Quick Takeaways:
Imagine you’ve just finished writing your research paper. Feeling proud, you get ready to submit it to your target journal. The submission system has an online form asking for the manuscript title and abstract. But wait! The next field asks for keywords, but you didn’t prepare any.
You’re tempted to leave that item empty or simply type the main words from the title. But then you realise keywords must have a special function, or the journal wouldn’t ask for them. You wonder: What do keywords do? How many should there be? Can keywords be two words, or more? Which ones should I use?
This blog aims to answer those questions. We’ll discuss strategies to identify appropriate keywords for your research paper and to maximise their effectiveness.
In general, keywords are simply the words you type in to perform a search online.
You’ve used keywords if you’ve ever typed words into a search engine to look for something online. The more matches that a search engine finds between a user’s keywords and an article’s content, the more likely that article will appear in the results. The keywords that you provide to a journal (hereafter, authors’ keywords) are used to improve this match.
As an author, you want people to find, read, and cite your work. Journal editors also want people to find, read, and cite their journals. Hence, journals typically ask authors to provide 3-8 keywords. These keywords will be listed on the article page and may be included in the hidden coding as part of the article’s metadata.
Authors’ keywords may even help database indexers find or create relevant indexing keywords. For example, in new research fields, the authors’ keywords may be more well-known among researchers because an appropriate official indexing word may not exist yet.
Keywords might also be used by the journal office for administrative purposes. Editors screen your work (including keywords) for relevance and fit with the journal’s aims, scope, mission, and readership. A good clear match will reduce the chance of desk rejection. The journal office may also use your keywords to find the most relevant handling/section editor or peer reviewers.
So, it’s worth investing time and attention to choose keywords that truly represent your hard work and will help make it findable in the future.
Keywords can be single words, but they can also be multiple-word phrases.
Appropriate keyword selection is both a science and an art. It should be done in conjunction with writing the title and abstract, as part of your article marketing strategy to entice readers to read your full paper.
You can choose the optimal keywords for your research paper or abstract by following these simple guidelines.
To identify keywords and choose the best ones, a good place to start is your target journal’s author instructions.
Your target journal may specify certain rules about keywords. Relevant issues may include:
⦁ Minimum or maximum number of keywords allowed
⦁ If keywords must be single words or if they can be phrases (possibly with a word or character count limit)
⦁ If abbreviations are allowed (avoid these anyway, unless they are well known)
⦁ If generic terms are preferred to brand names
⦁ Punctuation, spelling (UK or US), and capitalisation
⦁ If some or all of the terms must come from a special wordlist or database (specific to the journal, publisher, or discipline)
Regarding the last point, some journals will accept only approved terms using a controlled vocabulary. For example, if a journal insists on using Medical Subject Headings (MeSH) keywords, ‘skin cancer’ would need to be ‘Skin Neoplasms’.
For medical keywords, the online MeSH tool called MeSH on Demand is extremely useful. As a starting point, copy–paste your draft text into the text box and click “Search” to find relevant MeSH terms. In addition, the MeSH Browser gives up-to-date MeSH equivalents of possible key terms typed into the search box.
As another starting point, you could visualise the most frequent terms in your paper by inserting the text into a wordcloud generator. The size of each word in the wordcloud reflects the word frequency. Many free worldcloud generators, such as MonkeyLearn Wordcloud Generator (for phrases) and TagCrowd (for single words), are available.
Aim for a mixture of general and specific terms that potential readers will look for. However, avoid single words that are too general and vague to be useful, such as ‘Data’ or ‘Study’. Also avoid repeating keywords. If allowed, use a logical hierarchy or tree structure similar to that used by MeSH, where general terms are followed by specific aspects. Examples are ‘Skin Neoplasms / therapy’ or ‘Melanoma / pathology’.
Select keywords that closely represent and summarise your text, data, and illustrations. Some terms might clearly be main features of your study and paper, such as the topic, variables, population, and setting.
