How to Do Research

Last updated:




How to Do ResearchHow to do research for an academic paper is a question that extends well beyond typing keywords into a search engine. Research for a college-level research paper involves systematic engagement with scholarly sources — peer-reviewed journal articles, government reports, archival documents, original datasets, and other materials that have been vetted by experts in the field. The process requires identifying what types of sources the project demands, locating those sources through academic databases and library systems, evaluating each source for credibility and relevance, and organizing the material so that it feeds directly into the drafting process. This guide covers the principles and strategies that apply across disciplines, while iResearchNet’s discipline-specific research guides — linked throughout and listed in the Research by Discipline section below — provide field-level guidance on databases, journals, and methodological conventions particular to each area of study.

Understanding the Research Process

Research is not a discrete stage that begins after topic selection and ends before writing. In practice, research, reading, and writing overlap from the earliest phases of a project through final revision. Students who treat research as a single block of time — gather all sources first, then write — often discover gaps in their evidence only after a draft is underway, forcing them to return to the library under time pressure. A more effective approach treats research as iterative: read a cluster of sources, draft a section, identify what the argument still needs, search for additional material to fill that gap, and revise the draft accordingly.

The scope and depth of research depend on the assignment’s parameters. A five-page paper for an introductory course may require five to eight sources, most of them secondary — books and articles by scholars who have already analyzed the topic. A twenty-page seminar paper may require twenty or more sources and may demand engagement with primary materials — original documents, datasets, or artifacts that the student analyzes firsthand. A senior thesis or honors project may require months of sustained research across multiple source types. Regardless of scale, the underlying logic is the same: every factual claim and analytical interpretation in the paper must be grounded in evidence that the reader can locate and verify.

Three activities constitute the core of the research process, and each requires a different set of skills. Source identification — finding relevant materials — depends on knowing which databases, catalogs, and archives serve your discipline and how to construct effective searches within them. Source evaluation — determining whether a particular article, report, or dataset is credible, relevant, and current — depends on understanding how scholarly publishing works and what distinguishes peer-reviewed research from opinion, advocacy, or journalism. Source integration — incorporating evidence into the paper’s argument without merely summarizing or stringing quotations together — depends on analytical reading and note-taking practices that separate the source’s claims from the student’s own interpretation. The sections below address each of these activities in detail.

Planning the research phase before beginning it saves considerable time. Students who sit down at a database and begin searching without a plan tend to follow tangents, collect sources that turn out to be irrelevant, and lose track of what they have already reviewed. A brief research plan — listing the databases to be searched, the keywords to be used, the types of sources needed (empirical studies, theoretical articles, policy reports, primary documents), and the approximate number of sources required — imposes structure on what can otherwise become an unfocused and inefficient process. This plan should be revisited and adjusted as the research progresses, but having a starting framework prevents the aimless searching that consumes hours without producing usable material.




↑ Back to top

Types of Research

Academic research falls into several categories, and understanding these categories helps students match their approach to their assignment’s requirements. The most fundamental distinction is between primary and secondary research.

Primary Research

Primary research involves working directly with original sources or generating new data. In the humanities, primary sources include literary texts, historical documents, letters, diaries, photographs, films, legal records, and other artifacts produced during the period or by the person under study. In the social sciences, primary research may involve conducting surveys, interviews, experiments, ethnographic observations, or statistical analyses of raw datasets. In the natural sciences, it typically involves laboratory experiments, field observations, or computational modeling. Upper-division and graduate courses in the United States frequently require some form of primary research; introductory courses usually do not, though they may ask students to analyze a primary source that the instructor provides.

The demands of primary research vary by discipline. A history student working with eighteenth-century court records faces different challenges than a psychology student designing an experimental survey or a biology student collecting water samples. Each requires discipline-specific training in methodology, ethics (particularly when human subjects are involved — see the Office for Human Research Protections for federal guidelines), and data management. Students undertaking primary research for the first time should consult their instructor and, where applicable, their institution’s Institutional Review Board (IRB) before beginning data collection.

For extended guidance, see iResearchNet’s guide to conducting original research.

Secondary Research

Secondary research involves analyzing, synthesizing, and building on work that other scholars have already published. Most undergraduate research papers in the United States are built primarily on secondary sources — peer-reviewed journal articles, scholarly books and edited volumes, government reports, and systematic reviews. Secondary research requires the student to locate relevant scholarship, understand how different authors approach the same question, identify areas of agreement and disagreement in the literature, and position the student’s own argument in relation to this existing conversation.

The quality of a secondary-research paper depends heavily on the quality of the sources selected. A paper built on five carefully chosen articles from leading journals in the field will produce a stronger argument than one built on twenty sources drawn from general-interest websites, opinion columns, and encyclopedia entries. Learning to distinguish between peer-reviewed scholarship and other types of publication is among the most important research skills a college student can develop.

