How Should You Search When Your Research Question Is Highly Local?

What changes when your research question is highly local?

Suppose your study concerns public-school teachers in one province, flooding in a particular municipality, healthcare access in a specific rural district, or the implementation of a national policy within one city. You search a major scholarly database, add the place name, and retrieve only a handful of papers. Does that mean little research exists?

Not necessarily. A highly local question creates an information-retrieval problem as much as a research problem. Relevant evidence may be scattered across international databases, national and regional journals, university repositories, government websites, institutional reports, theses, conference materials, and other sources that do not share the same indexing standards. Some useful studies may concern your setting without naming it in the title or abstract at all.

The challenge, then, is not simply to search more. It is to design a search that reflects how evidence about your setting is actually produced, described, indexed, and stored.

Use a layered search rather than one geographically restricted search

A local question changes both where and how you search

First decide what “local” means for your question

Locality is not a fixed geographic scale. Depending on the question, “local” might mean a country, region, province, state, city, municipality, district, school system, hospital network, Indigenous community, watershed, or even one institution. Evidence about local conditions has similarly been defined in evidence-informed decision making as evidence from the particular setting in which a decision or action will occur.

Define that boundary before constructing the search. Otherwise, the search can drift between several meanings of relevance. A study from the same country may be geographically local in one project but too distant from the relevant context in another.

Separate topical relevance from geographic relevance

A useful way to think about a highly local search is to separate two questions: What evidence is relevant to the phenomenon? and what evidence is relevant to the setting?

These questions overlap, but they are not identical. If you search only for studies that mention your locality, you may miss important evidence about the phenomenon itself. Conversely, a large international literature may explain the phenomenon well while saying relatively little about how it operates under your particular institutional, cultural, environmental, economic, or policy conditions.

This is why a highly local question usually benefits from a broader evidence layer as well as a local one. The broader search helps establish what is already known, while the local search helps determine what is known about the setting and which contextual conditions may matter. The question of how global and local searching should be sequenced deserves separate consideration rather than assuming that one can substitute automatically for the other.

Do not rely on one version of the place name

Geographic terminology is surprisingly unstable. A location may be described by its city, municipality, district, province, region, country, administrative abbreviation, historical name, or a broader geographic label. A study conducted in your locality may instead identify only the participating university, hospital, school division, health authority, watershed, or other institution.

Build a geographic vocabulary before searching. Depending on the topic, it may include:

• the official place name and common spelling variants;
• province, state, region, district, or other parent administrative units;
• older geographic or administrative names when relevant to the search period;
• common abbreviations;
• major local institutions associated with the topic;
• locally used population or community terms; and
• local-language equivalents where relevant.

You do not necessarily need to place every term into one enormous Boolean string. Different sources have different search capabilities. The Campbell Collaboration's information-retrieval guidance emphasizes adapting search strategies to the characteristics of individual databases and other information sources rather than assuming that one search syntax transfers unchanged everywhere.

Search major scholarly databases, but do not treat them as a census of local research

Major bibliographic databases remain useful because they can reveal peer-reviewed studies from your setting, studies from comparable settings, systematic reviews, and terminology that improves later searches. Their weakness for a highly local question is not that they are inherently unsuitable, but that their coverage is selective.

Local journals may not be indexed in the databases you normally use. Institutional research may never become a journal article. Reports commissioned by government agencies or local organizations may be published directly on organizational websites. Theses and dissertations may sit in university repositories. Conference outputs may be available only through society or institutional archives.

Accordingly, a sparse result in Scopus, Web of Science, PubMed, ERIC, or another field-specific database is evidence about what that database retrieved. It is not, by itself, evidence that your locality has barely been studied.

Add sources that are structurally closer to the setting

Once conventional database searching is underway, move closer to the local research ecosystem. Which sources matter will vary by discipline and country, but potentially useful places include national bibliographic databases, regional journal platforms, university repositories, government portals, statistical agencies, professional associations, research institutes, local scholarly societies, and institutional websites.

Grey literature deserves deliberate consideration. Cochrane's guidance on evidence searching identifies reports, dissertations, theses, and conference abstracts among relevant grey-literature sources, while also recommending reference-list searching and other supplementary approaches. Grey literature can be difficult to search systematically because bibliographic control and search functionality vary substantially across sources.

The appropriate response is not to search every conceivable website. Identify the source types most likely to contain evidence relevant to your question. Separate guidance is useful when deciding how to approach local government reports, theses and dissertations, and local conference proceedings.

Use known relevant records to learn the local vocabulary

Your first useful local paper is more than a source to cite. Treat it as a map.

Inspect its title, abstract, keywords, author affiliations, references, funding information, and terminology. Which name does it use for the population? How does it describe the locality? Which institutions appear repeatedly? Which journals publish similar work? Which authors or research groups work on the topic? What reports or local sources does it cite?

Those clues can generate additional searches. Citation chasing can also reveal records that keyword searching missed. Searching backward through reference lists and, where supported by a database, forward through papers that cite a relevant record can be particularly useful when indexing is inconsistent.

Search local repositories with simpler strategies

A sophisticated database search may use controlled vocabulary, proximity operators, field codes, truncation, and nested Boolean logic. A university repository or government website may support little more than a keyword box.

Do not force the same search string into every system. Break the question into short combinations and run several searches. You might search the topic plus the municipality, the topic plus the province, the population plus a local institution, and important synonyms separately. Campbell guidance similarly recommends documenting searches of non-bibliographic sources in enough detail that readers can understand how those sources were explored.

Consider local-language searching where it could change retrieval

If research in the setting may be written or catalogued in more than one language, an English-only search can create an avoidable blind spot. Translate the important concepts carefully rather than relying mechanically on literal translations. Terminology used by local researchers, government agencies, and communities may differ from the vocabulary used in international scholarship.

This matters especially when searching repositories, government documents, local journals, and older records. Test candidate terms against known relevant documents and adjust them when the local terminology differs from what you initially expected.

Document searches that cannot be perfectly reproduced

Local searching often involves websites and repositories with changing interfaces, unstable result ordering, weak filtering, or limited export functions. Perfect reproducibility may therefore be unrealistic. Transparency is still possible.

Record the source or website, URL, date searched, terms used, filters or sections searched, and any relevant limits. If you browsed rather than searched, document what you browsed. If you screened only a defined number of search-engine results, report that decision. Campbell's current information-retrieval guidance specifically recommends detailed documentation for additional sources so readers can assess the reliability and comprehensiveness of the search.

What a layered local search can look like in practice

The result may still be that very little directly relevant research exists in the province. The difference is methodological: that conclusion now rests on a search designed around the local evidence environment rather than on the coverage of one database.

Common mistakes when searching for highly local evidence

Match the depth of your local search to what your research question requires

Not every project needs an exhaustive search of every local repository. Search intensity should follow the role that locality plays in the research question.

One useful principle is to ask what would have to be true for evidence from elsewhere to apply to your setting. Differences in baseline conditions, institutions, resources, implementation capacity, population characteristics, or other effect-modifying factors may matter. Evidence-informed policy literature has long distinguished the value of broad evidence about effects from the need for local evidence about conditions and implementation.

In other words, “local” should not become either a prestige marker or an exclusion rule. It is a dimension of relevance whose importance depends on the question you are asking.

Before concluding that you have searched the local evidence adequately

Questions researchers ask about highly local literature searches

A highly local question needs a search strategy built for local evidence

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.