Primary Sources vs Secondary Sources: Differences, Examples, and How to Identify Them
Understand primary, secondary, and tertiary sources across history, science, literature, law, social sciences, and digital research. Learn how to classify, find, evaluate, and cite evidence for an academic assignment.
Need help selecting, evaluating, and organising academic sources?
Explore research paper support →Classify a source by its relationship to your research question. Original records, firsthand testimony, creative works, and newly reported data can be primary. Interpretations and syntheses are secondary. Reference tools that collect information are often tertiary. A source’s format alone does not determine its category.
What Is the Difference Between Primary and Secondary Sources?
A primary source provides original evidence, firsthand testimony, original data, or the work being studied. A secondary source analyses, interprets, evaluates, or synthesises primary material. A tertiary source organises information from both. The correct classification depends on the discipline and the research question.
For example, a diary written during an event is primary evidence; a historian’s book interpreting the diary and other records is secondary. An original scientific article reporting the authors’ experiment is primary research; a systematic review comparing many studies is secondary. A novel is primary in literary analysis, while an article critiquing the novel is secondary.
The Meaning of Primary and Secondary Sources
Primary and secondary sources are two broad categories used to describe the relationship between evidence and the question a researcher is investigating. A primary source provides direct evidence, original data, a firsthand account, or the work being examined. A secondary source interprets, evaluates, compares, or explains primary material. The distinction helps researchers decide what evidence can support a claim, how close a source is to the subject, and what kind of analysis is being offered.
A primary source is not automatically accurate, neutral, or complete. A diary records its writer's perspective; an official report reflects the institution's methods and priorities; an experiment has a design, sample, and limitations. A secondary source is not automatically less valuable. A carefully researched scholarly analysis can compare multiple records, identify contradictions, explain context, and test interpretations. The categories describe a source's function in relation to a research question, not a universal ranking of quality.
The Library of Congress describes primary sources as original documents and objects created at the time under study, while secondary sources retell, analyse, or interpret events. University library guides also emphasise that disciplinary conventions matter: original research is primary in many scientific assignments, while a novel is primary evidence in literary analysis. Always connect the label to the question rather than relying on the format or publication date alone.
Primary Sources vs Secondary Sources: The Main Difference
The central difference is proximity to the evidence. A primary source supplies the original material, observation, testimony, data, creative work, or record. A secondary source makes a claim about that material by interpreting it, synthesising it, criticising it, or placing it in a wider context. For example, a 1963 speech recording is primary evidence of what the speaker said; a historian's article explaining the speech's political significance is secondary analysis.
Consider a scientific study. The article in which researchers report the participants, methods, measurements, results, and limitations of their own experiment is usually a primary research source. A systematic review that gathers findings from many experiments and evaluates the overall evidence is secondary research. A textbook chapter that summarises the field for students is generally tertiary or secondary depending on its depth and purpose.
Neither category means 'good' or 'bad'. Primary evidence can be biased, incomplete, misleading, or difficult to interpret. Secondary analysis can be rigorous, transparent, and methodologically strong. A sound research project often uses both: primary sources establish what the original evidence contains, and secondary sources help explain its meaning, context, debates, and limitations.
Primary, Secondary, and Tertiary Sources
Tertiary sources organise, condense, or point readers towards information drawn from primary and secondary sources. Encyclopedias, dictionaries, general reference works, bibliographies, subject guides, and many introductory textbooks are common examples. They are useful for learning terminology, identifying important people and events, understanding a broad topic, and discovering keywords for further research.
The boundaries are not absolute. A scholarly handbook may provide an original argument in one chapter and a broad synthesis in another. A bibliography is a tertiary tool for locating publications, but it may itself become primary evidence in a study of publishing history or scholarly networks. A textbook is usually a secondary or tertiary source for learning biology, but it can be primary evidence in a study of how biology has been taught.
Use tertiary material to orient yourself, not automatically as the final evidence for every academic claim. If an encyclopedia states that a law changed in a particular year, locate the statute or official record when the legal wording matters. If a reference work summarises a scientific consensus, consult current systematic reviews and original studies where the assignment requires detailed evidence.
Why the Classification Depends on Your Research Question
A source does not carry a permanent primary or secondary label independent of its use. The same item can perform different roles in different investigations. A newspaper editorial from 1968 is primary evidence for a study of public opinion and political rhetoric in 1968. It may function as a secondary source if a researcher uses it to recount an event it did not witness. A school textbook is secondary for a study of the French Revolution but primary for a study of how the Revolution is represented in school curricula.
Start by stating what you are studying. If the subject is an event, ask whether the item records the event directly or analyses it afterwards. If the subject is a creative work, the work itself is primary evidence and criticism of it is secondary. If the subject is the development of a scientific idea, original laboratory papers may be primary historical documents even when they are not current evidence for a clinical decision.
Before classifying a source, write a one-sentence description of its role: 'I am using this interview as direct testimony about the participant's experience' or 'I am using this review article to compare findings from several studies.' That statement makes the classification explicit and helps you justify source selection in your methodology.
Primary and Secondary Sources in History
Historical research uses primary sources to investigate people, institutions, events, beliefs, and conditions in a particular period. Examples include letters, diaries, government records, court proceedings, treaties, census returns, maps, photographs, posters, newspapers, speeches, military orders, business ledgers, oral histories, and material objects. Their value comes from the evidence they preserve, not from an assumption that they tell the whole truth.
Secondary historical sources include scholarly monographs, peer-reviewed journal articles, biographies, historiographical essays, and chapters that interpret historical evidence. Historians compare records, assess provenance, reconstruct chronology, examine power relations, and debate explanations. A historian's account of the causes of the French Revolution is secondary analysis; a contemporary pamphlet or tax register from the period is primary evidence for particular aspects of that society.
Historical primary sources require contextual reading. Identify who created the item, when and where it was made, its intended audience, purpose, genre, and survival history. A royal proclamation communicates an official position; it does not prove that every person obeyed it. A soldier's letter can reveal personal experience but may omit information because of censorship, memory, or the writer's assumptions. Corroborate important claims across independent records.
Primary and Secondary Sources in Science
In the natural and applied sciences, primary sources commonly include original research articles, laboratory notebooks, experimental datasets, field observations, technical reports, registered trial results, patents, and conference papers reporting new findings. An original article usually explains the research question, methods, sample or materials, analysis, results, and limitations. These details allow other researchers to examine how the evidence was produced.
