Using the right tool for your literature searching makes all the difference for your searching efficiency and for the quality of information you find. When you're searching for information, you want to feel confident that you’ve found the most important and relevant literature. Unfortunately, if you are not looking in the right places, even the best search won't find you the literature you need.
To consider where to search, it is helpful to think about the tools you use in terms of whether they are better for accessing some research or if they are better for discovering all the research you need.
Discovery means finding out about the existence of research. Access means getting the full text of research. It is a very common mistake to use tools that are better for access for the task of discovery, but doing so essentially means you are doing your research backwards, wasting time and negatively affecting the overall quality of your information.
Databases, and especially subject focused databases, are customised for discovery to lead you to the information you need.
The best tools for conducting your literature searches comprehensively and systematically are subject-specific databases.
Databases abstract and index the content of academic journals from multiple publishers, and when appropriate other publication types such as trade journals, standards, reports, conference papers and patents. They are designed for discovery—i.e. finding out that a piece of research exists and giving you the bibliographic details you need to find that piece of research.
The difference between subject-specific databases and general databases
Databases can either be multidisciplinary or focused on a discipline like chemistry, or sports and sports medicine, or the sciences of food and health. The focus defines the database's scope--what information is included within it--and how you find that information.
A subject focused database is built around a thesaurus of subject terms based in its discipline. Because of their breadth of coverage, multidisciplinary databases don't have thesauri, and this means that they are far more likely to return what are called false hits, or noise, where the term you search is not used in the sense you need.
Sometimes even when a database does have a thesaurus but one focused on a different discipline than where your topic falls, you will still get false hits with a term. Information retrieval in the area of food is complex because of the broadness of the field. For example:
Searching pig in a general database will bring back content where an animal has been used in preclinical trials, livestock research, and pork as food.
In a health-focused database, the search options and filters will have been developed for the human health field, which may not be helpful for searching food science topics not related to human health.
Searching spirits in PubMed (which does have a thesaurus focused on biomedical terms) or the multidisciplinary databases Web of Science or Scopus, (which do not have thesauri) you get moods and the supernatural mixed in with research focused on alcoholic beverages.
FSTA, focused on food science, not only doesn’t bring back supernatural false hits, it brings back many more relevant ones about distilled alcoholic beverages because of how each record has been tagged, or indexed, with the subject specific term spirits, even when that term does not appear in an article’s title or abstract.
What is indexing and why is it helpful in searching?
Databases that use a thesaurus, or controlled vocabulary, for indexing content pull all the different terms referring to a topic under a single heading. This helps users navigate the variations in language and terms used by researchers.
For example, in FSTA, if you search the thesaurus term aroma it pulls together all the results where the authors used the word aroma to describe an important element of the research, but also works by authors who used the words odor, odour or smell.
Similarly, research about Baijiu, Luzhou-flavor liquors, Luzhou-flavour liquors, Moutai liquors, and Moutai-flavor liquors are all gathered under the thesaurus subject heading Chinese liquors.
Some databases rely on machine learning to do the indexing, while others like FSTA have editorial teams of experts who do the work more accurately.
Search engines like Google allow you to find all sorts of information on the internet, but they are not designed specifically for finding scholarly information, so are terrible for literature searches.
However, they are good for finding governmental information like U.S. Department of Agriculture research funding instructions, scientific reports from the European Food Safety Authority, or guidelines from organizations like the World Health Organization. Academic search engines and most databases do not include this type of document. The database FSTA is an exception, since it indexes legislation, standards and reports (but not funding instructions).
Unlike general search engines, academic search engines like Google Scholar do focus on scholarly information, but they:
DO NOT allow precise control over searches, even with advanced search options.
Search engines can be useful for accessing the full text of articles and patents but using them for discovery is an inefficient--and potentially hazardous--way to research.
Platforms like Academia.edu and ResearchGate allow researchers to create profiles to showcase their work and share their articles. Both can be useful for acquiring full-text articles; however, because researchers create and maintain their own profiles, searching these platforms will not give you a comprehensive overview of a field—you’ll only find the work of researchers who have chosen to participate.
Don’t confuse these platforms with discovery services, such as databases, which are specifically designed to be comprehensive in the subject area they cover in order to help researchers find relevant information.
Wikipedia can be a good place to start building knowledge and start gathering terminology around a new subject. Most of the time the information found on Wikipedia is accurate, though how well-developed entries are varies across disciplines.
Bear in mind, however, that Wikipedia should only serve as a starting point for a search. There is always a risk, because any one can edit any article, regardless of their expertise, that the information is not accurate, or is outdated or somehow skewed. If a Wikipedia entry has references, those are generally worth following, and they can also give a sense of useful terms to use in a comprehensive search on the topic.
Never cite Wikipedia directly--you never know when an entry will be edited and if what you've cited will still exist.
BEST PRACTICE RECOMMENDATION: familiarise yourself with the databases you have access to and invest time in learning how to search them, including using the thesaurus functions if available.
BEST PRACTICE RECOMMENDATION: always consider whether or not you are using the right tool for finding the information you need.
BEST PRACTICE RECOMMENDATION: remember that research for a literature review is a two-step process—first is discovery of research, and the second is accessing the research you’ve determined you need. Don’t switch the order of the steps! If you limit your search to the research outputs that you think that you have easy access to, you will almost certainly end up with a biased review that is neither systematic nor comprehensive.