The Literature Review is a chapter common to dissertations that report the results of an empirical study. I find that it is a chapter a lot of writers get stuck on. And it seems to me that perhaps they get stuck because they don't have a clear idea of what they're trying to accomplish in the literature review.
It seems to me that there is a common way of looking at a literature review: a report on relevant studies. But this common way of looking at it is not very useful. It doesn't help you find structure, or focus. How is the relevant distinguished from the irrelevant?
There is another way of looking at it: the literature review is a discussion of the sources in the literature that have shaped your study and have led to its current form, and a discussion of the major issues that those sources discuss and debate (including representation of salient opposing voices). I strongly believe that this is a more productive way of looking at the literature review. I also believe that it leads to a better written work--one that is more focused and less prone to getting lost in tangential details.
The literature is not meant to cover everything ever said about the general topic that you are using. It is not even necessary to cover everything ever said about a specific theory that you are using. It is necessary to cover the voices on which you rely, as well as the voices that provide a foil for the ones you wish to use: you want to use other sources that allow you to reflect on the sources you rely on the most. It is worthwhile to show opinions that disagree with the theories you are using: it provides an opportunity to discuss what is particularly salient about the ideas that you are using.
Start the literature with what you know and with the sources that you already have in hand. If you've only read three articles and your textbook, you can still write up a rudimentary literature review that will help you outline the ideas that you are using. If you have set up a good structure for the literature review, one based on the ideas that you are using, not one based on an attempt to describe all the different literature on the subject, then adding new sources to the literature review is pretty easy.
It seems like the literature review should be about reviewing what has been published, and therefore should somehow be structured by what has been published. But it's more productive to think about the literature review as discussing the ideas that provide a background for your work. What ideas led into your choice of method and focus within the topic? These are the questions that you have to answer in the literature review; that way the literature review follows up on the introduction--which gives the reader an overview of the whole scope of the project--by giving detail that the introduction couldn't accommodate, and then it sets up the methodology chapter that follows. The literature should be structured by your reasoning: what are the ideas that shape the study and how are these ideas debated in the discourse of your field?
I labeled this post under the "your voice" category, because ultimately the literature review is about revealing your voice, and showing how your voice reflects the larger discourse.