Choosing a dissertation topic ....

Discussion in 'General Distance Learning Discussions' started by EllisZ, Jun 24, 2002.

Loading...
  1. EllisZ

    EllisZ Member

    Question: (And I feel kind of ignorant in asking this ... so don't beat up on me too badly.)

    As I review some schools that I would consider enrolling in for doctoral level work I find that many want to know my dissertation topic right on the application.

    Believe it or not I had not thought about it in any granular level as of yet.

    Ideally I'd like to revisit some research that I performed and published during my undergrad years dealing with the use of multi-media courseware in education. (Which would combine my IT background along with my aspirations for an educational doctorate.)

    Now for the questions:
    HOW do you choose a topic? How do you articulate this? Where can one find published dissertations for review to get some ideas?

    Also: My ability to perform research on live-subjects is limited at the moment. I assume there are creative ways to pursue this?

    Thanks for the feedback. Please be gentle. :)
     
  2. Peter French

    Peter French member

    The process is different out here to oer there - have you looked up UMI to get some 'ideas'?

    We are requried to do a 2-3 paragraph outline which gets us in and then it is developed into a proposal that cna be as long as 10,000 words in some cases.

    One of the popular supervisors I worked with in the EdD program at Latrobe used to hand out small seed enveloped at our Thursday lunch meeting where everyone qas required to do a 30 minute presentation on their topic and later their research mthodology, and findings etc as time went on. These envelopes were about 2 inches by 1 inch and the mystification disappeared when he said - '...no one is ready to proceed until they can get the question they are seeking the answer to, written on one side of this envelope.

    Probably doesn't help much - but you are not the first person with the same problem - woudl you like me to send you a seed envelope?

    Peter French
     
  3. Rich Douglas

    Rich Douglas Well-Known Member

    This can be tough, considering you haven't done a thorough research of the literature at this point. Still, you should have some idea on what you'd like to research. It sounds like a few words would do it.
     
  4. telfax

    telfax New Member

    This is how it will work....!

    Firstly, you have to be PASSIONATE about the topic you wish to research. If you are not then forget it! You have to be passionate because there will be many troughs as you develop your thesis/dissertation and it is only the passion and interest that keeps you going.

    Secondly, you will need to identify your topic and then do an aweful lot of reading around it and around themes that may be related to it. Although most topics/subjects are capable of further development you may find that your slected topic has been 'done to death' and there is actually little else to be researched about the topic! This backround reading will help you idetify a theme that is worth looking at.

    Thirdly, get a copy of the Modern languages Association handbook re doing research and look at Yin's (and many others) work if you are considering a single (or more) case study approach. Not until you've looked at the various research methodolgies available will you really be able to determine what you will be doing.

    Fourthly, remember the word KISS - keep it simple, stupid! There is a great deal of truth in this! Ask yourself these simple questions at this stage:

    What is it I want to look at?

    Why?
    When? (period of time)
    Where?
    How?

    If you can asnwer these simply questions you'll find a format starting to develop.

    My own supervisor used to ask me as the thesis and research developed:

    WHAT IS THE LAST SENTENCE OF THE LAST PARAGRAPH OF THE LAST CHAPTER? IN OTHER WORDS, THAT FINAL SENTENCE OR TWO AT THE END AND THE ONE PAGE ABSTRACT AT THE BEGINNING OF THE THESIS HAS TO SAY IT ALL.

    One final suggestion. Begin EVERY chapter with a short introduction to tell the reader where s/he is going. At the end of the chapoter provide a summary of the chapter and a link to the next chapter. These summaries and links will change as each chapter develops. You must also be prepared to go back and revise chapters, especially in the light of your data analysis.

    Best wishes,

    telfax
     
  5. tcnixon

    tcnixon Active Member

    One thing to bear in mind is that you can fail and still pass. One of my graduate school professors ultimately proved his hypothesis was false. He still got the doctorate. Why? Because proving that something does not happen can be just as important as proving something is (even if you do it by accident).

    It's all about the process and not whether you're right or wrong.




    Tom Nixon
     
  6. telfax

    telfax New Member

    Agreed!

    Agreed! You can prove something you set out to do is not as you thought it would be! Providing you can support your asseryions with sound research evidence, that's fine!

    Set up a 'straw man' and knock it down - that's the essence of a doctorate!

    telfax
     
  7. Andy Borchers

    Andy Borchers New Member

    I heartily agree that not supporting your hypothesis isn't the kiss of death on a dissertation - although it is easier to publish results where you support the hypothesis. On my dissertation some hypotheses were supported and others were not. Indeed, if dissertations had to include support for a hypothesis, the researcher's objectivity would be biased. You are supposed to let the data talk - whatever it may say.

    As a minor technical point, however, no one "proves" a hypothesis. Statisticians are a timid lot. They "support" or "fail to support" their hypotheses. Given the nature of statistics we never "prove" anything.

    Regards - Andy, a late night statistics teacher...

     
  8. Gary Rients

    Gary Rients New Member

    If the hypothesis were mathematical or logical in nature then I suppose that a proof may be just what the Dr. ordered. ;) Also, some hypotheses can certainly be physically proven. For example, if your hypothesis were to state that it is possible to create a stable cold fusion reactor, then building and demonstrating such a device would seem to proove it. Research that relies upon statistical probability to support a hypothesis will only produce just that though: a probability.
     
