At the moment I’m working on a PhD, so my methods are biased towards resources available at a major research university. I have a list of different things to try when I want to be as comprehensive as possible. I’ll flesh out my list in more detail. You can do many of these if you are not at a university, e.g., if you can’t access online journal articles, try the Less Wrong Help Desk.
In terms of sources, the internet and physical libraries will be the main ones. I wrote more on the process of finding relevant prior work.
This process can be done in any particular order. You probably will find doing it iteratively to be useful, as you will become more familiar with different terminologies, etc.
Here are some things to try:
Searching Google, Google Scholar, and Google Books. Sometimes it’s worthwhile to keep a list of search terms you’ve tried. Also, keep a list of search terms to try. The problem with doing this alone is that it is incomplete, especially for older literature, and likely will remain so for some time.
Searching other research paper databases. In my case, this includes publisher specific databases (Springer, Wiley, Elsevier, etc.), citation and bibliographic databases, and DTIC.
Look for review papers (which often list a lot of related papers), books on the subject (again, they often list many related papers), and also annotated bibliographies/lists of abstracts. The latter can be a goldmine, especially if they contain foreign literature as well.
Browsing the library. I like to go to the section for a particular relevant book and look at others nearby. You can find things you never would have noticed otherwise this way. It’s also worth noting that if you are in a particular city for a day, you might have luck checking a local library’s online catalog or even the physical library itself. For example, I used to live near DC, but I never tried using the Library of Congress until after I moved away. I was working an internship in the area one summer after moving, and I used the opportunity to scan a very rare document.
Following citations in related papers and books. If something relevant to your interest was cited, track down the paper. (By the way, too many citations are terrible. It seems that a large fraction of researchers treat citations are some sort of merely academic exercise rather than a way for people to find related literature. I could insert a rant here.)
Searching WorldCat. WorldCat is a database of library databases. If you’re looking for a book, this could help. I also find browsing by category there to be helpful.
Asking knowledgeable people. In many cases, this will save you a lot of time. I recently asked a professor a question at their office hours, and in a few minutes they verified what I spent a few hours figuring out but was still unsure of. I wish I asked first.
Looking for papers in other languages. Not everything is written in English, especially if you want things from the early 20th century. If you really want to dig deep, you can do this, though it becomes much harder for two reasons. First, you probably don’t know every language in the world. OCR and Google Translate help, thankfully. Second, (at least in the US) many foreign journals are hard to track down for various reasons. However, the benefits could be large, as almost no one does this, and that makes many results obscure.
It should be obvious that doing a detailed review of the literature can require a large amount of time, depending on the subject. Almost no one actually does this for that reason, but I think it can be a good use of time in many cases.
Also, interlibrary loan services can be really useful for this. I submit requests for anything I have a slight interest in. The costs to me are negligible (only time, as the service is free to me), and the benefits range from none to extremely substantial. You might not have access to such services, unfortunately. I think you can pay some libraries for “document delivery” services which are comparable, though maybe expensive.
Finally, you probably would find it to be useful to keep notes on what you’ve read. I have a bunch of outlines where I make connections between different things I’ve read. This, I think, is the real value of the literature review, but verifying that an idea is original is another value you can derive from the process.
At the moment I’m working on a PhD, so my methods are biased towards resources available at a major research university. I have a list of different things to try when I want to be as comprehensive as possible. I’ll flesh out my list in more detail. You can do many of these if you are not at a university, e.g., if you can’t access online journal articles, try the Less Wrong Help Desk.
In terms of sources, the internet and physical libraries will be the main ones. I wrote more on the process of finding relevant prior work.
This process can be done in any particular order. You probably will find doing it iteratively to be useful, as you will become more familiar with different terminologies, etc.
Here are some things to try:
Searching Google, Google Scholar, and Google Books. Sometimes it’s worthwhile to keep a list of search terms you’ve tried. Also, keep a list of search terms to try. The problem with doing this alone is that it is incomplete, especially for older literature, and likely will remain so for some time.
Searching other research paper databases. In my case, this includes publisher specific databases (Springer, Wiley, Elsevier, etc.), citation and bibliographic databases, and DTIC.
Look for review papers (which often list a lot of related papers), books on the subject (again, they often list many related papers), and also annotated bibliographies/lists of abstracts. The latter can be a goldmine, especially if they contain foreign literature as well.
Browsing the library. I like to go to the section for a particular relevant book and look at others nearby. You can find things you never would have noticed otherwise this way. It’s also worth noting that if you are in a particular city for a day, you might have luck checking a local library’s online catalog or even the physical library itself. For example, I used to live near DC, but I never tried using the Library of Congress until after I moved away. I was working an internship in the area one summer after moving, and I used the opportunity to scan a very rare document.
Following citations in related papers and books. If something relevant to your interest was cited, track down the paper. (By the way, too many citations are terrible. It seems that a large fraction of researchers treat citations are some sort of merely academic exercise rather than a way for people to find related literature. I could insert a rant here.)
Searching WorldCat. WorldCat is a database of library databases. If you’re looking for a book, this could help. I also find browsing by category there to be helpful.
Asking knowledgeable people. In many cases, this will save you a lot of time. I recently asked a professor a question at their office hours, and in a few minutes they verified what I spent a few hours figuring out but was still unsure of. I wish I asked first.
Looking for papers in other languages. Not everything is written in English, especially if you want things from the early 20th century. If you really want to dig deep, you can do this, though it becomes much harder for two reasons. First, you probably don’t know every language in the world. OCR and Google Translate help, thankfully. Second, (at least in the US) many foreign journals are hard to track down for various reasons. However, the benefits could be large, as almost no one does this, and that makes many results obscure.
It should be obvious that doing a detailed review of the literature can require a large amount of time, depending on the subject. Almost no one actually does this for that reason, but I think it can be a good use of time in many cases.
Also, interlibrary loan services can be really useful for this. I submit requests for anything I have a slight interest in. The costs to me are negligible (only time, as the service is free to me), and the benefits range from none to extremely substantial. You might not have access to such services, unfortunately. I think you can pay some libraries for “document delivery” services which are comparable, though maybe expensive.
Finally, you probably would find it to be useful to keep notes on what you’ve read. I have a bunch of outlines where I make connections between different things I’ve read. This, I think, is the real value of the literature review, but verifying that an idea is original is another value you can derive from the process.