A comprehensive and systematic literature search will underpin your literature review and, by extension, your entire research project. So, it’s important to construct one which appropriately reflects your topic. Developments in technology have improved accessibility of research, but you still must apply a rigorous, structured, effective, repeatable, and achievable methodology. 

This guide will be especially useful for PGRs and anyone at UWS conducting research.

A well-constructed search should strike an appropriate balance between sensitivity and specificity when searching for literature and often involves iterative refinement and testing. Taking the time to do this at the beginning of your work will pay dividends later.

Start by speaking to an Academic Librarian or Research Librarian - they can help you develop your question and build a strong search from the start of your research.

Cooper, C., Booth, A., Varley-Campbell, J., Britten, N. and Garside, R. (2018) 'Defining the process to literature searching in systematic reviews: a literature review of guidance and supporting studies', BMC Medical Research Methodology, 18(1), pp. 85. Available at: https://doi.org/10.1186/s12874-018-0545-3 (Accessed: 23 January 2024).

A search strategy is a planned way of finding research by entering a combination of text words and special characters into a database. These are chosen to match the words used in article titles, abstracts, and subject headings, helping you find the most relevant records without too much irrelevant material.

It’s important because a well-designed search strategy helps you find the right research without missing key studies or being overwhelmed by irrelevant results. It makes your search more efficient, repeatable, and transparent - especially if you're doing a systematic review or academic project where you need to demonstrate how you found your evidence.

For instance, as you engage in your search, you may:

• Discover fresh keywords, synonyms, or substitutes.
• Encounter situations where your results are excessively abundant or insufficient, prompting adjustments to your search strategy or the choice of search sources.
• Realise that your obtained results lack the desired relevance, necessitating the fine-tuning of your search criteria.
• Need to tailor your search approach to accommodate the unique characteristics and constraints of different databases or search engines.

Before you begin searching, it’s important to set some boundaries. This helps you stay focused and avoid being overwhelmed by irrelevant material.

Start by making sure your research question is clear and well defined - this will shape everything that follows. Then think about:

• What kinds of sources will be useful? (e.g. journal articles, books, reports, policy documents)
• How recent should they be? (e.g. only the past 10 years, or a wider historical view?)
• Are there specific inclusion or exclusion criteria you already know? (e.g. certain countries, age groups, or study types)
• Will your review be a brief overview or a more comprehensive project?

Do some early scoping searches in tools like One Search (the UWS library catalogue) to get a feel for your topic and test whether your idea is feasible. From there, keep a simple log of your searches, draft search terms, and the databases you've used – Word, Excel, plain text or a notebook is fine.

Finally, consider how you plan to present what you find. Are you mapping out key themes? Looking for patterns? Comparing approaches? Knowing this in advance can guide both how you search and what you include.

Tip: Scope ≠ Search Strategy. Setting your scope isn’t the same as running a search - but it lays the foundation for one. It defines what you're looking for, and why.

Once you're done searching, review how well your strategy worked. Were your results useful? Is anything missing? Finally, save a copy of your final strategy - as an appendix or a standalone document - so that others can see how your search was done and build on it in future work.

1. Author Inclusion Check
Take a known author - ideally someone prominent in the field or from a paper you found in your background searching - and run an author search in the database. Then combine that line with your full search strategy using AND.

• If relevant papers by that author appear in the results, it suggests your search is sensitive enough to capture known, relevant work.

• If they don’t appear, revisit your concepts and keywords - something may be missing.

2. Title Inclusion Check
Use the exact title of a key paper you've already identified and run a title field search (e.g., TI "The revolution will not be televised: Spoken word, protest, and post-civil rights rhetoric" . You should get a single result (n=1). Then, AND this line with your full search strategy. Make sure that reference is in the database you're using!

• If the result is still n=1, it shows your strategy is picking up that key paper.

• If the result is n=1, your search may need adjusting- it could be missing important known material.

3. Concept Line Isolation
Run each concept or Boolean search line from your strategy separately to check that it returns a reasonable volume of results and behaves as expected. Testing each concept separately helps you spot problems before combining everything together. It’s much easier to fix an issue at this stage than after you’ve built the full strategy.

• If one line is returning too few results, it might be too narrow or missing key terms.

• This is especially useful before combining all lines into a full strategy.

