eduKate Singapore

VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

Next useful routeCanonical hubHELP

Why English? | Writing a Literature Review Search Strategy

—

by

eduKate Asia
Three learners review open books together at a classroom table, with stacks of textbooks, stationery and a whiteboard in the bright room.
WHY ENGLISH? · PRACTICAL READING AND WRITING

Make a literature search explainable, repeatable and proportionate to the review question

A search strategy is not a list of keywords remembered after the review is written. It is a documented set of choices connecting the question to concepts, synonyms, subject headings, databases, limits, screening and updates. English matters because relationships among words determine which evidence becomes visible.

1. Define the reviewSet question, review type, eligibility and evidence needs.2. Build concept blocksDevelop synonyms, variants and subject headings.3. Write the logicUse operators, fields, proximity and truncation carefully.4. Choose and test sourcesSelect databases, grey literature and citation routes.5. Document the processRecord dates, translations, exports, deduplication and updates.

Choose the route closest to your task, or read straight through for the complete system.

Searches for literature review search strategy, Boolean search, systematic review keywords and database search often focus on one clever string. The more important task is building a transparent route from question to evidence. The strategy should show what concepts were searched, which variants were included, where the search ran, when it ran and what limits were applied.

Recall and precision create a practical tension. A broad search retrieves more potentially relevant records and more noise. A narrow search is easier to screen but may miss relevant language, spelling, disciplines or older terminology. The solution is not a universal balance; it is a reasoned choice for the review purpose, supported by pilot searches and known relevant records.

English is especially important because databases do not understand intention the way a human reader does. Quotation marks, parentheses, AND, OR, NOT, truncation, proximity and field codes change the logic. A small punctuation error can exclude thousands of records or combine concepts incorrectly. Search writing is therefore a form of executable reasoning.

This article is educational. A course literature review, scoping review, systematic review and rapid review have different expectations. Consult the assignment, librarian, supervisor, protocol and database help pages for the required method.

Did You Know? NUS Libraries separates systematic-review work into question development, searching and result management. Its search guidance emphasises concepts, synonyms, controlled vocabulary and database-specific translation. One master string pasted everywhere may look efficient, but different databases use different fields, operators and subject headings; accurate searching requires documented adaptation.

Find the section you need

Define the review · 2 chapters
  1. Name the review purpose and question
  2. Break the question into searchable concepts
Build concept blocks · 2 chapters
  1. Expand synonyms and language variants
  2. Use subject headings with free text
Write the logic · 2 chapters
  1. Write Boolean logic visibly
  2. Use phrases, truncation and proximity carefully
Choose and test sources · 2 chapters
  1. Choose fields and databases for the evidence domain
  2. Plan supplementary searching
Document the process · 3 chapters
  1. Test against known records and noise
  2. Apply limits only with a defensible reason
  3. Document, translate, export and update
CHAPTER 1 OF 11 · Define the review

1. Name the review purpose and question

Back to contents

Search design depends on whether the review maps, estimates, compares, explains or synthesises experience.

Write the review question and preliminary inclusion criteria before composing keyword blocks.

Evidence checkpoint

A scoping review may intentionally include diverse evidence while an intervention review may require a more specific population and comparison.

What to check

  • Name review type
  • Write question
  • Draft eligibility
  • Identify stakeholders

Create a four-column evidence record for chapter 1, “Name the review purpose and question”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Do not call a review systematic merely because it uses several databases.

Learning transfer: Purpose language guides proportionate method.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 2 OF 11 · Define the review

2. Break the question into searchable concepts

Back to contents

Not every word in a question belongs in the query.

Identify two to four central concepts and separate them from context that may be better handled at screening.

Evidence checkpoint

Adding every eligibility detail to the search can remove papers that use unexpected wording.

What to check

  • Circle core concepts
  • Remove outcome if harmful
  • Separate screening rules
  • Keep rationale

Create a four-column evidence record for chapter 2, “Break the question into searchable concepts”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: The right number of concepts depends on the topic and database.

Learning transfer: Concept extraction is useful in all information searches.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 3 OF 11 · Build concept blocks

3. Expand synonyms and language variants

Back to contents

Authors can name the same idea differently across fields, regions and time.

Build a concept table containing everyday terms, technical terms, acronyms, spelling variants, older labels and closely related phrases.