Secondary scientific sources include systematic reviews, meta-analyses, narrative literature reviews, evidence summaries, and scholarly articles that interpret previous research. A systematic review follows a defined process to search for, select, and assess relevant studies. A meta-analysis statistically combines compatible results. These sources can help establish the state of evidence, identify disagreement, and explain how findings vary across populations or methods.
Distinguish original research from a review even when both appear in a peer-reviewed journal. The journal name does not determine source type. Read the abstract and methods: does the paper report a study conducted by its authors, or does it synthesise studies conducted by others? In fast-moving areas such as medicine, public health, and artificial intelligence, check publication dates, corrections, retractions, and whether newer evidence changes the interpretation.
Primary and Secondary Sources in Social Sciences
Social science primary sources include survey responses, interview transcripts, focus-group recordings, ethnographic fieldnotes, observation logs, census data, administrative records, policy documents, organisational records, and datasets collected for a particular study. A research article reporting original survey findings is usually primary research. The underlying dataset and questionnaire may also be primary materials when they are available and appropriate to reuse.
Secondary sources include articles that reinterpret existing datasets, literature reviews, theoretical analyses, comparative studies, and books that synthesise earlier research. A report analysing national census trends is secondary analysis if it uses data collected by a statistical agency. The original census tables and questionnaires are primary data for a study of population measurement or demographic change.
Pay attention to the distinction between data collection and data analysis. A dataset may be primary for the question it was designed to answer, but secondary for a new project that reuses it. Researchers using secondary data should inspect the original variables, sampling frame, definitions, missing values, collection dates, and consent conditions. A dataset's availability does not mean it is suitable for every new question.
Primary and Secondary Sources in Literature and the Arts
In literary studies, the literary work being analysed is normally the primary source. A novel, poem, play, short story, screenplay, song, painting, sculpture, photograph, film, or performance can be primary evidence when it is the object of interpretation. A scholarly article examining symbolism in Toni Morrison's Beloved is a secondary source; the novel itself is primary evidence for that analysis.
Secondary literary sources include book chapters, peer-reviewed criticism, reviews, biographies, theoretical essays, and studies of an author's style or historical reception. They help locate a work within literary movements, genres, historical contexts, and critical debates. A review published when a novel first appeared may also serve as primary evidence in a study of contemporary reception, even though it comments on the novel.
Specify the object of analysis. If you study Shakespeare's language, cite the play and the edition used. If you study how a production adapts Shakespeare, the performance recording, director's notes, programme, and production reviews may each have a different evidentiary role. A source's format does not settle its category; the relationship between the source and your research focus does.
Primary and Secondary Sources in Law
Legal research distinguishes primary legal authority from secondary legal commentary. Primary legal sources include constitutions, statutes, regulations, treaties, court judgments, administrative decisions, legislative debates, and official legal instruments. They establish or record legal rules and decisions within a particular jurisdiction and date. A judgment from a supreme court is primary authority for the court's reasoning and holding.
Secondary legal sources include law review articles, legal textbooks, practitioner commentaries, encyclopedias, case notes, and treatises that explain or critique legal rules. They can clarify doctrine, trace legal development, compare jurisdictions, and identify unsettled questions. They do not replace the wording or authority of the legislation or judgment being interpreted.
Check jurisdiction, court level, date, subsequent history, and whether a case remains good law. A commentary may accurately describe a decision but become outdated after an appeal or statutory amendment. In a legal essay, use secondary scholarship to frame the debate and primary authority to support statements about what the law says. Follow the legal citation system required by the institution or jurisdiction.
Primary and Secondary Sources in Journalism and Media Studies
News reports, broadcast recordings, photographs, interviews, press conferences, social media posts, and official statements can be primary sources when they provide direct evidence of an event, a person's public statement, or media coverage at a particular time. A live broadcast recording is primary evidence of what the broadcaster transmitted. A journalist's eyewitness report can be primary evidence of what the reporter observed.
Later explanatory articles, retrospective documentaries, media criticism, and scholarly studies of news coverage are usually secondary sources. They analyse events, compare accounts, or examine how a story was framed. A newspaper editorial can be primary evidence in research about the newspaper's institutional position, but secondary commentary on the event it discusses.
Media research often studies both the event and its representation. Keep those questions separate. To establish what happened, compare contemporaneous reporting with records, testimony, and other evidence. To study framing, examine headlines, image selection, language, source choices, publication timing, and omissions. A media item can be evidence of public communication without being a complete or neutral record of the underlying event.
Primary and Secondary Sources in Business and Economics
Business and economics primary sources include annual reports, audited financial statements, company filings, earnings-call transcripts, original market data, central-bank releases, national accounts, trade statistics, policy announcements, contracts, and records from an organisation. A company's Form 10-K filed with the U.S. Securities and Exchange Commission is a primary corporate disclosure for research into that company's reported performance and risks.
Secondary sources include analyst reports, academic studies, economic commentary, market reviews, textbooks, and articles interpreting company or macroeconomic data. These sources can compare firms, explain economic mechanisms, evaluate policy effects, and identify limitations in reported figures. A newspaper article summarising an earnings announcement is not the same as the original filing or transcript.
Verify the reporting period, accounting basis, currency, units, definitions, and source organisation. A company's reported revenue, an analyst's forecast, and a journalist's estimate are different kinds of information. When using a secondary analysis, trace important numbers back to the original filing or dataset and check whether the analyst has adjusted, annualised, or reclassified them.
Primary and Secondary Sources in Education Research
Education research primary sources may include classroom observations, student work, lesson plans, curriculum documents, assessment results, teacher interviews, school policies, original survey data, and experimental or quasi-experimental studies. A study in which researchers test a teaching intervention and report their own results is primary research. A recorded lesson can be primary evidence when the question concerns classroom interaction.
Secondary sources include systematic reviews of teaching interventions, meta-analyses of learning outcomes, scholarly critiques of educational policy, and books synthesising research on pedagogy. They help educators understand patterns across settings and identify which conclusions are supported by multiple studies. A government policy review may analyse primary administrative data and function as a secondary source for that analysis.
Education studies are highly dependent on context. Record the age group, subject, school setting, country, intervention, comparison condition, outcome measure, and study period. A finding from one school system may not transfer directly to another. Use primary studies to understand how an intervention was implemented and secondary reviews to evaluate whether similar effects appear across a wider evidence base.
