by using current information, either from others who have been involved in similar research, or using texts, papers in journals, or other information sources (e.g. the internet), or a combination of these. Be aware of possible biases in the information used, especially where this is obtained from websites belonging to individuals (rather than respected organisations) that have not been independently validated. Most papers in reputable journals and many textbooks have been examined by independent referees, although even these may contain factual inaccuracies and personal opinions that may not conform to current opinion. Although considered the gold‐standard of information sources, even peer‐reviewed journals are subject to bias against the publication of negative results. It is important to start your review of the literature as early as possible, since it is an ongoing process throughout your research and should inform each stage of your project. At the very least you should begin by reading the literature to establish that your proposed idea has not been already published and to define the gaps in knowledge that you will attempt to fill. It is likely that as you read one paper, you will find references to other work that may be important. Expect to read more than 100 papers in your field and to actively use a third to half of these in your thesis.
If you are new to a subject matter, you should first try and locate seminal piece(s) of work in the field or a recent review. Typically, this will be close to the top of a search list of highly cited papers and can be found by ordering a search by ‘times cited’. Take a detailed look at the seminal paper(s), the reference list, and who is citing that paper. In journal databases (e.g. Web of Knowledge), citation networks can be viewed to examine the connectedness between a seminal paper and all those papers that cite it. This is useful because it can elucidate key papers in the field and reduce the search effort dramatically. Typically, your first search should include seminal works and a collection of the most recent papers in the field (i.e. from the last few years). It might be helpful to order these by journal impact factor (if available), since parochial journals may not contain as high quality science, although sometimes smaller research papers with less apparent impact can provide valuable information in the form of species lists, new methods, and negative findings that are often not reported in more mainstream journals. An additional word of warning: highly cited papers can also be poor papers in the field since other authors might simply be referencing them to make an example of that piece of work (e.g. ‘Black and White's (2000) experimental design has been shown here and by others to be flawed’). Knowledge of the literature can assist in avoiding ‘blind alleys’ and unfruitful lines of enquiry or techniques. There are two main types of literature: primary and secondary.
Primary literature
This is first‐hand information; for example, articles in specialist journals, reports, MSc and PhD theses by those who have done the work and generated data. Journals that publish only refereed papers (i.e. those that have been through a peer‐review process) are the most important sources of primary, up‐to‐date information, and where possible your literature review should focus on this type of source. Other primary sources include technical reports, management plans, consultancy reports, and species lists (e.g. from annual recorder reports). All of these can be useful sources of information for ecological projects, but you should be aware that they may not have been edited or their quality controlled.
Secondary literature
This is prepared from other sources of information – including textbooks, review articles, opinion pieces, and letters to the editor – which often express views on primary research. If you are lucky, there may be specialised books covering your subject area. These may provide a good starting point since books are secondary sources of information, while journals are a preferred source of reference for most research projects. The coverage in student textbooks is usually rather more superficial than that in specialist texts. If there are review articles on your subject, these may be useful to obtain an overview and as a source of new references. Reviews are found in edited book sections, journals that specialise in reviews (e.g. Trends in Ecology and Evolution)3 and journals that have occasional review papers. Review papers in established scientific journals have usually been subject to peer review.
Other sources of information
Maps, personal notes, databases, museum collections, and archives may all contain information that can be useful in supporting your research. There are several software systems used within professional ecology that can be useful for research projects. These can help to set your work into a context grounded in practical conservation issues, as well as supplying data on either a wider spatial or temporal basis. One such package, the Conservation Management System (CMS) software,4 is used by many countryside managers to produce management plans, whilst Recorder5 is the software used by the UK National Biodiversity Network6 to record, manipulate, and map biological records. Note: software is often subject to changes in version and may indeed stop being supported, either by the producers or by computer operating systems, over time.
Search terms
The best way to locate relevant papers is to systematically search journals using your library's electronic databases using key words. The use of targeted search terms can identify appropriate works and avoid too many articles or irrelevant papers from being selected. First, identify a list of key words or phrases that could be included in the title, abstract, or key words of an article. Begin by using simple combinations of terms (e.g. woodland beetles, hedgerow birds). If you are finding too many papers, then either restrict your search to more recent publications or use more complicated combinations (e.g. by adding ‘predation’ to the key terms ‘woodland beetles’ you can significantly reduce the number of articles returned by a search). Note that most databases allow wildcard entries (e.g. an asterisk) to truncate key works (so that space, spaces, and spatial can all be covered by spa*), although such terms can increase irrelevant returns (e.g. spa* will also cover any word beginning with spa; e.g. Spain, Spanish, etc.). You can use Boolean operators to refine your search and reduce the number of irrelevant works identified. For example, the search ‘woodland beetles’ OR ‘hedgerow birds’ will select those papers which have either ‘woodland beetles’ or ‘hedgerow birds’ in the title, abstract, or key words. Conversely, the search ‘beetles’ AND ‘predation’ will select only papers with both terms in them. Once appropriate papers have been acquired, you can use them to find others of interest in two ways. First, you can examine the literature that the paper cites to see if any older works are relevant. Second, you can use most search engines (e.g. Google Scholar) to identify those works that come after the paper in question and have cited it. If you have access to a librarian, they can help you to set up a search. This search can be saved and re‐run during your project so that you can detect any new papers published whilst you have been carrying out your research.
Reading papers
You will find more source papers than you have time to read. It is easy to get bogged down in the wealth of published material. Keep your subject area in mind and do not read everything indiscriminately. Skim‐read a new reference to decide how much attention it deserves. Start by reading the abstract, skimming the subheadings and then the first paragraph or so of the introduction and the last paragraph of the discussion. Only read in detail those papers that are particularly relevant.
Keep a copy (photocopy, scanned image, or electronic copy) of key references and make notes of (or highlight on photocopies) any useful information. Save the full reference, since all material cited in a research report must be listed in full in the reference list; there are few things as annoying as having to re‐find the details of a reference that you
