first started and what kind you would have liked to have.

Here's a few tips and tools I plan to make use of:

  • A tutorial appendix: This is a great way to make sure that you satisfy both the expert reader, and the novice who's willing to work. Make it short, ideally five to ten pages that will require some dedicated work but just explain the required concepts for your thesis. End with some pointers to longer tutorials. This also gives you the opportunity to write the tutorial you wish you'd had when you started out.
  • An informal introduction: I like the idea of trying to write an introduction that my mother could read. Something that describes the ideas in such informal language that everybody who's willing to put the time in can get through it. This gives you one part of the book that you know absolutely everybody can read, and for the more expert readers, it gives them a motivation to guide them through the technical matter. Perhaps the same could go for the conclusion, to create a bookending effect.
  • Err on the side of caution: I think it's better to insult people by explain too much than to lose them. So when in doubt, make sure things are properly explained.
  • Pretend you have a page limit: The danger of erring on the side of caution is that you'll just ramble on and on. Nobody will read your work if you don't get to the point as quickly as possible. This is the basic contradiction you have to deal with when writing. Explain as much as possible, and get to the point as fast as possible. See if you can keep it under a hundred pages.
  • 2
    I'd say speak to your advisor before erring on the side of caution re: explaining too much, some take the view that it's a waste of you and your examiner's time. – blmoore Sep 4 '15 at 16:22
  • 1
    No. "It's your book" -- a dissertation is not a book, it's a dissertation. You might turn your dissertation into a book but it's not a book. It's not intended to be read by novices or by parents, or even be sold. It's proof that you can conduct independent research at the doctoral level, that's all. – RoboKaren Sep 5 '15 at 14:11
  • 2
    p.s. p.s. As a woman and a scholar, I find the phrase "could be read by my mother" to be insulting in its implication. You'll most probably argue that you're talking your specific mother, but it has become a turn of phrase that is deeply sexist. Ask yourself, why didn't you say, "couldn't be read by my father" -- or how many times you've heard that latter version. – RoboKaren Sep 5 '15 at 14:12
  • 1
    @RoboKaren, you're right about that phrase. In my case, my mother serves as better benchmark, because my father has some mathematical background. "Could be read by my father" also has equally sexist implications, because it paints a picture of the men talking about science amongst themselves. I suppose "could be read by an interested, unschooled family member" is more neutral, but it's also a little sterile. Still, you're right that I should retire the phrase; it does more harm than good. – Peter Sep 5 '15 at 14:58
  • I prefer "could be read by anyone" or "accessible to anyone" or something like that. No need to use family. – RoboKaren Sep 5 '15 at 14:59
18

Generally, the target audience of your doctoral dissertation is your doctoral dissertation committee. Write to their level.

In disciplines where your dissertation gets turned into a book (or series of articles), then you may consider writing at a lower level or with explanation for people outside of your specific subfield. You can certainly do this if you have the time or luxury to do so, but if you are short on time or funds, then it is perfectly acceptable to do what is noted in paragraph #1.

  • 4
    I think this is the best answer, in our field a dissertation is mostly a waste of time, nobody reads them, the only people who care about them are the committees. Advisors mostly care about the papers which students publish. I honestly don't even understand why we do the book form dissertations anymore, since they don't matter for an academic career in our field. It does have one benefit, increasing the understanding of the person writing it, but that time could be better spent writing another publishable paper, especially considering how competitive the academic market is today. – daaxix Sep 4 '15 at 15:13
  • 1
    @daaxix Some places allow for a thesis to be a collection of peer-reviewed published articles. – user8001 Sep 4 '15 at 17:21
  • 1
    I like this two-step approach. If you have the time and inclination later, you may start adding appendices and a lay introduction. I would just add: have a chat with some good listener in your general field, and give him/her an overview of what you're going to cover in your write-up; as you are writing, imagine someone like this good listener reading your text. This will give you a focal point for your imaginary readership. – aparente001 Sep 5 '15 at 15:15
3

This is one place where I think even with a single field, explanation can be key (remember, not every math person is up-to-date on every field of math, etc).

For your example with the chi-squared test, you could just say "To reach this conclusion, we used a chi-squared test". But it is beneficial for everyone (statistician or not) to understand why you did it. So you could rewrite it as "To reach this conclusion, we used a chi-squared test which is the normal method used for calculating X given conditions Y with the desire to interpret Z". People unfamiliar with statistics need not know the underlying methods, but they will fully understand what you're aiming to do and why and the statistician can then further verify correct usage of the method.

For terminology issues, this can often happen within the same field (at least, in my field of literature). Different authors or schools of thought may use two terms that have effectively the same meaning. It's common to see something like "...what is effectively an X, or what so-and-so termed Y, due to Z". Notice that because we indicate Z, we establish to the people who do know X/Y (people in-field) can verify we know what we're talking about, and people who don't know X/Y (out-of-field) can still follow along because Z gives them the gist of it. In fact, if your paper will regularly be cycling through fields or integrating them, it's highly advisable to establish a standard set of terms in early on (indicating their amongst fields then only) to aid the flow of the paper later on.

Also, don't hesitate to use footnotes quite liberally initially. As you go revising your dissertation in conjunction with your committee —or simply having people from various fields reading it— you'll get feedback letting you know which footnotes would be best integrated into the main paper (very important information for the out-of-field people), which ought to remain as footnotes ( pertinent, but not super necessary for out-of-fielders to know), and which can be removed entirely (way too basic, or not really important at all).

3

The answers given by @Peter and @RoboKaren are good, and therefore my answer is an addition to theirs.

My Experience

My dissertation is very interdisciplinary and so is my committee. The core problem I needed to solve was how to give clear, explicit definitions to many key terms (incl. references) without cluttering the main body of the dissertation (which is long already). It would not be sufficient for me to simply give a reference next to each key term.

My solution was to have a Glossary at the end, with hyperlinks in the main text pointing to glossary entries. This means I can nearly always put all of my definitional text in the glossary and focus the main text on exposition. This can help my readers (the committee) because they will know to look to the Glossary for all key definitions, and not the main text.

General Advice:

Figure out what needs to be accomplished relative to the 'basic' or 'common' knowledge, and then choose a design and writing strategy to meet this goal.

"What needs to be accomplished" is specific to your dissertation and to your committee. For example, do you need to simplify the reading process? Minimize confusion? Justify your methods? Provide tutorial assistance to a reader outside of a given field or specialty? Each of these goals might call for a different design and writing strategy such as those listed in the answer by @Peter.

  • +1, I forgot about the glossary, that's good advice. Also, if applicable, a list of common mathematical notation with explanations/definitions can really help. – Peter Sep 5 '15 at 15:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.