Literature research

We recommend two search engines at the moment – google scholar and pubmed. Sometimes, if you are dealing with exotic species, Zoological Records directs you to articles you would not find otherwise.
The important step is to select good “search terms”. There are no general rules apart from that you should try as many as you can think of.
The papers you find in this way will then provide you, most likely, with further literature. This goes in both directions – to the past, in terms of the reference list of sources cited; and to the time following that paper, in terms of other papers that cited it.
Both, the reference lists of the papers found in the first step, and the list of material that cited those papers you found in the first step, are the most important sources for your literature research. In order to be efficient, follow the strategy
1. Search forward:
a) check the title, or abstract, of what you found using your “search terms”, and if it is relevant for you, check out what other, evidently more recent sources, cited this work.
b) do this over and over again until you hit those sources that are so recent that they have not been cited yet.
c) remember to do this several times during your research project, to check out what new papers have been published since your last did the literature research.
2. Search backward:
a) Check the abstract – especially the last 2-3 sentences where the conclusion is – is the paper relevant for you?
b) Scan the Introduction for references relevant to your topic
c) Scan the Discussion for references relevant to your topic
d) Put the paper away, search for the references, repeat the process with the next papers you get
e) Later when you have a small collection, sort the papers mentally, then read the abstracts and the text body again carefully
As a rule of thumb, I scan all pages (up to a max. of the first 30 pages) of hits in pubmed, and the first 5-8 in google scholar. Do not stop after the first 1-2 pages. If in pubmed you have an excessive number of hits, check the titles and figure out how to refine your search by using different search terms.
You will realize that you are on a good track once you recognize most of the references in the list of a paper.
Do not trust a negative finding – usually, everything has already been published on at least once.

It is good scientific practice that you follow important information back to its original source. For some information, you could use a review: For example, if your own study is on the length of the tongue on ruminants, and you want to link this data to information on the natural diet of ruminants, it would be acceptable to use, as a source for the natural diet, a review paper (although it is always best to do everything with original data, sometimes this may not be feasible). However, in the example mentioned above, it is not acceptable to use data on tongue length from a review paper - because in this example, this is your topic! Which means you have to check whether anything that is written in a review actually matches the original sources. This can be a cause of delay for a study (if you first use a review paper and your supervisor only realizes this quite late, making you then repeat the data collection from the original literature). Always ask your supervisor when you want to use a review source.