I'm here for another year, but need to restructure my thesis and draw up more rigourous research questions. I'm beginning to do so - but any tips people. I feel like have lots of interesting sources and vague notions but seperating them into chapter with 'points' seems difficult..I'm finding it difficult to develop threads that follow through...

Start writing - even if you chuck these drafts away, it will focus your thoughts and make you see how your ideas will fit together. You can't tell how your arguments will work until you start to use them.

So true. Also talk endlessy to people - I've often heard ideas coming out of my mouth that I didn't realise I had until I spoke them (if that make sense).

All good advise above. One suggestion I had on a course about overcoming writers block was to 'free write'. Take a piece of paper and start writing and try not to stop, let the pen follow your train of thought. I was very suprised at how cogent the writing that came out was. I'm sure a cursory Google search will give you more information on this technique.

Hi,

I agree with the above, yet I think that it is also vitally important to concentrate just on structure! Once you know the structure your writing will be a lot better. The difficulty being that you do not know at this stage what your structure is.

I use the following method: first just brain storm: write whatever you like, use broad ideas and put them randomly on paper.Also ask other people to provide other ideas. Then look at how these are interrelated or subparts of other ideas. Re-arrange, look again, add, re-arrange etc. until you really know how you can put them in different catagories. Once you have done this start writing the actual text.

i agree with the brainstorming-then-structuring approach. another way would be to go from large to small: for example from broad topics to subtopics, or from centuries to decades... think about if you want to proceed "chronologically" or by "topic/theme"...
another thing i find useful:
distinguish between "type a questions", "type b questions", and "aspects of the case" (where type a is very general - you want to address but not answer this question as it is too broad, type b is more specific - the questions you want to answer so that you can address type a, and aspects are the things you actually need to find out in order to answer type b)

or you can think about the narrative structure: if you are studying an event, you can distinguish between things that happen before the event, the event itself, things that are contemporary to the event, and outcomes after the event.

Yep, I do the same as Rick & Shani described; Initially just start typing a list of questions (What, how, who) under my section or chapter heading, then start answering those questions. While answering those questions I get many ideas and remember many other concepts, important statements and start writing in pointers; then organize and expand those pointers, answers. Main key to start a structure is, I think, to start writing what you want to, your ideas and not what your advisor/supervisor expectations. Once the structure is formed then work on those expectations.