Primary and Secondary Sources in Medicine and Public Health
Medical primary sources include randomised controlled trials, cohort studies, case-control studies, cross-sectional surveys, clinical case reports, laboratory experiments, surveillance datasets, and original diagnostic research. The authors report the study they conducted and the data they collected or analysed. Clinical guidelines and consensus statements are generally secondary or evidence-synthesis sources because they evaluate multiple studies and translate evidence into recommendations.
Secondary medical sources include systematic reviews, meta-analyses, evidence-based guidelines, health technology assessments, and narrative reviews. They can help assess consistency, strength of evidence, clinical relevance, and uncertainty. A public health dashboard may present original surveillance data, while an epidemiological report interpreting trends in those data is an analysis. The distinction depends on who collected the data and what the document does with them.
Do not treat a single primary study as definitive proof. Examine sample size, population, randomisation, blinding, follow-up, outcomes, confounding, and risk of bias. Check whether results have been replicated and whether a systematic review reaches a different conclusion. For clinical decisions, use current professional guidance and qualified medical advice; an academic source classification is not a substitute for clinical judgement.
Examples of Primary Sources
Written primary sources include diaries, letters, emails, speeches, meeting minutes, manuscripts, contracts, legislation, court records, government reports, patents, original research papers, fieldnotes, and unpublished archival documents. Their shared feature is that they preserve original communication, observation, evidence, or a record produced by a person or institution connected to the subject.
Visual and material primary sources include photographs, paintings, posters, maps, architectural plans, clothing, tools, buildings, product packaging, archaeological artefacts, and physical specimens. Audio and audiovisual examples include oral-history recordings, interviews, radio broadcasts, films, music recordings, surveillance footage, and recordings of public events. Digital primary sources may include websites captured at a particular date, social media posts, emails, software versions, and archived web pages.
Data-based primary sources include raw experimental measurements, survey datasets, census tables, financial records, environmental readings, laboratory results, and original statistical releases. The same dataset may be reused by later researchers, but the original data collection remains primary evidence for the study that produced it. Check documentation and metadata before relying on numerical records.
Examples of Secondary Sources
Common secondary sources include scholarly books, peer-reviewed review articles, biographies, historical monographs, literary criticism, book reviews, film reviews, policy analyses, legal commentaries, newspaper retrospectives, and research reports that interpret existing evidence. They explain what primary materials may mean, compare multiple accounts, or place an event or work within a broader framework.
Secondary sources can be academic or popular. A peer-reviewed journal article may offer a detailed interpretation based on primary records. A reputable newspaper explainer may summarise a public policy debate. A museum catalogue essay may interpret an artwork. The credibility of each source depends on its author, evidence, methods, publication process, transparency, and relevance—not merely its category.
Look for the evidence behind the interpretation. Does the author cite original documents, explain how studies were selected, acknowledge contrary findings, and distinguish evidence from opinion? A secondary source that makes a strong claim without showing its basis may be less useful than a narrower analysis with transparent methods. Use the reference list to identify primary documents and further scholarship.
Examples of Tertiary Sources
Tertiary sources collect and organise information from other sources. Examples include encyclopedias, dictionaries, general handbooks, bibliographies, indexes, directories, chronologies, almanacs, subject guides, and many introductory reference websites. They are designed to provide an accessible overview or help readers locate more detailed material.
A tertiary source is useful at the beginning of research because it can clarify unfamiliar terminology, introduce major debates, identify important people and institutions, and suggest search terms. For example, an encyclopedia entry on the Cold War may help identify the Berlin Blockade, NATO, the Warsaw Pact, and key dates before you search archival collections or scholarly histories.
Do not cite a tertiary summary when the assignment expects direct engagement with original evidence or specialist scholarship. Follow its references to the underlying source, check the original publication, and cite the item you actually consulted. Tertiary works can still be primary evidence when the research question concerns the reference work itself, its editorial choices, or the way knowledge was categorised.
Primary vs Secondary Research: Do Not Confuse the Terms
Primary research describes the process of collecting or generating new data for a research project. Secondary research uses data or findings that already exist. Primary and secondary sources describe the relationship between an information item and the research question. The terms are related, but they are not interchangeable.
A student who interviews ten nurses for a dissertation is conducting primary research and producing interview transcripts that serve as primary data. A student who analyses an existing national nursing workforce dataset is conducting secondary data analysis, even though the dataset itself is a primary source created by the statistical agency. A systematic review is secondary research because it synthesises previously published studies.
An original research article is usually a primary source for a researcher studying the scientific finding, but it may be a secondary source for a project investigating how scientific claims are reported in newspapers if the article is being discussed as an object of representation. Explain whether you mean the research method, the data source, or the source's evidentiary role.
Primary Sources Are Not Always More Reliable
Primary sources provide direct access to evidence, but directness does not guarantee truth. A witness may misremember, a diary writer may conceal information, a government record may reflect institutional bias, and a dataset may contain measurement errors. An original experiment can be poorly designed or underpowered. Researchers must evaluate the conditions under which the evidence was created.
Assess authorship, purpose, audience, date, provenance, completeness, and internal consistency. Ask what the creator could know, what they could not know, and what incentives or constraints shaped the record. For data, examine definitions, sampling, measurement instruments, missingness, and data cleaning. For photographs and recordings, consider editing, framing, metadata, and whether the file is authentic.
Corroboration is essential. Compare an eyewitness account with other witnesses, official records, physical evidence, and contemporaneous reporting. Compare a scientific finding with replications and systematic reviews. Secondary sources can improve interpretation by identifying limitations that are not obvious from the primary item alone. The goal is not to choose one category as superior but to assess the evidence for the claim being made.
Secondary Sources Are Not Automatically Less Credible
A secondary source may be more useful than an isolated primary record when the research question requires synthesis, context, or evaluation. A high-quality systematic review can assess dozens of trials. A scholarly historian can compare archives across regions and decades. A legal treatise can explain how several court decisions interact. These contributions require expertise and transparent reasoning.
Evaluate the author's qualifications and perspective, the publisher's standards, the date, the evidence cited, the method of selection, and the treatment of disagreement. A secondary source should distinguish what the evidence demonstrates from what the author infers. It should acknowledge uncertainty and avoid presenting a contested interpretation as settled fact.
Use the secondary source for the work it performs best. It can help you identify scholarly consensus, map competing explanations, and locate original documents. When a specific detail or quotation matters, consult the primary source where possible. If you cannot access it, cite the secondary source you actually read and follow your institution's rules for secondary citation.
