“If I have become great, it is by standing on the shoulders of Giants.”
Sir Isaac Newton
Newton probably said it best (or actually, whoever Newton was citing here): Science is not just one person discovering something, but — as with any creative endeavor — it must be new and useful. This means you must know the work that was already done (to prevent you from reinventing the wheel), and you must build on it and participate in the scientific discussion on the issue you are working on. To do this you need to know the right questions and how to answer them with methods that are up-to-date. As scientific communication is (usually) done in writing, this means you must find and read the relevant literature.
There are a lot of scientists who did and do currently work on topics that are similar to the ones you are interested in. In many cases, there are even multiple disciplines asking the same or similar questions — and answer them with the same to widely different methods. Depending on how hot the topic is there can be a lot of rapid communication going on — on conference, in workshops, in journals and special issues of journals, etc. Getting up-to-date on the topic you want to work on is no easy feat — but you need to do it to really contribute.
So, you want to stand on the shoulders of giants and have to take care not to be buried under a mountain of papers. How do you do it? The following tips might help:
- Consider the Topic: First of all, consider the topic you are working on. Does it really interest you? You have to read a lot about it — if you are not inherently interested in the topic it will be an uphill battle all the way — for years or even decades. See the topic posting for more information.
- Read with a sense of discovery: If you are interested in a topic, chances are you have some ideas of what to do. It can be very … painful to find out that the question was already answered and other researchers had the same or even better ideas. However, be open for the thoughts and ideas of others. Nothing is worse than closing your eyes to existing research and finding out too late that the question you researched was already answered — esp. if you consider that you could have contributed to the current state of research if you had followed their line of thought as far as possible and added your own ideas. Chances are, when you get to the end of their research you have some ideas to continue — or you have some ideas for a different path, or a turn on the road. That will be your actual research question — but you can only find it if you open your eyes to the works of others.
- Use the database: The first step in literature research is often using a database. Many scientific domains have their own specialized databases (e.g., PsycINFO in psychology). Finding the right keywords to search for can be difficult but there are usually help pages available (and look at the keywords in the articles you have already found). Other possible sources are Google Scholar (which you can also use to create alerts; or Google in general) or databases like ScienceDirect. Take notes which keywords and search terms you used.
- Select deliberately: In almost every topic in science there are hundreds if not thousands of papers and books (somewhat) related to your topic. Like Heinlein said in “Have Space Suit will Travel”: “If I had found out anything, it was that they could print it faster than I could study it.” Many students make the mistake of trying to read everything — and when they notice that they cannot keep up they try to read faster. Unfortunately, this ends up like Woody Allen’s: “I took a speed reading course and read ‘War and Peace’ in twenty minutes. It involves Russia.” It just does not work this way. You have to select deliberately which literature you read. Ask yourself: What do you expect from it? An overview? An idea for good dependent variables? Something else? Where are you at when it comes to your research topic? In general, first you want overview papers, later — when you have a mental model of the state of research (and know what you want to research), papers that answer the specific questions that you have (then). Whereas you should get all the literature you can get your hand on (in case you need it — take care to store it in a way you can deal with it, see next posting), if you are swamped beyond belief you might have to narrow down your topic.
- Screen before reading: Most papers have an abstract and a discussion/conclusion section. Read those first. Skim the other sections to gauge its value. You might find that — once you become familiar with the topic — you can skip most of the introduction/theoretical background and focus on the unique contribution of the paper, but this will take a while.
- Get an Overview first: Your first step should be to get an overview about the topic. You should start with literature reviews, meta-analysis, or even books on the topic. They will allow you to get an overview and then (later) focus on the aspects that interest you specifically, and will give you an idea about the important concepts, theories, empirical findings and works. Ask your adviser for literature, colleagues in the workgroup should have some common papers — otherwise question whether they really work on the same thing (see also Positions and Advisers whether you are in the right place). If you have no-one in your department who is an expert on the topic, consider asking externally for advice/tips. It is very helpful here to have someone who knows the topic and can separate the wheat from the shaft. On the other hand, take care that you do not narrow down your view to quickly, you have to add your own contribution (e.g., by combining it with something that others have not seen yet, asking different questions, etc.). Take care that reviews simplify findings from other papers and can sometimes interpret the results different than the authors had intended. You have to read the original literature (later), which will — in turn — lead you to other sources.
- Use (Overview) Papers as Seed Papers: Ideally the overview papers/books will provide you with the necessarily literature references and lead you to more specific literature. While often overlooked (and sometimes not copied by student assistants), the reference section is often the most valuable part in the first papers you read.
- Visit the researchers websites: Take note which people work in the (sub)domain you are interested in. Visit their websites to find out what they are currently working on. As good as literature reviews and books are, chances are that the researchers who wrote them or were cited in them have already conducted new research. Most websites of scientists will show you their recent papers, often including those which are submitted or in press, and — most importantly — which conferences they have visited. This is often the place where the current research is shown.
- Screen the conference websites/proceedings: The current research is often shown on conferences. Make sure you identify the conference that are relevant for your topic and screen them for relevant contributions. Many conferences have proceedings where the conference submissions (similar to a short paper) are printed. Specialized conferences are usually much more useful than the general conferences.
- Consider other sources: Especially in less organized, more independent areas (e.g., museum research) there are other sources that are easy to overlook. Check for organizations, workshops, special interest databases (e.g,. InformalScience), blogs, etc. pp. Many researchers blog, some disciplines have mailinglists or newsletter. Find them and subscribe to them.
- Monitor the Current Developments: It’s hard to keep up-to-date with literature once the daily tasks swamp you. However, there are a couple of helpful techniques. Make sure you get a good RSS-reader (many Mail programs can do it, personally, I recommend DEVONthink) and subscribe to the relevant blogs and journals. You are automatically informed when new papers are available for the journals that you monitor. It’s also possible to have alerts for specific search terms (e.g., “critical thinking” and “museum”) — you get an alert whenever a paper becomes available with these terms. More information in this posting. However, this likely covers only a part of the available outlets, so make sure you get into the community.
- Work your way backwards through the journals: Once you have identified and monitor the relevant journals and other outlets, work yourself backwards through the issues to find papers you might have missed with database searches/seed papers/by checking out the key authors. Often reading the title of the papers is enough for a quick check, if in doubt, check the abstract. Makes notes which journals you have checked back to which issue.
- Get into the community: A lot of information is shared in professional/scientific organizations and special interest groups. Find out what the relevant organizations are in your (sub)domain and join them as early as possible. You will get a feel for field and make contacts that can lead to crucial information, shared grant applications or research projects. You might also want to consider to join Mendeley [Update: Until it becomes clear how Elsevier treats Mendeley, I no longer recommend using Mendeley (currently looking for another solution).]. Whereas I am not a fan of this kind of literature management, the social networking function that is targeted at academics is quite interesting. You might find others who read the same literature as you do and work on the same topics — and they might have additional literature you do not know (yet).
- Keep the Global Perspective in Mind: It’s easy to get lost in the literature — while reading you can get lost in the details, so keep the global perspective in mind. Switch back occasionally to the global picture — it will help you to keep the use/goal in mind (and align your work to it). It’s also easy to be swamped by the literature. The next posting will deal with managing literature.
- Booth, W. C., Colomb, G. G., & Williams, J. M. (2003). The Craft of Research. (Second Edition). Chicago: The University of Chicago Press.
Next up: Managing Literature