4. Relevance Spot-Check
Scan the first few pages of your combined search results. Pay special attention to the record titles and abstract fields as you look.

• Are at least some of the records relevant to your topic?

• If not, it could indicate problems with specificity (too broad) or concept alignment (missing the mark).

5. Compare Against a Known Review

If there’s a published systematic review on a topic similar to yours, it can be a helpful reference point. You can try replicating or adapting their search strategy in the databases you’re using to see what results it brings back.

You don’t need to copy it exactly, but you can:

• Use it to test your own search and check whether your strategy retrieves some of the same key papers.especially if their strategy has already been peer reviewed or published.
• Borrow useful terms or subject headings - especially if their strategy has already been peer reviewed or published.

This can give you confidence that your search is on the right track and that you're using terminology that works well in your subject area.

When you first start to write your protocol search strategy, choose one relevant database for your field and build a search. Test it, revise it, and adapt it for use in other databases, checking that the logic still works. It’s good practice to ask someone to peer review your search. Make sure you're saving your results as you go - using a tool like EndNote or Zotero can help manage your references, remove duplicates, and get ready for screening what you’ve found.

Once the protocol search has been finalised and tested in the primary database, the next step is to adapt it for use across other relevant databases. While the core logic and keyword structure of the search remain unchanged to ensure consistency, modifications may be required to accommodate differences in syntax, field codes, and controlled vocabulary between platforms.

Only the technical expressions - such as truncation symbols (*, $, #), proximity operators (adjn, NEAR/n, W/n, Nn), and field tags (.tw, Title, TS, XB) - should be adjusted. The search terms themselves, conceptual structure, including Boolean operators and term groupings, should remain consistent to preserve the integrity of the strategy.

Each adapted version should be:

• Tested to ensure it executes correctly in the target database.
• Saved as a record of the search.
• Documented, including the database name, date run (crucial when the search is rerun/ updated), and any specific notes or issues encountered.

All searches should be run in close succession. It is recommended that you schedule dedicated time to carry out all searches in close succession, one after the other. This improves consistency, reduces error, and supports transparent documentation of the process. All results in the final line should be exported where possible, and results downloaded in a structured format to support subsequent screening and analysis. Remember and give the output file a sensible name, e.g. “20250616_MEDLINE_1-1349” (search run on the 16^th of June 2025 in MEDLINE and the file contains 1349 results).

Once you're done searching, review how well your strategy worked. Were your results useful? Is anything missing? Finally, save a copy of your final strategy - as an appendix or a standalone document - so that others can see how your search was done and build on it in future work.

Maintaining a log of searches can be done using simple methods such as:

1. Personal Accounts: Many database interfaces offer personal account features that allow you to save searches and search histories. These personal accounts often provide options to save searches, set up alerts for new publications matching your search criteria, save references you’ve found, and organise saved searches into folders for easy retrieval.
2. Word (.doc or .docx, similar rich text document): A Word document allows clearer formatting, like headings, tables, and comments, making complex search strategies easier to read and explain. It supports richer context with notes and images. However, Word files are larger and may include hidden formatting that can cause issues when copying search terms, making them less universal than plain text.
3. Spreadsheet (.xls, .xlsx, .csv): An Excel file is useful for organising search strategies in a structured way, such as separating databases, search terms, filters, and notes into columns. It makes sorting, filtering, and updating parts of the strategy easier. However, Excel files are larger and less universal than plain text and copying search strings directly may require extra steps to avoid formatting issues.
4. Notebook or text file (.txt): A plain text document is useful for storing a database search strategy because it is simple and universally compatible, ensuring it can be opened and edited on virtually any device or software without formatting issues. It contains no hidden characters or complex metadata that might interfere with copying and pasting search terms accurately. Additionally, the files are very small in size, making them easy to store, share, and archive. This simplicity also aids transparency and reproducibility, as the search strategy can be clearly reviewed and edited by anyone involved.
5. Email drafts: Use your email account to draft messages to yourself with search details. Include the search strategy, databases searched, and any relevant notes.

Choose the method that suits your preferences and guarantees easy access to your search history when required. It's crucial to accurately document the details. These logs serve not only as an audit of your search activities but also as a record of skill development.

Once you have reached the end of your work, it can be reassuring to see a full record of the work you have conducted and check whether you have missed anything important.