Quantitative and Qualitative Approaches

Within both primary and secondary research, approaches divide further into quantitative and qualitative methods. Quantitative research collects and analyzes numerical data — survey responses measured on scales, experimental results, census figures, economic indicators — and uses statistical techniques to identify patterns and test hypotheses. Qualitative research collects and analyzes non-numerical data — interview transcripts, field notes, textual passages, visual media — and uses interpretive techniques to identify themes, meanings, and relationships. Many research papers in the social sciences and health sciences employ mixed methods, combining quantitative and qualitative evidence to address a research question from multiple angles.

The choice between quantitative and qualitative approaches is not a matter of preference; it is determined by the research question. A question about the prevalence of food insecurity among U.S. college students calls for quantitative data — survey results, institutional reports, demographic statistics. A question about how food-insecure students experience and cope with their situation calls for qualitative data — in-depth interviews, narrative accounts, observational studies. A comprehensive study of the topic might use both. Students should discuss their methodological approach with their instructor early in the research process, particularly if the assignment does not specify a method.

↑ Back to top

Finding and Accessing Sources

Academic Databases

Academic databases are the primary tools for locating peer-reviewed scholarship, and familiarity with the databases relevant to your discipline is essential for efficient research. Most U.S. university libraries provide access to hundreds of databases through their websites, and subject librarians can help students identify which databases serve their field. The major general-purpose databases include JSTOR, which archives complete runs of leading journals across the humanities, social sciences, and sciences; Google Scholar, which indexes scholarly literature broadly and provides citation tracking; and ProQuest, which aggregates dissertations, theses, newspapers, and journal content.

Discipline-specific databases offer deeper and more targeted coverage within particular fields. PubMed, maintained by the National Library of Medicine, indexes biomedical and life sciences literature. PsycINFO, maintained by the American Psychological Association, covers psychology and behavioral sciences. ERIC, sponsored by the U.S. Department of Education, indexes education research. SSRN (Social Science Research Network) hosts working papers and preprints in economics, law, and the social sciences. HeinOnline provides full-text legal journals, law reviews, and government documents. MLA International Bibliography indexes scholarship in literature, linguistics, and folklore. Students working in any of these fields should learn to search their discipline’s primary database as part of the research planning process.

Constructing effective database searches requires attention to keyword selection, Boolean operators, and filtering tools. A search for “immigration” in a social science database will return tens of thousands of results; a search for “immigration AND wage effects AND low-skilled workers AND United States” will return a manageable set of directly relevant articles. Most databases also allow filtering by date range, publication type (peer-reviewed only), methodology, and language. Learning these search refinement techniques early in the project saves substantial time and produces more relevant results.

Library Resources

University libraries offer resources and services that no database search can fully replicate. Physical collections include books, bound journals, government document depositories, microfilm archives, special collections of rare and primary materials, and reserve materials placed by instructors for specific courses. Libraries also provide interlibrary loan services, which allow students to request books and articles held by other institutions — a critical resource when a needed source is not available locally.

Reference librarians and subject-specialist librarians are among the most underutilized resources on any campus. A thirty-minute research consultation with a subject librarian can identify databases the student did not know existed, suggest search strategies that improve result quality, and locate primary source collections relevant to the project. Most academic libraries in the United States offer these consultations at no cost, either in person, by appointment, or through online chat services. Students who have already identified their topic and done some preliminary searching will benefit most from these consultations, because the librarian can offer targeted advice rather than general orientation.

For extended guidance, see iResearchNet’s guide to library research.

Internet Sources

The open internet — material accessible through general search engines without a library subscription — is useful for certain research tasks but unreliable for others. Government websites maintained by agencies such as the Bureau of Labor Statistics, the Census Bureau, the National Center for Education Statistics, and the Centers for Disease Control and Prevention provide authoritative statistical data and policy documents. Professional association websites — the American Psychological Association, the American Sociological Association, the American Historical Association — publish position statements, ethical guidelines, and links to disciplinary resources. News outlets with established editorial standards can provide current-event context for a research question, though they are not substitutes for peer-reviewed analysis.

Sources that lack clear authorship, institutional affiliation, editorial oversight, or citation practices — personal blogs, anonymous wiki articles, content farms, and social media posts — should generally be excluded from academic research papers unless the assignment specifically calls for analyzing such materials as primary data. The credibility threshold for inclusion in an academic paper is whether the source has been vetted by someone with relevant expertise, either through peer review, editorial review, or institutional accountability.

For extended guidance, see iResearchNet’s guide to online research.