How to Identify Whether a Source Is Primary or Secondary
Begin with the research question, then identify the creator and the source's purpose. Was the creator directly involved in the event, did they collect the data, or did they produce the work under study? Does the item present original observations or records? Or does it interpret, compare, summarise, or evaluate evidence created by others?
Read the abstract, introduction, methods, conclusion, captions, and reference list. An original empirical paper normally explains its sample, data collection, and analysis. A review article describes a search or selection process and synthesises earlier studies. A newspaper report may contain direct interviews and contemporaneous observations, while a retrospective feature may rely on later research and commentary.
Check the source against a library guide for your discipline. The University of Florida, Cornell University, the Library of Congress, Purdue University Libraries, and other academic libraries provide examples, but local course instructions take priority. If the classification remains uncertain, describe the source's role in your assignment and ask your lecturer or librarian how the discipline uses the terms.
A Practical Source Classification Checklist
Ask five questions. First, what exactly is my research question? Second, what is the source itself: a record, original work, dataset, report of new research, interpretation, review, or overview? Third, who created it and how close were they to the event, subject, or data? Fourth, what does the source contribute: evidence, analysis, synthesis, or orientation? Fifth, how will I use it in my argument?
Then check the publication details. Record author or creator, title, date, publisher or institution, document type, version, archive or database, and stable identifier. For digital materials, save the URL and access date when required. For archival items, note the collection, box, folder, item number, and repository. Accurate source records make it easier to verify claims and prepare citations.
Write a short annotation for each source: 'Primary—original interview transcript with a participant' or 'Secondary—peer-reviewed synthesis of 18 intervention studies.' This classification should be justified by the research question. Avoid labelling sources solely because they are books, journal articles, websites, or old documents.
How to Find Primary Sources in Library Databases
Search the library catalogue and subject databases using the name of the person, event, organisation, period, or research topic. Add source-type terms such as archive, manuscript, correspondence, diary, oral history, transcript, photograph, government record, dataset, court decision, legislation, fieldnotes, or original study. Use date ranges and subject headings to narrow results.
Discipline-specific databases can reveal original material. Historical newspapers and archival catalogues provide contemporaneous records. Government portals provide legislation, statistics, and official reports. Clinical databases identify original trials. Social science repositories provide survey data and codebooks. Film and literature collections provide original works, editions, and production records.
Read the database record carefully. A catalogue entry may describe an item without providing the full text. Check whether the result is the original document, a digitised copy, a transcript, a finding aid, or a later commentary. Follow repository links and preserve the collection details needed to cite the source.
How to Find Secondary Sources in Scholarly Databases
Use academic databases such as JSTOR, Scopus, Web of Science, PubMed, ERIC, PsycINFO, HeinOnline, and subject-specific library collections where relevant. Search the topic together with terms such as review, systematic review, meta-analysis, historiography, literature review, critical analysis, interpretation, commentary, or scholarly debate. Not every database uses the same indexing terms.
Use filters carefully. A 'review' filter can identify review articles in scientific databases, but it may exclude relevant books or chapters. Peer-reviewed filters can narrow journal results but do not guarantee that every item is a secondary source. Read the abstract and methods to establish whether the authors report original findings or synthesise existing evidence.
Follow citation trails in both directions. The references at the end of a useful article lead to earlier research; 'cited by' tools reveal later studies that discuss it. Citation counts are discovery aids, not quality scores. Check whether the later article agrees, challenges, replicates, or merely mentions the earlier work.
Using Google Scholar and Library Discovery Tools
Google Scholar can help discover scholarly articles, books, theses, conference papers, and citations. Search exact phrases in quotation marks, combine a key concept with a source type, and use date filters when appropriate. The 'Cited by' and 'Related articles' links can reveal connected research. Check whether a result is the publisher's version, a repository copy, a preprint, or a duplicate record.
Library discovery tools connect catalogue records, journal subscriptions, databases, ebooks, and institutional repositories. They may display a citation or abstract without providing full text. Use the library's access links, interlibrary loan, open-access repositories, or contact a librarian when an item is unavailable. Confirm the version and publication status before citing a preprint or accepted manuscript.
Search tools do not decide whether a source is primary or secondary for your project. They retrieve items based on metadata and text. You still need to inspect the creator, document type, date, research method, and relationship to your question. A search result marked 'article' could be an original study, review, editorial, commentary, or news item.
How to Evaluate Primary Sources
Evaluate provenance first: who created the source, when, where, and under what circumstances? Determine whether you have the original, a copy, a transcript, an edited edition, or a digitised reproduction. A transcription may introduce omissions or errors; a scan may not include the reverse side, annotations, or attached materials. Archive descriptions and metadata can clarify these relationships.
Next examine purpose and audience. Was the source created to inform, persuade, record, entertain, regulate, sell, or defend a position? A campaign poster is evidence of political messaging, not necessarily evidence that the message was widely accepted. A corporate press release records the company's public position, not an independent audit of its claims.
Finally test content and context. Compare the source with other evidence, identify contradictions, consider what is absent, and distinguish observation from interpretation. For interviews, consider memory, power relationships, question wording, and recording conditions. For data, examine collection methods and definitions. Explain limitations when the source supports only a narrow conclusion.
How to Evaluate Secondary Sources
Check the author, institutional affiliation, publisher, peer-review process, publication date, and intended audience. Academic credentials matter, but expertise should be relevant to the subject. A reputable publisher and peer-reviewed journal provide quality controls, not guarantees. Look for transparent references and a clear explanation of how evidence was selected and interpreted.
Assess the argument. Is the research question explicit? Does the author engage with competing explanations? Are primary sources or studies represented accurately? Does the article distinguish correlation from causation, evidence from opinion, and established findings from speculation? In a systematic review, inspect search strategy, inclusion criteria, quality assessment, and possible publication bias.
Consider currency and context. A classic theoretical work may remain influential, while a clinical or technological overview may become outdated quickly. Check for corrections, retractions, later editions, and newer reviews. Use secondary sources to understand the debate, but trace critical claims to their original evidence when possible.
Bias, Perspective, and Historical Context
Every source is created from a particular position. Bias does not mean that a source is useless; it means the creator's perspective must be understood. A political speech may reveal a leader's public priorities, while a private letter may show personal concerns. Neither alone represents the views of an entire population. A government report may use standardised categories that exclude experiences outside its definitions.