Evidence checkpoint

“University student” may appear as undergraduate, tertiary student, college student or higher-education learner.

What to check

  • Add acronyms
  • Check spelling variants
  • Look for historical terms
  • Record exclusions

Create a four-column evidence record for chapter 3, “Expand synonyms and language variants”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Synonyms are evidence-based candidates, not a thesaurus dump.

Learning transfer: Vocabulary breadth improves research without keyword stuffing.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 4 OF 11 · Build concept blocks

4. Use subject headings with free text

Back to contents

Controlled vocabulary groups indexed concepts; free text finds words in titles and abstracts.

Look up the database’s own subject headings, explode or focus deliberately, and pair them with tested free-text terms.

Evidence checkpoint

A new technology may lack a mature subject heading, so free-text variants remain essential.

What to check

  • Use correct thesaurus
  • Check scope note
  • Combine heading and text
  • Record mapped term

Create a four-column evidence record for chapter 4, “Use subject headings with free text”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Indexing practices and coverage vary by database and date.

Learning transfer: Readers learn that classification systems shape discovery.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 5 OF 11 · Write the logic

5. Write Boolean logic visibly

Back to contents

OR usually joins alternatives within a concept; AND combines different concepts; parentheses preserve the intended order.

Format one line per concept block and test the logic in small parts before combining it.

Evidence checkpoint

(adolescent OR teenager) AND (sleep OR insomnia) is different from an ungrouped string whose platform precedence is unclear.

What to check

  • Group OR terms
  • Join blocks with AND
  • Inspect parentheses
  • Avoid reflexive NOT

Create a four-column evidence record for chapter 5, “Write Boolean logic visibly”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: NOT can remove relevant records containing the excluded word in another context.

Learning transfer: Explicit logic supports coding and argument structure.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 6 OF 11 · Write the logic

6. Use phrases, truncation and proximity carefully

Back to contents

Search operators can improve retrieval but also create false matches or missed variants.

Check each database’s syntax, test sample records and avoid stems so short that unrelated words flood results.

Evidence checkpoint

child* may retrieve child and children; a different stem can retrieve unintended technical words.

What to check

  • Read help page
  • Test truncation
  • Check phrase handling
  • Use proximity purposefully

Create a four-column evidence record for chapter 6, “Use phrases, truncation and proximity carefully”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Syntax does not transfer automatically between platforms.

Learning transfer: Punctuation becomes operational meaning.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 7 OF 11 · Choose and test sources

7. Choose fields and databases for the evidence domain

Back to contents

A title-only search is precise but can miss relevant work; an all-fields search can be noisy.

Select databases by disciplinary coverage, document types, dates and indexing, then choose title, abstract, keyword or subject fields deliberately.

Evidence checkpoint

A multidisciplinary education question may need education, psychology and general citation databases rather than one familiar platform.

What to check

  • Justify each database
  • Check coverage dates
  • Name fields
  • Avoid convenience-only choice

Create a four-column evidence record for chapter 7, “Choose fields and databases for the evidence domain”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: No database is complete, and access subscriptions affect interfaces.

Learning transfer: Source selection is part of evidence quality.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 8 OF 11 · Choose and test sources

8. Plan supplementary searching

Back to contents

Database queries may miss grey literature, reports, theses, trial records and records connected by citation.

Specify official websites, repositories, backward references, forward citations, author searches or registries relevant to the question.

Evidence checkpoint

An education policy review may need ministry reports that journals cite but databases index inconsistently.

What to check

  • Name grey sources
  • Define citation chasing
  • Set stopping rule
  • Record website path

Create a four-column evidence record for chapter 8, “Plan supplementary searching”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Web searching is less reproducible; document query, date and screening boundary.

Learning transfer: Information routes matter beyond a single search box.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 9 OF 11 · Document the process

9. Test against known records and noise

Back to contents

Pilot searches reveal whether important terminology is missing and whether irrelevant clusters dominate.

Use several known relevant papers as tests, inspect why they are found or missed, and sample excluded-looking result sets.

Evidence checkpoint

If a key study is absent because its abstract uses a different population label, add the justified variant.

What to check

  • Choose sentinel records
  • Inspect false positives
  • Revise one block
  • Save versions

Create a four-column evidence record for chapter 9, “Test against known records and noise”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Known records help diagnose but should not be used to tune until only expected evidence remains.