  9. Charley Tichenor

    Charley Tichenor New Member

    I think it is usually a very good idea to have a dissertation topic identified and at least informally approved by the university before you start a Ph.D. program. I always recommend this approach to others. I did, and having the topic informally approved from the start allowed me to think about it and develop it as I was conducting my regular course studies. When it came time for the formal dissertation Proposal, I was able to complete it relatively quickly and spent a much shorter time completing my dissertation than most other colleagues did. Also, this approach signaled to me that the university I was considering had the faculty with the appropriate expertise to understand the dissertation and work with me through its development.

    You may be an expert within your profession or at least on the development curve towards that expertise level. I would identify an interesting problem that has not yet been solved within your profession and focus your dissertation topic on that. This will kill two birds with one stone. You will have the chance to become the industry expert on that particular topic when you complete your Ph.D, and you will not have to start developing your topic "from scratch." You will already have much of the background technical knowledge from which to conduct your research. Your dissertation will go much more quickly (and when you are finished you have the chance to be recognized in your industry as an expert in the field).
     
  10. EllisZ

    EllisZ Member

    Thanks for the advice all. It is most appreciated.

    It would seem that I am about to take several weeks and research what has been done in this topic since I looked at it back in '97 ... I'll see if I can get a general direction from there.

    BTW: Someone mentioned UMI. *HOW* do you find dissertations on that thing? I keep getting sales pitches ...
    (I did find quite a few good resources using a broad google search though including some extremely relavent book/publication lists ... so I've got something to start with.)
     
  11. Rich Douglas

    Rich Douglas Well-Known Member

    As a doctoral candidate, I want to support Andy's contention that the direction of your data--and, thus, the support or non-support of your alternate hypothesis(es)--is not the issue. Rejecting the alternate and accepting the null is still a contribution to the field. However, Andy is also quite right in that it is when the alternate hypothesis is supported that we have new and interesting--and more publishable--outcomes. But the dissertation isn't sexy; it's a demonstration of your ability to conduct original and significant research.

    As someone who's taught inferential statistics for two universities, I also agree with the distinction between proof and support. We accept the null hypothesis not because we've proved it (or disproved the alternate). Rather, we simply don't have enough data to support the alternate. When we do, we say we accept the alternate and reject the null. Theoretically, nothing is ever "proven." But in real-life application, some theories are so supported by the evidence that we accept them as "fact."

    Inferential statistics--decisions inferred about a population based upon a sample of it--is subject to sampling error and the probability of that occuring. We can control the likelihood of sampling error ("confidence level"), but we can't eliminate it without abandoning sampling and conducting a full census in its place. The possibility exists that we will reject the null hypothesis--based on our observations--when there is no difference in the population itself. And the possibility exists that we will accept the null because we do not have enough evidence to reject it, despite the fact that the population HAS that difference. We don't know when we make this mistake, but we can control for it, and even lessen its likelihood to a tiny proportion. No, you don't "prove" things with statistics, as Andy clearly notes. But you can have such strong evidence that to reject it is unreasonable.

    OT point: Evolution is often called a "theory," as if this distinction somehow makes it unsure whether or not evolution occurs. It does, and it is observable. Beings evolve. The real theory behind evolution is natural selection, the explanation of those observations. It is well supported by observations, but like any theory, it is subject to re-examination upon further evidence. And whether or not natural selection explains the situation well enough for your blood is a personal decision, just like whether or not gravity explains other observations well enough for you.
     
  12. Mike Albrecht

    Mike Albrecht New Member

    Go to http://wwwlib.umi.com/dissertations/ click on continue and then do your search. Unfortunately this "seems" to only cover the last two years.
     
  13. EllisZ

    EllisZ Member

    Mike,
    Thanks. I'll give this a try this evening.

    - Ellis
     
  14. Ike

    Ike New Member


    Bear in mind that it is easier to build on top of an already established and accepted concept than to initiate a new concept. Remember the famous statement of Isaac Newton. Newton said that he accomplished his great feats by standing on the shoulders of giants. I believe that it will be easier to start your dissertation by standing on the shoulders of giants. Below are steps you could follow when choosing a dissertation topic.

    1. Think of a topic you will like to work on (such as information visualization, Online Instruction Delivery Systems, Object-Oriented applications, etc. You can start with your current topic (Multimedia Courseware in Education).

    2. Search UMI database (or your school’s database) for dissertations that relate to the topic of your interest.

    3. Read chapter1 (maybe chapter 2), and then read the section that outlines recommendations for further studies.

    4. Pick one of the recommendations and develop a title based on the recommendation.

    5. Write a one-page description of what you intend to do.

    6. Approach the faculty member that supervised that dissertation and have a discussion with him/her.

    7. And then ......

    NB: Remember to avoid trying to solve the world’s problems with your dissertation. Make sure that your goal is achievable. Your goal will be to ensure that your work improves or adds to the named field of inquiry.

    Ike
     

Share This Page