Context includes the creator's social position, institutional role, audience, available knowledge, language, technology, legal environment, and historical moment. A photograph's composition may reflect the photographer's choices. A colonial archive may preserve the administration's perspective more fully than the voices of colonised communities. Researchers should identify whose experiences are documented and whose are missing.
Use multiple sources to test an interpretation. Compare official records with oral histories, community archives, material evidence, and contemporary journalism where appropriate. Do not assume that several sources are independent if they all repeat the same original report. Trace claims back to their earliest available source and explain the limits of the surviving record.
Triangulation: Combining Different Kinds of Evidence
Triangulation means examining a question through more than one source, method, dataset, or perspective. A historian might compare letters, census records, newspapers, and oral histories. A public health researcher might compare survey responses, hospital data, and interviews. A media scholar might analyse broadcast footage, transcripts, and audience research.
Different sources can confirm, qualify, or contradict one another. Agreement increases confidence only when the sources are sufficiently independent and measure the same thing. A newspaper that repeats a government press release is not independent confirmation of the press release's claim. A survey and an administrative dataset may use different definitions, so apparent disagreement may reflect measurement rather than error.
Document the relationship between sources. Explain which evidence establishes the event, which provides context, and which offers interpretation. When sources conflict, do not hide the discrepancy. Investigate timing, authorship, definitions, incentives, methods, and missing information. A transparent account of uncertainty is stronger than an unsupported claim of certainty.
How to Use Primary Sources in an Academic Essay
Introduce the source and explain why it is relevant before quoting or describing it. Identify the creator, date, document type, and context where those details matter. A quotation should be followed by analysis: explain what the wording reveals, what it does not establish, and how it supports or complicates your argument. Do not assume the reader will draw the intended conclusion without explanation.
Use primary sources as evidence, not decoration. A historical essay should interpret the language, purpose, and circumstances of a letter rather than insert it as an isolated quotation. A science paper should explain how a study's design and results relate to the claim. A literary essay should analyse the text's form, imagery, structure, or narrative choices.
Preserve accuracy. Quote exactly, use ellipses and brackets according to the required style, and provide page, paragraph, timestamp, archive, or dataset locators where possible. Paraphrase faithfully and cite the source. If the original is difficult to access and you rely on a secondary author, make that limitation clear and follow the required secondary-citation format.
How to Use Secondary Sources in an Academic Essay
Use secondary sources to establish scholarly context, define a debate, compare interpretations, and position your argument in relation to existing research. A strong literature review does more than list what authors have said. It groups scholarship by theme, method, period, theory, or disagreement and explains how the contributions relate to one another.
Represent each author's argument accurately. Distinguish the author's central claim from examples, qualifications, and limitations. If two historians disagree about the causes of an event, explain the evidence and assumptions behind each interpretation. If studies report different outcomes, compare their samples, measures, methods, and settings before concluding that the research is contradictory.
Do not allow secondary sources to replace your own reasoning. After presenting a scholar's interpretation, explain whether it is persuasive, limited, or relevant to your question. Use primary evidence where the assignment expects original analysis. Cite all borrowed ideas, even when you have paraphrased them, and verify important quotations in the original publication.
Primary vs Secondary Sources in a Literature Review
A literature review is itself a secondary research product because it analyses and synthesises existing scholarship. The sources included in it can be primary or secondary depending on the review's purpose. A review of clinical trials uses original trials as primary studies. A review of theories may treat foundational theoretical texts as primary sources for the history of ideas.
Define the scope before searching. Specify the topic, population, period, discipline, types of evidence, and inclusion criteria. Search for original empirical studies when you need findings, and review articles when you need a map of the field. Use citation chaining to locate foundational works and recent publications. Record why each source was included or excluded.
Organise the review around relationships, not a sequence of summaries. Compare findings, identify methodological differences, explain gaps, and show how the evidence informs your research question. A literature review should make clear which claims come from original studies and which are interpretations or syntheses by other scholars.
Primary vs Secondary Sources for a Research Proposal
A research proposal needs both source types for different reasons. Primary studies, datasets, documents, or artefacts demonstrate what evidence exists and may reveal a gap. Secondary sources establish the state of knowledge, explain theories, compare findings, and show why the proposed question matters. The balance depends on the discipline and project design.
Begin with credible reviews and scholarly overviews to understand the field, then identify key original studies and primary materials. If the project involves archival research, identify repositories and collections. If it involves interviews or surveys, explain what new data will be collected and how participants will be recruited. If it uses existing data, identify the dataset and its limitations.
Make the evidence chain explicit. State which sources inform the background, which support the conceptual framework, which provide the data, and which methods literature justifies the design. Do not claim that a study is original research if it only summarises previous work. Cite the sources that actually support the rationale and method.
Primary vs Secondary Sources for a Dissertation or Thesis
Dissertations and theses can occupy different positions in the source hierarchy. A thesis reporting original interviews, experiments, archival discoveries, or data analysis may be a primary research source for later scholars. A thesis that synthesises existing scholarship may function as a secondary source. Its classification depends on its methods and use, not the document label.
Use a dissertation's abstract and methodology chapter to identify its contribution. Check the institution, degree, year, supervisor or department where relevant, repository record, and whether the work has been published in a revised form. A thesis may provide valuable detail but may not have undergone the same peer-review process as a journal article.
Your own dissertation may use primary evidence, secondary scholarship, or both. Explain your source selection and justify why each type is appropriate. In qualitative research, describe interviews, observations, and documents. In quantitative research, identify whether the data are newly collected or secondary. In a systematic review, specify databases, search terms, inclusion criteria, and synthesis method.
Primary Sources in Digital Research
Digital primary sources include archived websites, online forums, social media posts, email messages, digital photographs, video recordings, software repositories, discussion boards, electronic records, and platform data. They can document communication, public response, interface design, community practices, and changes in online culture. Their digital form does not make them secondary.
Digital evidence can be unstable. Posts may be edited or deleted, websites may change, links may expire, and platform interfaces may hide context. Preserve the date, author or account identifier where ethically appropriate, URL, access conditions, and an archived copy when permitted. For datasets, retain documentation, codebooks, version numbers, and collection dates.
Consider authenticity, consent, privacy, and platform terms. A public post is not automatically ethically unrestricted for research use. Avoid exposing private individuals unnecessarily, especially in sensitive contexts. Follow institutional ethics requirements and relevant data protection rules. Distinguish a platform's own record from a later article describing that record.