Learning transfer: Testing makes retrieval a transparent iterative process.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 10 OF 11 · Document the process

10. Apply limits only with a defensible reason

Back to contents

Date, language, document type, human or age filters can save work but can also introduce bias.

State each limit, source of the rule, database implementation and likely consequence.

Evidence checkpoint

An English-language limit may be practical for a course but should be disclosed as a boundary, not treated as evidence that other research does not exist.

What to check

  • Justify date range
  • Explain language choice
  • Check filter accuracy
  • Avoid hidden defaults

Create a four-column evidence record for chapter 10, “Apply limits only with a defensible reason”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Interface filters can behave differently from screening criteria.

Learning transfer: Constraint reporting supports honest conclusions.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

CHAPTER 11 OF 11 · Document the process

11. Document, translate, export and update

Back to contents

A usable strategy allows another reader to understand what was done at a particular time.

Save exact strings, platform, database, coverage, date, result count, export format, deduplication process, changes and update plan.

Evidence checkpoint

A strategy written for one database must be translated, not copied, and the translation should be preserved.

What to check

  • Save exact query
  • Record run date
  • Track results
  • Plan rerun

Create a four-column evidence record for chapter 11, “Document, translate, export and update”. Copy the exact source wording or observation, explain what it means in context, name the decision it affects, and state the boundary that prevents a larger conclusion.

Now explain the same point to a reader who has not seen the source. Preserve the governing noun, verb, population, comparison, time frame and uncertainty. If the shorter explanation changes the research decision, restore the missing detail.

Boundary: Searches age as databases grow and terminology changes; date the evidence.

Learning transfer: Audit trails strengthen research and workplace information tasks.

Try it on the document

Write one sentence that states the evidence, one that states its limit, and one that states the next responsible action. Then test each noun, number, condition and source against the controlled document. Replace vague confidence with a traceable reason.

PARENT AND STUDENT GUIDE

How to build this kind of English before the job title arrives

Back to contents

Students do not need professional authority to practise the underlying language. They can learn to define scope, annotate conditions, compare versions, rebuild a timeline, match a number to its unit and explain uncertainty without embarrassment. Parents can ask calm questions: What decision is this document supporting? Which sentence carries the strongest claim? What would make that claim false? Which source would you open next?

Use a three-pass routine. First, map the document: title, owner, audience, date, sections and decision. Second, trace one claim from source to conclusion. Third, explain the limit and next action. This routine strengthens comprehension, science, mathematics, humanities and workplace readiness because it turns English into an evidence tool.

A student should never imitate professional authority. The useful goal is to recognise when a text needs a qualified reviewer, current rule or official decision. Knowing when to pause is part of literacy.

WORKED SEARCH

From question to concept blocks

Back to contents

Question

How do first-year university students describe verifying AI-generated sources in research-writing assignments?

Concept blocks

Population: first-year student, freshman, undergraduate. Phenomenon: AI-generated, generative AI, large language model. Practice: source evaluation, verification, information literacy.

Pilot decision

Research-writing assignment is initially screened rather than required in the query because authors may describe the course context only in full text.

Record

Each database translation, field code, date and result count is saved; new synonyms are added only with a documented reason.

SEARCH RECORD

What a reader should be able to reconstruct

Back to contents

Why

Review purpose and question.

What

Concepts, terms, headings, fields and limits.

Where

Database and platform.

When

Coverage and run date.

Then

Export, deduplication, screening and update route.

PRACTICE CASES

Four situations that expose the hidden English decision

Back to contents

For each case, identify the tempting shortcut, the controlling evidence and the sentence that needs repair.

The sentence search

A full research question retrieves little. The writer builds concept blocks and variants.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

The copied syntax

One database string fails in another interface. Operators, fields and subject headings are translated.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

The tempting NOT

An exclusion term removes relevant mixed-topic papers. Screening replaces the risky shortcut.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

The undocumented update

A final rerun finds new papers, but no date or export is saved. The search log is completed.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

A parent or tutor can ask for the source and reasoning, but the student should make the final record in their own words.

RESEARCH WORKSHOP

Five deeper practices for a durable result

Back to contents

Use a live, current source where possible. Keep source wording separate from interpretation and preserve dates.

