Pick an area of interest. Some of you may have free reign when it comes to your topic, others will have it dictated by their home university. Make sure it is viable- there is no point in you picking an area that is too wide nor too narrow; speak to your tutor about it, that's what they're there for! Read up about the subject on the news, on blogs, in the library archives - you can even try Twitter, you might come across a gem or two!
You then need to make sure you understand your subject area, there's no point in reading up about something and trying to formulate a 5,000 word essay out of it, if you don't see the line of argument, the pros and cons, and most important of all, a clear conclusion to your essay title.
Then, after much trepidation, some light (and not so light) reading, and many a-coffee, you'll be ready to dress up a plan, the blueprint behind your dissertation. This is without a shadow of a doubt of utmost importance, as without a plan, you won't be able to blurb on for long, without going round in circles. Some of you may sneer, but many students have got a considerably lower grade in their dissertations due to the lack of STRUCTURE which is key to developing an argument. Here are some tips about how it should all bind together:
1) Title page
Not to be confused with the excuse of spending 30 minutes choosing which font will look best. You do need one though, with the ahem, title, your name and the date, all clearly laid out.
It shouldn't be too long (about 150-300 words) explaining what the reader will find in your dissertation, in summary form. It's best not to use any quotations in this bit, it should be all about your writing and your argument.
Write out a clear list of headings and subheadings, with page numbers, so as to make it easier for the reader (and yourself) to read through the piece of work.
Here, you need to first outline what subject exactly you're going to tackle; give a bit of background information, as it were, to put the reader at ease with what's at the centre of the dissertation. You need to map out what your objectives are, how you wish to tackle the issue, what you're going to do and how you're going to do it. You need to guide your reader.
5) Main body - evidence for and against
Here is where a lot of students get lost; where to begin, what with, how to structure it etc. Remark the cardinal rule: each argument needs evidence to support it, an example to illustrate a point. Whatever you write, you need to back it up with either a quote, a finding, a statistic etc. Whatever evidence you have amassed, use it, but use it effectively. There's no point in quoting Dr So-and-so and countless journals and periodicals if they are not speaking of your subject matter. So many students get bogged down with textbooks, citations and referencing (here's a pretty extensive list of how to reference, courtesy of the Anglia Ruskin University). Make sure your arguments have a clear trajectory and that you link them effectively. Whether you're writing in English, French or Esperanto, you need to get some linking words on the go, to make it all stick together. Most students prefer to have a sort of 'pros' and 'cons' approach, so arguments first in favour of the issue, then arguments against. You can criss-cross, but make sure you refer back to each argument whilst doing so, as you may just end up confusing the reader. This section is the largest in the work, so you need to make sure it goes with both your introduction and your conclusion, fitting in perfectly with your line of argument. You're looking in lyrical terms for an hour-glass shape. Would Beyoncé look as good if she was disproportioned? Nope. And nor would your dissertation.
6) The big finale - conclusion/recommendations
Don't skimp on the words here, but don't babble on either (you should have done that in the body of your essay). Make it concise, punchy and answer your initial question(s). You can, in some cases, expand the subject matter so as to 'recommend' as it were another area to further research on. This is a strategy that is particularly common in French dissertations, for example.
As we've mentioned before, the Anglia Ruskin University outlines the do's and dont's of Harvard referencing. There are no footnotes or endnotes here; this is to show the reader where you've got your information from. Don't add books/journals that you may have read, but, ultimately, found completely useless to your bibliography. You're wasting words, discrediting your work and it's not going to work in your favour. However, make sure that any piece of information that you have lifted/gathered across your research and slipped into your dissertation is mentioned, credited and credited properly. Plagiarism is not what you want to be stuck with.
Finally, you may want to have someone read over your dissertation. You may even find it more useful to have a friend/parent/teacher who is not familiar with the subject to give it the once-over, as they'll be able to give you some feedback as to structure and content. You also need to make sure your spelling, grammar and punctuation are up to scratch, as you will get marked down for this (and considerably so if you intend to write a dissertation in another language).
So, with all this info, you should be able to write a piece of literature Dante would be proud of! Good luck!