Primary and Secondary Sources in Oral History
Oral history interviews can be primary sources because they preserve a participant's testimony, memories, interpretations, and lived experience. They are especially valuable when written records are limited or when the research question concerns how people remember and understand events. The interview is created through interaction between interviewer and participant, so the questions and circumstances shape the record.
Oral histories are not transparent recordings of the past. Memory changes, narratives are reconstructed, and participants may interpret earlier events through later experiences. The recording can provide evidence of recollection and meaning even when a detail requires corroboration. A published article analysing several oral histories is a secondary source for that analysis.
Document the interview date, interviewer, participant consent, recording format, transcript, repository, and access restrictions. Preserve pauses, context, and transcription conventions where relevant. Compare testimony with other evidence without dismissing it simply because it is subjective. Explain what the interview can establish and where memory or sampling limitations apply.
Primary Sources in Government and Public Policy
Government primary sources include legislation, regulations, policy papers, parliamentary debates, official statistics, budgets, public consultations, agency datasets, court decisions, administrative records, and official correspondence. They provide direct evidence of what an institution enacted, recorded, funded, or publicly stated. A statute is primary legal text; an academic article evaluating its effects is secondary analysis.
Policy research often combines primary records with secondary evaluations. A ministry's policy document describes the government's stated objectives. A budget shows allocated resources. Administrative data may show service use. An independent evaluation interprets implementation and outcomes. These materials answer different questions and should not be treated as interchangeable.
Check the issuing body, legal authority, date, version, geographical scope, definitions, and data collection method. A policy announcement is not proof of implementation. A budget allocation is not proof that funds were spent. An official statistic is not self-explanatory: inspect the denominator, categories, revisions, and methodology. Use secondary evaluation to assess effects and primary records to verify the underlying policy.
Primary Sources in Art, Architecture, and Material Culture
An artwork, building, design object, photograph, sculpture, garment, tool, or archaeological artefact can be a primary source when it is the object being studied. Researchers may examine materials, construction, style, inscriptions, provenance, use, location, and physical condition. The object itself provides evidence that cannot be fully replaced by a written description.
Secondary sources include exhibition catalogues, scholarly articles, conservation reports, biographies, and critical essays that interpret the object. A museum label may offer an introductory interpretation; a conservation record may document technical examination. Both can help, but the researcher should distinguish the object, its documentation, and later interpretation.
Provenance is central. Establish who made or owned the object, where it was located, how it changed hands, and whether restoration or alteration has occurred. A photograph of an artwork is not identical to the artwork, although it may be primary evidence for a study of its reproduction or circulation. Record collection, accession number, medium, date, and location where available.
Primary Sources in Film, Music, and Performance Studies
Films, recordings, scripts, scores, concert recordings, theatre performances, production photographs, programmes, and interviews with creators can all be primary sources when they are the works or records being analysed. A film is primary evidence for a study of cinematography or narrative; a review of that film is secondary criticism. A recording of a concert is primary evidence of that performance, not necessarily of every performance of the composition.
Secondary sources include film criticism, musicology articles, theatre reviews, biographies, production histories, and scholarly analyses of performance. A review published at the time may be primary evidence in a study of reception, while it is secondary commentary in a study of the film's plot. Again, the research question determines the category.
Identify the exact version, performance date, edition, recording, director, composer, performer, production company, and platform where relevant. A remastered version or director's cut may differ from the original release. Use timestamps, track numbers, scene descriptions, or bar numbers to locate the material you analyse. Cite the work itself as well as critical commentary.
Primary Sources in Anthropology and Sociology
Anthropological and sociological primary sources can include ethnographic fieldnotes, participant observation, interview recordings, survey data, community documents, photographs, artefacts, census data, and records produced by social institutions. These sources capture social practices, identities, relationships, institutions, and lived experience in particular settings.
Secondary sources include theoretical interpretations, comparative ethnographies, systematic reviews, sociological analyses of existing datasets, and scholarly discussions of earlier fieldwork. They help connect local findings to concepts such as social class, gender, migration, institutions, power, and cultural change. A published ethnography may be primary evidence for a study of ethnographic writing and secondary analysis of the community it describes.
Ethical and positional questions matter. Consider informed consent, confidentiality, researcher influence, community representation, and the consequences of publication. Fieldnotes are shaped by what the researcher noticed and recorded. Survey results depend on sampling and question design. Explain these limits rather than presenting a source as a complete picture of a community.
Primary Sources in Technology and Computer Science
Computer science primary sources include original conference papers, technical reports, software source code, benchmark datasets, experiment logs, standards documents, patents, system documentation, and original user studies. In fields where major research is published at conferences, a peer-reviewed conference paper may be the first formal report of a new method or result.
Secondary sources include systematic mapping studies, literature reviews, textbooks, technical surveys, comparative evaluations, and articles that analyse or reproduce existing methods. A paper that introduces a new algorithm and reports experiments is primary research; a survey comparing algorithms across the literature is secondary. Documentation may be primary evidence for how a software system is intended to work.
Check version and reproducibility. Software changes over time, benchmarks may be revised, and preprints may differ from final proceedings papers. Record repository commit, release tag, dataset version, hardware, software dependencies, and experiment settings when relevant. Do not assume a benchmark result generalises beyond the tested tasks and conditions.
Common Mistakes When Classifying Sources
A frequent mistake is assuming that every book is secondary and every journal article is primary. A novel can be a primary source in literary analysis; a journal article can be a review. Another mistake is treating age as the deciding factor. A memoir published decades later can be primary testimony, while a recent textbook can be secondary or tertiary.
Students also confuse peer review with source category. Peer review concerns editorial evaluation; primary and secondary describe the source's relationship to evidence. A peer-reviewed systematic review is secondary research. A government document may be primary evidence but still require critical evaluation. A blog post may be primary evidence for a study of online discourse but unsuitable as factual authority for an unrelated claim.
Another error is classifying a source without stating the research question. A newspaper report, documentary, textbook, dataset, or biography can shift categories depending on what is being investigated. Avoid absolute lists that ignore disciplinary context. Explain how the source functions in the specific assignment and cite it accurately.
How Many Primary and Secondary Sources Do You Need?