Create a concept table

Add free text, acronyms, variants, headings and exclusions.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Test one block at a time

Inspect known records and irrelevant clusters before combining.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Translate across databases

Preserve logic while adapting syntax and fields.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Log every run

Save platform, database, query, date, limits and result count.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Plan the update

Set the final rerun and change-management method.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Finish with three lines: what is supported, what remains uncertain and what the next research action is.

TRANSFER LENSES

How the same English changes across five readers

Back to contents

The evidence should remain stable even when tone, detail and next action change with the reader.

Reviewer lens

The search should serve a defined synthesis.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Librarian lens

Vocabulary and database behaviour need testing.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Reader lens

Choices and limits should be visible.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Future updater lens

Exact records make reruns possible.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Student lens

Proportionate rigour matters more than imitation.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

Adaptation is useful only when the shortened version preserves design, population, comparison, uncertainty and boundary.

DEEPER READING

Three checks that catch polished but incomplete work

Back to contents

These checks slow the reader where a plausible summary can hide a consequential gap.

The retrieval trade-off

Record what broadening and narrowing changed. Precision is not a reason to erase a relevant concept, and recall is not a reason to accept uncontrolled noise.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

The translation audit

For each database, compare concept coverage rather than character-for-character strings. Save the platform-specific final query.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

The evidence-date boundary

A careful search is accurate for its run date. State the final search date so readers know where the evidence window ends.

Write a short evidence note naming the source, applicable date, decision owner and boundary. Then state the smallest next action that remains justified if one assumption proves wrong.

A strong final record shows not only the answer reached, but also the evidence boundary and the condition that would require revision.

TWELVE-QUESTION CHECK

Pause before you submit, rely or repeat

Back to contents

Answer each question in a complete sentence tied to a source, date, observation or authorised clarification. Write “unknown” when the record does not support an answer.

  1. What review type applies? Record the evidence and the conclusion it permits.
  2. What is the question? Record the evidence and the conclusion it permits.
  3. Which concepts enter the search? Record the evidence and the conclusion it permits.
  4. Which variants and headings are used? Record the evidence and the conclusion it permits.
  5. How are Boolean blocks grouped? Record the evidence and the conclusion it permits.
  6. Which field codes apply? Record the evidence and the conclusion it permits.
  7. Which databases are justified? Record the evidence and the conclusion it permits.
  8. What supplementary routes apply? Record the evidence and the conclusion it permits.
  9. Were known records found? Record the evidence and the conclusion it permits.
  10. Why are limits used? Record the evidence and the conclusion it permits.
  11. How are results deduplicated? Record the evidence and the conclusion it permits.
  12. When will the search be updated? Record the evidence and the conclusion it permits.

Reader reconstruction

Give the record to a careful reader who was not present. Ask them to identify the question, method, evidence, uncertainty and next action without extra explanation. If they cannot, repair the missing link rather than adding decorative formality.

Change test

Imagine that the database, sample, outcome definition, analysis or source page changes tomorrow. Mark every sentence that would need updating. This makes cycle-sensitive claims visible and prevents an old note from controlling a new decision.

FREQUENTLY ASKED QUESTIONS

Questions readers often ask

Back to contents

How many databases should I search?

Enough to cover the relevant disciplines and evidence types for the review, subject to the review method and resources. More is not automatically better.

Should I search the entire research question as a sentence?

Usually no. Break it into concepts and use the database’s syntax and vocabulary.

Is Google Scholar enough?

It can support supplementary discovery, but its coverage, ranking and reproducibility differ from curated bibliographic databases. Follow the review requirement.

Can I use NOT to remove irrelevant topics?

Use it cautiously and test whether relevant records also contain the excluded term.

Why record result counts?

Counts help audit translations, detect interface changes and explain the flow from search to screening.

When should I update the search?

For time-sensitive or formal reviews, plan an update close to completion or publication according to the protocol and guidance.

USEFUL NEXT READING

Continue with verified sources and connected eduKate guides

Back to contents

Cycle-sensitive facts were checked against current official pages on 11 October 2026. Always reopen the authority’s live page before a real decision.

Discover more from eduKate Singapore

Subscribe now to keep reading and get access to the full archive.

Continue reading