Remember to look at the bigger picture as well. What concepts, theories, or processes does your research paper cover, even if those specific names do not appear in the manuscript?
Say your paper is about sunblock effectiveness. The general concept of ‘Skin Neoplasms / etiology’ would be an appropriate keyword, despite not being mentioned in the text. Consider also whether the methodology or approach is a critical feature or a likely search term, such as ‘Cross-Sectional Studies’.
You need to think about the whole paper. The keyword list is part of your marketing strategy to attract readers. You want them to read the full paper (HTML or PDF file), which may be behind a paywall and available to subscribers only. The keywords should thus provide an idea of the topic and gist of the whole study, especially if only the abstract is freely accessible.
Some search engines do not search the whole text of a webpage or document (e.g., if the content behind a paywall), making your keyword choice even more important.
Your choice of keywords might have the effect of including or excluding certain readers. Bear in mind that:
⦁ Some people may not instantly think of the same formal or technical words as in your article
⦁ Your potential readers may be a mix of generalists and specialists
⦁ Your topic and terminology might be known by other names in different disciplines
⦁ Your study material (product, machine, drug) might be called something else in another country
Remember to test your keyword selection in the search engines and databases that your potential readers are likely to use. See if you to find similar papers to yours. Examine the keyword lists of similar papers to get ideas to refine your own selection.
Keywords represent your research paper and help make it visible from among the millions of other research papers in the literature. Carefully chosen keywords that match users’ likely search terms will improve the chance that your work will appear high up in search engine results.
Think about your keywords early on and keep a list of possible candidates while you draft your paper or conduct literature searches. Familiarise yourself with search engines in your field, as well as any special disciplinary word lists. At the same time, follow your target journal’s instructions and style closely. Finally, test and apply your selected keywords. Check you’ve included some, along with secondary keywords and synonyms, in your abstract and main text.
Our latest online workshop built on the success of face-to-face workshops we developed specifically for local universities. Over 30 faculty members joined the session, presented by our Chief Operating Officer, Mr Nick Case, to learn from our case studies on editing research proposals.
The response to our workshop, which included a constructive and insightful Q&A session, was very positive.Drawing on our extensive experience working with hundreds of Hong Kong researchers targeting the GRF and ECS every year, we used examples of poor and subsequently improved proposals to show the attendees how they can make their applications stand out.
Our latest online workshop built on the success of face-to-face workshops we developed specifically for local universities. Over 30 faculty members joined the session, presented by our Chief Operating Officer, Mr Nick Case, to learn from our case studies on editing research proposals.
The response to our workshop, which included a constructive and insightful Q&A session, was very positive.Drawing on our extensive experience working with hundreds of Hong Kong researchers targeting the GRF and ECS every year, we used examples of poor and subsequently improved proposals to show the attendees how they can make their applications stand out. The response to our workshop, which included a constructive and insightful Q&A session, was very positive.Drawing on our extensive experience working with hundreds of Hong Kong researchers targeting the GRF and ECS every year, we used examples of poor and subsequently improved proposals to show the attendees how they can make their applications stand out. The response to our workshop, which included a constructive and insightful Q&A session, was very positive.Drawing on our extensive experience working with hundreds of Hong Kong researchers targeting the GRF and ECS every year, we used examples of poor and subsequently improved proposals to show the attendees how they can make their applications stand out.
Wondering why some abbreviations such as ‘et al.’ and ‘e.g.’ use periods, whereas others such as CV and AD don’t? Periods are typically used if the abbreviations include lowercase or mixed-case letters. They’re usually not used with abbreviations containing only uppercase letters.
Dr Trevor Lane is a publishing and education consultant and an elected Council Member of the Committee on Publication Ethics. He has 25 years of experience helping authors publish their research in peer-reviewed academic journals.
Your Schedule, Our Prime Concern AsiaEdit takes a personalised approach to editing.
Suite 2101, 99 Hennessy Road,
Wan Chai, Hong Kong
9:00am – 6:00pm
(+852) 2590 6588