↑ Back to top

Evaluating Sources

Not every source that appears in a database search or on a library shelf is equally useful for a research paper. Source evaluation is the process of determining whether a particular source is credible, relevant, current, and appropriate for the argument being constructed. Applying a consistent evaluation framework to every source prevents the common problem of building an argument on evidence that turns out to be unreliable, outdated, or tangential.

Credibility depends on several factors: the author’s qualifications and institutional affiliation, the publication venue (peer-reviewed journals carry more weight than trade magazines or self-published material), the rigor of the methodology, and the transparency of the evidence presented. A source authored by a faculty member at a research university and published in a journal with a rigorous peer-review process is, by default, more credible than an unsigned article on a commercial website — though no source is immune from error, and critical reading remains necessary even for peer-reviewed material.

Relevance requires assessing whether the source actually addresses the research question or merely touches on the topic in passing. A student writing about the effects of standardized testing on educational equity in U.S. public schools does not benefit from a source that mentions standardized testing only in a single paragraph within a broader discussion of educational philosophy. The most useful sources are those that address the student’s specific question directly, provide evidence the student can analyze, or articulate a theoretical framework the student can apply.

Currency matters in fields where knowledge evolves rapidly. In the biomedical sciences, a study published fifteen years ago may have been superseded by subsequent research using improved methods or larger datasets. In the humanities, older scholarship may remain foundational — a literary critic writing in 1960 may still be essential reading for a paper on a Victorian novel. The student’s discipline and the nature of the research question determine how much weight to give publication date, but in general, recent sources should constitute the majority of the bibliography in the social and natural sciences.

Bias and perspective should be assessed rather than treated as automatically disqualifying. Every scholar writes from a particular theoretical, methodological, and political standpoint, and recognizing these standpoints is part of sophisticated reading. A source advocating for a particular policy position may provide useful evidence and arguments even if the student ultimately disagrees with its conclusions. The question is not whether a source is “objective” — no scholarship is entirely free of perspective — but whether its claims are supported by evidence that can be independently evaluated.

↑ Back to top

Organizing Your Research

Effective organization during the research phase prevents two costly problems later: losing track of sources that will be needed for the bibliography, and losing track of the student’s own ideas about what those sources mean. Both problems are avoidable with consistent practices established at the outset of the project.

Citation Management

Citation management tools — Zotero, Mendeley, and EndNote are the most widely used in U.S. higher education — allow students to save bibliographic information from databases with a single click, organize sources into folders and tag them by theme, annotate entries with personal notes, and generate formatted bibliographies in APA, MLA, Chicago, or other citation styles automatically. Learning a citation manager at the beginning of the research process eliminates the tedious and error-prone task of manually formatting dozens of references at the end. Most U.S. university libraries offer free workshops on Zotero and other tools; students who take advantage of these workshops early in the semester gain a productivity advantage that compounds over every subsequent assignment.

Note-Taking Strategies

Structured note-taking during the reading phase is the bridge between research and writing. Effective research notes distinguish clearly among three types of content: direct quotations (copied verbatim with page numbers), paraphrases (the source’s idea restated in the student’s own words), and the student’s own analytical observations (reactions, questions, connections to other sources or to the thesis). Failing to distinguish among these three types is a primary cause of unintentional plagiarism — students who do not clearly mark which words are their own and which belong to a source may inadvertently present borrowed language as original prose.

Several note-taking systems enforce this distinction. The Cornell method divides the page into a narrow left column for keywords and questions, a wide right column for detailed notes, and a bottom section for summary. Annotated bibliography entries combine a standard citation with a paragraph assessing the source’s argument, methodology, relevance, and relationship to other sources. Digital tools like Zotero’s note feature or dedicated apps like Notion allow students to tag and cross-reference notes across sources. The specific system matters less than the consistency with which it is applied — any method that reliably distinguishes quotation from paraphrase from original thought will serve the purpose.

Building an Annotated Bibliography

Even when not formally required by the assignment, building an annotated bibliography during the research phase produces compounding benefits. For each source, the student writes the full citation followed by a paragraph — typically 100 to 150 words — that summarizes the source’s central argument, evaluates its methodology and credibility, and notes its relevance to the student’s own research question. This exercise forces active engagement with each source at the point of reading rather than passive consumption. When drafting begins, the annotated bibliography serves as a ready-made inventory of evidence: the student can scan the annotations to identify which sources support which claims, which sources disagree with one another, and where gaps in the evidence remain. Many instructors at U.S. universities assign annotated bibliographies as preliminary steps toward a research paper precisely because the exercise builds the analytical habits that produce stronger final papers. For students working on longer projects — senior theses, capstone papers, or graduate seminar papers — the annotated bibliography also functions as a record of intellectual progress, documenting how the student’s understanding of the topic evolved as the research developed.

For extended guidance, see iResearchNet’s guide to taking notes during research.

