It can be as broad as you like, because this is just a starting point. If you are still picking your specific topic for your PhD, that’s fine, but you should at least know roughly what area you want to explore.
A quick google scholar search for your subject area could turn up as many as 1 million results. Clearly you can’t read them all, so you need to look for an easy way in.
The vast majority of academic papers are written for people already familiar with the subject. They will refer to theories and methodologies assuming that the reader knows what they are.
So to start with just any paper at random would be a demoralizing waste of time, as you’ll be overwhelmed by the jargon. Instead, you need something you can understand easily to give yourself a foundation of knowledge to build upon.
Textbooks and review articles can be good places to start, though even these can be highly technical. If you can’t find one you can understand easily, then look for a book written for the general, non-academic public.
The idea is to gain a quick, broad background knowledge before getting into more specialised technical detail.
The idea of a literature review is to give some background and context to your own work. You need to show how your research fits into the big picture, relating it to what has been done before.
You don’t need to write a comprehensive history of your subject, but it helps if you know roughly how it has developed over time.
There may be several angles you can take in your research, and you may have to explore many areas of the literature. So divide your literature search into sections to make it easier to manage. For each section, think of several keywords to try out in different combinations.
Even when you look at highly specialised sub-topics, there may still be thousands upon thousands of papers, so you need to filter them. Here are a few ways to reduce the numbers:
• Look at the number of citations as an indication of quality
• Make your keywords more specific
• Scan the abstract and make a quick decision as to whether it will be relevant or not
Don’t be afraid to reject papers. You can always come back to them later, but you have to start with something manageable.
You might not be able to read everything in depth immediately. From the papers you selected, give them a ranking A, B, or C.
A = must read, highly relevant, high quality
B = unsure, probably relevant, but not yet sure how
C = probably irrelevant, not what you thought it was when you read the title
If you’ve printed them , put the letter A, B, or C on the front so you can tell quickly when you come back to them (maybe months or years later)
Even if you can only find one good quality paper, read the introduction carefully and see who they cite. There may be a few gems there you didn’t find with the search engine.
Also see who else has cited that one paper since it was published (this is also a very quick way to update your bibliography if you are coming back to it a year or more later).
In any research field, no matter how specialised, there will be leading experts or competing research groups. Figure out who they are, and read their work.
As the saying goes, you can’t prove a negative. How can you prove that nobody else has done what you plan to do, without searching every paper ever published?
Well, it’s worth spending a day or two searching every keyword combination you can think of related to your specific research plan.
When you finally start writing your literature review, focus on ideas and use examples from the literature to illustrate them.
Don’t just write about every paper you have found (I call this the telephone-directory approach), as it will be tedious to write and impossible to read.
The aim should always be to cite the best and most relevant research, rather than going for sheer quantity.
So you can leave out big chunks. Write about what is relevant to your research.
When talking about a broad topic, only cite the very, very best papers. You’ll have a lot to choose from , so why choose anything but the best?
Then when you get into more specialised sections, you can include a larger number of less well-known papers (but still the highest quality you can find).
Don’t cite anything you haven’t read or don’t understand
Your perspective on the literature will be quite different once you have done your own research. If you are in your first year, get your literature review done quickly so you can move on with your own work, and don’t let it hold you back.
It takes time to figure out what makes a good paper and what makes a bad one, and that comes with experience of carrying out research, talking to other researchers, and just reading more.