There is no universal required ratio. The appropriate number depends on the assignment, discipline, research question, available evidence, and instructions from the lecturer. A history essay may require several primary documents and scholarly interpretations. A literature essay may centre on one or two primary texts and use critical scholarship to frame the analysis. A scientific literature review may prioritise original studies and systematic reviews.
Follow the assignment rubric first. If it specifies a minimum number of primary sources, check what the instructor means by primary in that discipline. Ask whether original research articles count, whether archival items are required, and whether sources must be peer reviewed. A source count alone does not demonstrate quality or balance.
Choose sources for relevance, credibility, coverage, and analytical value. A smaller set of well-chosen sources can support a coherent argument better than a long bibliography of loosely related items. Include evidence that represents important competing perspectives and explain why some available material was excluded.
How to Cite Primary and Secondary Sources
Citation style does not usually change simply because a source is primary or secondary. You cite the item according to its source type and the required system: a letter, dataset, court case, journal article, book, interview, archival document, or website each has its own details. APA, MLA, Chicago, Harvard, and discipline-specific legal styles organise those details differently.
Cite the source you actually consulted. If you read an original report, cite that report. If you encountered a quotation only in a secondary article and could not consult the original, do not imply that you read the original. Use the secondary citation procedure required by your style guide, and include the source you actually accessed in the reference list or bibliography as directed.
Provide precise locators when available: page numbers, paragraph numbers, section headings, timestamps, dataset identifiers, archive collection names, box and folder numbers, or legal paragraph references. Accurate locators allow readers to verify the evidence. Follow the current edition and institutional instructions for punctuation, author names, dates, URLs, and access information.
What to Do When a Source Could Be Both Primary and Secondary
Some materials combine original evidence and interpretation. A documentary may contain archival footage, interviews, narration, and editorial framing. A government report may publish original statistics and interpret their implications. A newspaper feature may include eyewitness testimony and analysis. Instead of forcing the entire item into one label, identify the component you are using and its role.
Ask whether you are analysing the source as an object or relying on it as a report about something else. A documentary is primary evidence for a study of documentary filmmaking, but it may be a secondary account of an earlier historical event. A biography is secondary for a study of its subject's life, but primary evidence for a study of biographical writing or the author's public reputation.
State the distinction explicitly in your methods or discussion. You might write: 'The documentary is treated as a primary source for its representation of the event, while its narration is evaluated as a secondary interpretation of the archival footage.' This precision avoids category disputes and makes the analytical purpose clear.
Building a Balanced and Credible Research Base
A credible research base connects direct evidence with scholarship that explains its significance. Begin with a clear question, identify the primary materials that can answer it, and locate secondary research that establishes context and debate. Use tertiary sources to orient yourself and discover terminology, but verify important claims against more direct evidence.
Assess whether the sources are current enough, representative enough, and methodologically appropriate. A study of present-day technology needs current technical evidence. A study of a historical event may need contemporary records and later scholarship. A study of public opinion may need survey instruments and sampling details. A legal analysis needs authoritative materials from the correct jurisdiction.
Keep a source log with full citation details, source type, key claim, evidence, limitations, and intended use. This makes it easier to distinguish original findings from interpretations, avoid accidental plagiarism, and construct a coherent argument. Revisit the classification if your research question changes.
Frequently Asked Questions About Primary and Secondary Sources
Is a journal article a primary or secondary source? It depends on the article. A report of original research is usually primary; a review, meta-analysis, or commentary is secondary. Is a book a primary source? A novel or memoir may be primary for certain questions; a textbook or scholarly analysis is usually secondary or tertiary. Is a website a primary source? It can be, if it contains an original record, statement, dataset, or work being studied.
Are interviews primary sources? An interview recording or transcript is generally primary testimony for research into the participant's experience or statements. Is a documentary a primary source? It may be primary for a study of filmmaking or representation and secondary for its interpretation of an earlier event. Are newspapers primary sources? Contemporary eyewitness reporting and editorials can be primary evidence for media or public opinion research; later analysis is generally secondary.
Are primary sources more reliable? Not automatically. Are secondary sources acceptable in academic writing? Yes; they are essential for scholarly context and interpretation. Should you use both? Often, but the balance depends on the assignment. When uncertain, check your lecturer's instructions and your university library's subject guide.
Final Research Checklist: Choosing the Right Source
Before using a source, confirm that it is relevant to the research question and identify its creator, date, purpose, audience, and format. Decide whether it provides original evidence, original research, interpretation, synthesis, or general orientation. Explain the classification in relation to the task rather than treating it as a fixed label.
Evaluate credibility and limitations. Check provenance, methods, references, publication process, currency, bias, and context. Compare important claims with independent evidence. For digital materials, record version and access details. For archival materials, preserve repository and collection information. For research data, inspect the codebook and collection method.
Use primary sources to engage directly with evidence and secondary sources to understand analysis, context, and debate. Use tertiary sources to begin exploring a topic and locate more authoritative material. Cite what you actually consulted, provide precise locators, and follow the citation style required by your institution.
Quick Comparison Table: Primary vs Secondary vs Tertiary
Primary sources are original records, works, observations, or data. Secondary sources interpret or synthesise primary evidence. Tertiary sources organise and summarise information from primary and secondary material. These are functional descriptions, not quality rankings. A source may move between categories when the research question changes.
Example: for a study of a protest, a participant's video, police record, and contemporaneous photograph may be primary sources; a historian's or sociologist's analysis may be secondary; an encyclopedia overview may be tertiary. For a study of how protests are represented in documentaries, a documentary becomes a primary object of analysis.
Example: for a study of a clinical intervention, an original randomised trial is primary, a systematic review is secondary, and a general health encyclopedia is tertiary. For a study of the history of medical textbooks, a textbook edition itself may become a primary artefact. Classification follows use.
Source Types at a Glance by Discipline
History: primary sources include letters, diaries, official records, photographs, newspapers, and oral histories; secondary sources include historical monographs and historiographical articles. Literature: the poem, novel, or play is primary; literary criticism and scholarly interpretation are secondary. Art history: the artwork and its material record are primary; interpretive scholarship is secondary.
Science: original experiments, observations, and datasets are primary; systematic reviews and meta-analyses are secondary. Social sciences: original interviews, surveys, and fieldnotes are primary; analyses of existing studies or data are secondary. Law: statutes and judgments are primary authority; textbooks and law review commentary are secondary.
Media studies: original broadcasts, posts, and contemporaneous coverage can be primary; retrospective analysis and media criticism are secondary. Business: filings, audited accounts, and original market data are primary; analyst interpretation is secondary. These examples are starting points; consult discipline-specific guidance.
