Keyword Generation
Once we have our concepts mapped out, we can start generating keywords. Keyword generation is an important next step. Not every researcher is going to use the exact same language in their article, and so it's important to use a variety of terms in your search to make sure that you don't miss out on valuable evidence.
You can start by thinking of synonyms on your own, or you can use a tool like ChatGPT or Gemini to help you generate keywords. Additionally, the database Embase is great for generating synonyms, especially for clinical interventions. See these methods in action, check out the video below:
Controlled Vocabulary
To make our search as effective as possible, we need to add controlled vocabularies to our search, which are different depending on the database that you're using. Embase uses Emtree terms, while PubMed used MeSH (Medical Subject Headings) terms.
Check out the video to see how to add controlled vocabulary terms to your search.
Putting it together
Next, we'll combine our controlled vocabulary with our keywords to perform a search in the databases:
Common mistakes
Below is a checklist that can be helpful to consult in case you run into errors (you aren't finding relevant results, you have too many results, you have too few results) as you look through the databases.
| Checklist Questions | Yes | No | How might you improve? |
|---|---|---|---|
| Does the search strategy match the concepts you mapped out? | |||
| Are Boolean operators used correctly (AND and OR operators) | |||
| Are the subject headings relevant? | |||
| Are the keywords relevant? | |||
| Are words and phrases spelled correctly? | |||
| Are all limits and filters used appropriately? (Ex: Limiting by year, or filtering by title/abstract) | |||
| Was the search strategy effective and efficient? |