↑ Back to top

Research by Discipline

Research methods, conventions, and source types vary significantly across academic fields. The guides below provide discipline-specific guidance on key databases, major journals, professional associations, source types, and methodological norms for research papers in each field. Each guide is tailored to the needs of U.S. college and university students conducting research in that discipline.

Humanities and Arts

Social Sciences and Policy

Business and Economics

Health and Natural Sciences

Interdisciplinary and Applied Fields

↑ Back to top

Common Research Mistakes

Several research errors appear with striking regularity in student papers, and most are preventable with awareness and planning. Relying too heavily on a single source — even an excellent one — produces a paper that reads as a summary of someone else’s argument rather than an original contribution. A research paper should synthesize multiple sources, placing them in conversation with one another rather than deferring to any single authority.

Beginning the research process too late compresses the time available for source gathering, evaluation, and integration. Students who start searching for sources two days before the paper is due can only use whatever they find immediately — which is rarely the best available evidence. Starting research early also leaves time for interlibrary loan requests, which typically take several days to fill but can provide access to sources not held by the student’s own institution.

Searching too narrowly or too broadly wastes time in different ways. A search that is too narrow returns few or no results and may lead the student to conclude — incorrectly — that insufficient scholarship exists on the topic. A search that is too broad returns thousands of results that cannot be reviewed within the available time. Iterative refinement of search terms, combined with use of database filtering tools, resolves both problems. If an initial search for “climate change education” returns too many results, adding terms like “K-12” or “curriculum reform” or “United States” will reduce the set to a more manageable size while preserving relevance.

Failing to track sources systematically during the research phase creates a frustrating scramble at the bibliography stage. Students who read an article, take notes, but do not record the full citation information may find themselves unable to relocate the source when the reference list is due. Using a citation manager from the first day of research eliminates this problem entirely. Students who prefer not to use software should at minimum maintain a running document with complete bibliographic information — author, title, journal, volume, pages, DOI — for every source consulted.

Over-relying on abstracts rather than reading full articles is a subtler error. Abstracts provide a compressed summary of a study’s purpose, method, and findings, but they omit the nuance, limitations, and contextual information that the full article contains. A student who cites an article based solely on its abstract risks misrepresenting the source’s claims, overlooking qualifications that the authors themselves acknowledge, or missing methodological limitations that affect how the findings should be interpreted. Reading the full text is the only way to use a source responsibly.

↑ Back to top

Frequently Asked Questions

How many sources do I need?

The number of sources depends on the assignment’s length, complexity, and disciplinary conventions. A general guideline for U.S. undergraduate courses is one to two peer-reviewed sources per page of the final paper — a ten-page paper typically requires ten to twenty sources. However, some instructors specify a minimum number, and some assignments (particularly literature reviews or historiographical essays) require substantially more. Quality consistently matters more than quantity: five carefully selected, thoroughly analyzed sources contribute more to an argument than twenty superficially cited ones.

Should I use books or journal articles?

Both have roles, and the balance depends on the discipline and topic. Journal articles report specific studies or arguments and are typically more current than books. Books provide broader context, deeper treatment of complex topics, and synthetic overviews of a field’s development. In the social and natural sciences, journal articles usually form the backbone of a research paper’s bibliography. In the humanities — particularly history and literary studies — monographs (scholarly books) often carry more weight than individual articles. Most strong research papers include a mix of both.

Can I use Wikipedia as a source?

Wikipedia should not appear in a research paper’s bibliography, because its articles are collaboratively edited, lack consistent authorship credentials, and can be modified by anyone at any time. However, Wikipedia can be useful as a starting point for identifying key terms, major scholars, and preliminary bibliography. The references listed at the bottom of well-developed Wikipedia articles often point toward credible scholarly sources that the student can then locate through academic databases and evaluate independently.

What do I do if I cannot find enough sources?

Insufficient results usually indicate a search strategy problem rather than a genuine absence of scholarship. Broadening search terms, trying synonyms, searching in a different database, removing filters, or consulting a reference librarian can often surface sources that an initial search missed. If the topic is truly too new or too specialized for sufficient peer-reviewed literature to exist, the student should discuss the situation with the instructor — the solution may involve adjusting the topic, incorporating primary sources or gray literature (government reports, working papers, conference proceedings), or acknowledging the limited state of the literature as part of the paper’s framework.

How do I know if a source is peer-reviewed?

Most academic databases include a filter that limits results to peer-reviewed or “refereed” publications. The database Ulrichsweb provides a directory of periodicals that identifies which journals use peer review. When in doubt, checking the journal’s website — specifically its “About” or “For Authors” page — will typically state whether the journal uses a peer-review process. Peer-reviewed sources have been evaluated by independent experts in the field before publication, which makes them the gold standard for academic research, though they are not the only type of source that belongs in a research paper.

↑ Back to top

↑ Back to top