The Role of Original Research Articles
An original research article normally reports a study conducted by its authors. It usually identifies a research problem, describes methods, presents results, discusses implications, and cites earlier scholarship. In science and social science, this is commonly treated as a primary source because it is the first formal publication of new data or findings.
Not every article in a research journal is an original study. Journals also publish reviews, editorials, letters, commentaries, protocols, corrections, and perspectives. A protocol describes a planned study and may be primary evidence for research design, but it does not report outcome findings. An editorial expresses a position and should be classified according to the question.
Look for labels such as Original Article, Research Article, Review, Systematic Review, Meta-analysis, Editorial, or Commentary. Then verify the content. The abstract may say 'we conducted' or 'we analysed' for original work, while reviews describe searching and synthesising published studies. Journal labels are helpful but not conclusive.
The Difference Between a Review Article and a Literature Review
A review article is a published secondary source that examines existing research. It may be narrative, systematic, scoping, integrative, or meta-analytic. A literature review is also the process and written section in which a researcher synthesises relevant scholarship. A literature review inside an original research article does not make the entire article a review article.
Systematic reviews use explicit methods to identify, select, and assess studies. Scoping reviews map the breadth and characteristics of evidence. Meta-analyses statistically combine results when studies are sufficiently comparable. Narrative reviews may provide expert synthesis without the same formal search and selection requirements. Identify the review type before judging what it can establish.
Use review articles to understand the field and find primary studies. When a review makes a claim about a specific experiment, locate the original study if that detail is central. A review's conclusion is based on its search scope, inclusion criteria, and appraisal process; it should not be treated as an exhaustive account unless its method supports that claim.
Archival Sources, Finding Aids, and Digitised Collections
An archive holds records created or received by individuals, families, organisations, and public bodies. Primary archival materials may include correspondence, minutes, photographs, maps, plans, ledgers, recordings, and unpublished reports. A finding aid is a descriptive guide to a collection; it helps researchers locate records but is not usually a substitute for the records themselves.
Digitisation makes archival material easier to access but does not remove the need to assess provenance. A scanned letter is a digital surrogate of a physical item. The scan may omit edges, annotations, envelopes, or related documents. A transcription may normalise spelling or punctuation. Record the repository, collection title, series, box, folder, item, and digital identifier where supplied.
Respect access restrictions and handling conditions. Some collections contain private, sensitive, or copyrighted material. Follow repository rules for reproduction and quotation. When citing a digital collection, provide a stable link or catalogue identifier and enough archival detail for another researcher to locate the original.
Datasets as Primary and Secondary Evidence
A dataset is often primary evidence when it contains original observations or measurements collected for a study. It can include survey responses, sensor readings, experimental measurements, census records, financial transactions, or administrative data. A dataset's documentation explains how variables were defined, collected, cleaned, and stored.
When a researcher reuses an existing dataset to answer a new question, the project is secondary data analysis. The dataset remains an original record from its source organisation, but it is secondary data relative to the new research process. This distinction matters for research methods, consent, data quality, and interpretation.
Check who collected the data, the target population, sampling method, response rate, measurement units, missing values, weighting, version, and access conditions. A large dataset is not automatically representative. A public dataset may exclude groups or variables that matter to your question. Cite the dataset itself and any documentation used.
Primary Sources and Ethical Research Practice
Access to a primary source does not automatically mean it is ethically appropriate to reproduce or analyse it. Interviews, private messages, health records, social media posts, and identifiable photographs may contain personal information. Research involving human participants must follow institutional ethics review, consent procedures, and relevant privacy requirements.
Historical and archival research also raises ethical questions. A public record may expose vulnerable people, reproduce discriminatory language, or contain information shared under conditions that differ from modern publication. Researchers should consider harm, representation, cultural protocols, and whether a quotation or image is necessary for the argument.
Use only the information needed for the research purpose, protect identities where appropriate, and explain limitations. Follow repository restrictions, copyright rules, platform terms, and institutional guidance. Ethical handling strengthens the credibility of the research and respects the people represented in the evidence.
Frequently Asked Questions About Primary and Secondary Sources
What is a primary source?
A primary source is an original record, firsthand account, original dataset, work, or report of research that provides direct evidence relevant to a question. Examples include letters, speeches, photographs, laws, interviews, original research articles, and raw data.
What is a secondary source?
A secondary source analyses, interprets, evaluates, or synthesises primary evidence. Examples include scholarly books, literature reviews, biographies, historical analyses, legal commentary, and research articles reviewing earlier studies.
What is the difference between primary and secondary sources?
Primary sources provide original evidence or material; secondary sources explain or analyse that material. The distinction depends on how the source relates to the research question and varies across disciplines.
Is a journal article a primary or secondary source?
An original research article reporting the authors' own study is generally primary. A review article, meta-analysis, editorial, or commentary is generally secondary. Read the article's methods and purpose to identify its type.
Is a book a primary or secondary source?
A book can be either. A novel, memoir, or historical document may be primary for a relevant study. A textbook, biography, or scholarly interpretation is usually secondary or tertiary.
Are newspapers primary or secondary sources?
A contemporaneous eyewitness report, editorial, advertisement, or interview can be primary evidence for research into the event or media coverage. A later retrospective or analysis is usually secondary.
Are interviews primary sources?
An interview recording or transcript is usually a primary source when it provides direct testimony or original research data. An article analysing several interviews is secondary.
What are tertiary sources?
Tertiary sources compile or organise information from primary and secondary materials. Examples include encyclopedias, dictionaries, bibliographies, indexes, and many introductory reference works.
Are primary sources more reliable than secondary sources?
Not automatically. Primary sources can be biased or incomplete, while secondary sources can provide rigorous synthesis. Evaluate authorship, purpose, methods, evidence, context, and limitations.
Can one source be both primary and secondary?
Yes. A documentary can be primary evidence for a study of filmmaking and a secondary account of an earlier event. Its role depends on the research question and the aspect being analysed.
What is the difference between primary research and primary sources?
Primary research is the process of collecting new data. A primary source is an item that provides original evidence. A researcher can conduct secondary data analysis using an existing primary dataset.
How many primary sources should an essay include?
There is no universal number. Follow the assignment rubric and disciplinary expectations. Choose sources for relevance, credibility, and analytical value rather than meeting an arbitrary count.
