As you start exploring your topic, it is useful to read around it for some background knowledge. Prof Johannes Cronje has some good tips on his free online doctoral programme:
Get introduced to the field.
Start with Wikipedia. Do not listen to the snobs who say you shouldn't. It's how you use it that matters. You begin by reading the article on the topic of your choice. If it is completely strange and hard to understand, then you cannot do your doctorate in that field. Then, if you feel that you are so familiar with the contents that you could have written it yourself, you proceed. Now click on the "Talk" tab at the top of the article, and read the comments made by the various participants in the article. It is here where you will start finding the issues that may or may not be worth researching. What debates are going on there.
Find out who's who in the field.
Take note of the authors who commented on the Wikipedia article. Then scroll down to the reference section of the article itself and make a list of all the authors cited. Now download Harzing's Publish or Perish and check the "H Score" of each author. Make a list of these authors and sort them according to this score. Now take the top five authors and for each, determine which are the top-scoring journals in which they publish. Now, when you have the top five authors, and the top five journals, identify the top five articles published by each of the top five authors, and identify the top five articles published in each of the top five journals. Now copy the keywords of all these articles into a spreadsheet. Sort them alphabetically and determine the five most popular keywords. You may want to draw a graph of these. Now take the abstracts of all the articles and paste them into Wordle. The resultant word cloud will give you a good intial view the key concepts of your field.
Ask the right questions.
One of the most common mistakes made by doctoral candidates is to ask their potential supervisors for a topic. Otherwise, their initial research questions are ones to which we already have the answers. Check to ensure that you do not fall into the second category by typing your own research question, as you formulated it, into Google and see how many times your question (and its answer) actually appear. A good research question is one that is both relevant and unique. To find one you need to know what other people have asked (i.e. relevant) and what other people have not asked yet (i.e. unique). To do that you need to analyze the articles that you have identified even futher. Make a table. In the first column, put the full citation of the article. in the next column, put the keywords, then the main research question, followed by the answer to that research question, and finally the author's recommendation for further research. Now cluster the questions together around common themes, and arrange them in the order in which they appeal to you.
Meet your heroes.
Register with Academia.edu and Researchgate.net and follow the key authors that you have identified. Get to know who else is following them and get to know whom they follow. Find out which conferences they attend and what keynote addresses they have delivered. Visit their websites, and generally form a picture of the people whose work you follow.
An excellent lecture by CPUT's Dean of the Faculty of Informatics and Design.
Mendeley is a free reference manager and academic social network that can help you organize your research, collaborate with others online, and discover the latest research. It's a tool that helps you store your readings, keep organized and reference correctly.
Get Mendeley at www.mendeley.com.
For help in getting set up, see the Mendeley LibGuide.
The sources below are databases where you can check on the uniqueness of your topic.