01 · The Question
When do local government reports deserve a systematic search?
Local governments produce a great deal of potentially useful information: programme evaluations, surveillance reports, needs assessments, policy documents, statistical reports, technical studies, planning documents, implementation reviews, and administrative data summaries. Much of it never becomes a journal article.
If your research concerns a particular locality, ignoring these sources can leave an important part of the evidence base invisible. Yet the opposite strategy, attempting to search every page of every government website, can consume enormous amounts of time while producing little relevant evidence.
The question is therefore not simply whether government reports count as evidence. It is whether your research question makes them sufficiently important that they should be searched deliberately and systematically.
03 · What You Need to Know
Government reports can fill evidence gaps that journal databases cannot
Government reports are part of the grey literature
Grey literature generally refers to material produced outside traditional commercial or academic publishing channels. Government reports are a major part of this landscape, alongside theses, dissertations, conference materials, technical reports, working papers, and other research outputs.
Current Cochrane guidance recommends considering relevant grey literature sources, including reports, when comprehensive searching is required. One reason is publication bias: limiting a search to conventional journal publications can systematically exclude research or results that were never published through those channels.
Local governments may hold evidence that universities do not
The importance of government sources depends on the question. Local authorities often collect information because they administer services, regulate activities, allocate resources, monitor programmes, or are legally required to report particular indicators.
For a study of local education, for example, relevant government material might include enrolment statistics, programme implementation reports, school performance summaries, infrastructure assessments, or local policy documents. Public-health research might encounter surveillance reports, epidemiological profiles, service-utilization data, or programme evaluations. Environmental research may draw on monitoring reports, hazard assessments, land-use plans, or technical studies.
These documents may answer questions that journal articles were never designed to answer, particularly questions about conditions within a highly specific locality.
Decide what kind of government document actually counts as evidence
“Government report” is too broad to function as an eligibility criterion. A statistical bulletin, commissioned evaluation, annual accomplishment report, policy memorandum, planning document, consultation paper, and mayoral press release do not provide the same kind of evidence.
Before searching, identify what document types could legitimately answer your research question. If you need estimates of local prevalence, an official statistical report may be relevant. If you are examining programme effectiveness, you may need evaluation reports containing methods and outcome data. If your question concerns policy implementation, administrative and implementation documents may be directly informative even if they do not evaluate outcomes.
Evidence about a local condition or outcome
Documents containing empirical data, analyses, evaluations, monitoring results, or other information capable of supporting a factual claim about the setting.
Evidence that a government adopted a policy or position
Official policies, regulations, plans, memoranda, and similar documents that establish what an authority formally decided or stated, but do not necessarily show whether the policy worked.
Map the responsible agencies before searching
Searching becomes much more manageable when you begin with the administrative structure surrounding the topic.
Ask which government bodies would plausibly collect the information or commission the research. Depending on the setting, these might include municipal or city governments, provincial or state authorities, education offices, health departments, planning agencies, statistical offices, environmental authorities, social-welfare agencies, transport departments, or specialist regulatory bodies.
Then identify their official websites, document repositories, publication pages, open-data portals, research divisions, and relevant archives. This creates a defensible source list rather than an open-ended web search.
Use several retrieval routes because government websites are inconsistent
Government websites vary greatly in their search functionality. Some maintain well-organized publication databases. Others provide only a site search, document archive, or collection of PDF links. Older reports may sit on archived pages or under the department that originally commissioned them.
A practical search may therefore combine the agency's own search function, browsing of relevant publication sections, targeted web searching, exact-title searching for documents discovered in references, and citation or reference chasing from known reports.
Use simple queries when the website has a basic search engine. Long Boolean strategies developed for bibliographic databases often perform poorly on ordinary websites.
Search by programme and policy names as well as research concepts
Academic researchers and governments may describe the same subject differently. A journal article might discuss “digital learning adoption,” while the government documentation is organized under the official name of a technology programme. A study of food insecurity may need searches using the names of specific assistance programmes or surveys.
Build search terms from both the scholarly literature and the administrative language of the setting. Useful terms can include programme names, legislation, agency terminology, service names, geographic units, abbreviations, and older names when policies or agencies have changed.
Systematic searching does not require perfect reproducibility
Government websites present a practical difficulty: their content and search interfaces change. Results may not have stable ordering, archives may be reorganized, and documents can move or disappear.
This makes exact replication difficult, but it does not eliminate the need for transparent methods. Record the agency, website or portal, URL, date searched, search terms, relevant filters, sections browsed, and any limits you imposed on screening.
Recent research assessing Campbell systematic reviews found that grey-literature searches were often less completely reported than bibliographic database searches, including omissions involving URLs and other search details. That matters because readers need enough information to understand where and how non-database evidence was sought.
Watch Out
Do not write only “government websites were searched.” That tells the reader almost nothing about the search. Name the relevant agencies or portals and explain how they were searched or browsed.
Official status does not guarantee methodological quality
A document being published by a government establishes its provenance. It does not automatically establish the quality of its methods or the validity of every conclusion it contains.
Ask what evidence supports the report's claims. Where did the data come from? How were observations selected? Are definitions clear? Is the analysis described? Is there enough information to judge limitations? Was the report an independent evaluation, an internal administrative assessment, or a communication document?
The appropriate appraisal depends on the type of evidence. A census table, qualitative consultation report, programme evaluation, and policy document require different questions.
Be alert to institutional interests without dismissing the evidence
Government agencies may evaluate programmes that they designed, funded, or implemented. This creates a potential institutional interest that should be considered when interpreting methods and conclusions.
That does not make the evidence unusable. Similar considerations arise with industry-funded research, commissioned evaluations, advocacy reports, and even academic studies whose investigators have intellectual commitments to an intervention. The appropriate response is appraisal and transparency, not automatic exclusion.
A government report and a journal article may describe the same study
Comprehensive searching can retrieve multiple reports from one underlying study or evaluation. Cochrane guidance emphasizes that the study, rather than each individual report, is the primary unit of interest in a systematic review.
If a government evaluation later becomes a journal article, do not count the two documents automatically as independent evidence. Link reports that appear to describe the same underlying study and use all relevant reports to obtain the most complete information available.
07 · A Quick Checklist
Before searching local government reports
Before and during the search, check:
Decide what kinds of government-produced evidence could actually answer your research question.
Identify the government agencies responsible for the relevant policy area, service, programme, population, or data.
Locate their official publication pages, repositories, research units, archives, and open-data portals where relevant.
Search both academic terminology and official programme, policy, agency, and administrative terminology.
Record the agency, URL, search date, terms, filters, sections browsed, and any screening limits.
Distinguish empirical reports from policy statements, communications, and purely descriptive administrative documents.
Appraise relevant reports according to their methods and purpose rather than assuming quality from official status.
Check whether government reports and journal articles describe the same underlying study before counting them separately.
