Science and technology research spans an enormous range of disciplines — biology, chemistry, physics, computer science, engineering, earth science, materials science, and dozens of subfields within each — unified by a shared commitment to the scientific method: formulating testable hypotheses, collecting empirical evidence through observation and experimentation, analyzing that evidence using appropriate quantitative techniques, and subjecting findings to peer review by independent experts. A research paper in science or technology might require published experimental results from peer-reviewed journals, datasets from government and institutional repositories, preprints from discipline-specific servers, patents and technical standards, or computational tools and software libraries. This guide, part of iResearchNet’s How to Do Research series, covers the databases, data repositories, and methodological conventions that U.S. college and university students need for research papers across the natural sciences and technology. For discipline-specific guidance on mathematics, see iResearchNet’s guide to mathematics research; for environmental science, see the guide to environment research; for health and medical sciences, see the guide to health and medicine research.
Academic Databases for Science and Technology
Web of Science
Web of Science (Clarivate) is one of the two major multidisciplinary citation databases for scientific research. Its Science Citation Index Expanded covers more than 9,500 journals across the natural sciences, engineering, and technology. Web of Science’s defining feature is citation indexing — it tracks which papers cite which other papers, allowing researchers to trace the influence of a specific study forward through time (who has cited it since?) and backward (which earlier studies did it build on?). This citation network is one of the most powerful tools for mapping a research area, identifying the most influential studies, and finding recent work that extends or challenges earlier findings.
Scopus
Scopus (Elsevier) is the other major multidisciplinary citation database, indexing more than 27,000 journals, conference proceedings, and book series across science, technology, medicine, social sciences, and the arts and humanities. Scopus provides citation tracking, author profiles, institutional metrics, and journal analytics. Its coverage of conference proceedings and non-English-language publications is somewhat broader than Web of Science’s, making it particularly useful for research in engineering and computer science, where conference papers carry significant weight.
Discipline-Specific Databases
Each major scientific discipline has one or more specialized databases that provide deeper coverage than the multidisciplinary databases.
PubMed covers biomedical and life sciences, including biology, biochemistry, genetics, microbiology, pharmacology, and neuroscience. Chemical Abstracts Service (CAS, maintained by the American Chemical Society) provides the most detailed coverage of chemistry literature and chemical substance information; its database SciFinder is available at most university libraries. Inspec (Institution of Engineering and Technology) covers physics, electrical engineering, electronics, computer science, and information technology. GeoRef covers geology, geophysics, hydrology, and earth sciences. Compendex (Engineering Village, Elsevier) is the primary database for engineering research.
IEEE’s digital library (Institute of Electrical and Electronics Engineers) provides full-text access to IEEE journals, conference proceedings, and technical standards — essential for research in electrical engineering, computer science, telecommunications, and related fields. The ACM Digital Library (Association for Computing Machinery) provides full-text access to ACM publications covering all areas of computing.
Preprint Servers
The sciences have a well-established preprint culture in which researchers post manuscripts to open-access servers before or simultaneously with journal submission. arXiv (Cornell University) covers physics, mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical engineering, systems science, and economics. bioRxiv and medRxiv cover the biological and medical sciences. ChemRxiv covers chemistry. EarthArXiv covers earth and planetary sciences. TechRxiv covers engineering and technology.
Preprints provide the fastest access to current research — often months or years before the peer-reviewed version appears — but they have not undergone peer review, and their findings may change substantially during the review process. Students should note the preprint status when citing such papers and should check whether a peer-reviewed version has been published.
Google Scholar indexes preprints alongside published articles and provides citation tracking across both. Students should configure Library Links to connect Google Scholar to their university’s subscriptions for full-text access.
Data Repositories and Open Data
Modern scientific research increasingly relies on shared datasets, and many funding agencies and journals now require researchers to deposit their data in publicly accessible repositories. Students can access these datasets for secondary analysis, visualization, or to verify the findings of studies they cite.
Zenodo (CERN) is a general-purpose open-access repository that hosts datasets, software, and other research outputs from all scientific disciplines. Dryad specializes in datasets underlying scientific publications, particularly in the biological and environmental sciences. Figshare hosts datasets, figures, and supplementary materials from published research. GitHub hosts code and software used in computational research, including analysis scripts, simulation code, and software tools.
Discipline-specific repositories include the Protein Data Bank (PDB) for molecular structure data, GenBank (NCBI) for nucleotide sequence data, the Cambridge Structural Database (CSD) for crystal structure data, and NASA Earthdata for satellite-derived environmental observations. The Registry of Research Data Repositories (re3data) catalogs more than 3,000 data repositories across all scientific disciplines, helping students identify the relevant repository for their field.
Scientific Writing and the IMRAD Structure
Scientific papers follow a standardized structure known as IMRAD: Introduction, Methods, Results, and Discussion. This format organizes the paper around the logical sequence of the scientific method.
The Introduction establishes the research question, reviews the relevant literature, identifies the gap that the study addresses, and states the hypothesis or objectives. The Methods section describes the experimental design, materials, procedures, and analytical techniques in enough detail that another researcher could replicate the study. The Results section presents the findings — typically through tables, figures, and statistical summaries — without interpretation. The Discussion section interprets the results, compares them to previous findings, addresses limitations, and suggests directions for future research.
Students writing science papers should follow this structure unless their instructor specifies otherwise. Even review papers and literature-synthesis papers (which do not report original experiments) typically use a modified version of this format, with a methods section describing the literature search strategy and an analysis section replacing the results.
Peer Review, Reproducibility, and Scientific Integrity
Understanding how peer review works helps students evaluate the quality of the sources they cite. When a scientist submits a paper to a journal, the editor sends it to two or three independent experts (peer reviewers) who evaluate the paper’s methodology, analysis, and conclusions. Reviewers may recommend acceptance, revision, or rejection. Papers that pass peer review have been vetted by experts, which is why peer-reviewed journal articles are the gold standard for scientific evidence in research papers.
The reproducibility of scientific findings has become a major concern across multiple disciplines. The replication crisis — the finding that a significant proportion of published results cannot be reproduced when the experiments are repeated — has prompted reforms including preregistration of study protocols, open data requirements, and increased emphasis on replication studies. Students should be aware that a single published study, even in a prestigious journal, does not constitute settled science; confidence in a finding increases when it has been replicated by independent groups using different methods. Review articles, systematic reviews, and meta-analyses that synthesize evidence across multiple studies provide stronger foundations for claims in a research paper than any individual study.
Scientific integrity requires that researchers and students report data accurately, cite sources appropriately, and distinguish between their own contributions and those of others. Data fabrication (making up data), data falsification (manipulating data to produce desired results), and plagiarism (presenting others’ work as one’s own) are the most serious violations of scientific ethics and carry severe consequences in both academic and professional contexts.
Citation Conventions in Science
Citation practices vary across scientific disciplines, and students should follow the style required by their course or target journal.
The natural sciences most commonly use numbered citation styles — CSE style (Council of Science Editors) in biology and related fields, ACS style (American Chemical Society) in chemistry, and IEEE style in engineering and computer science. These styles use superscript numbers or bracketed numbers in the text, with corresponding numbered entries in the reference list. APA style (author-date format) is used in some areas of the social and behavioral sciences that intersect with science, including neuroscience and health sciences.
Regardless of the citation style, scientific papers require precise and complete citation of all sources that informed the research. Every claim of fact should be traceable to a specific source. The reference list should include full bibliographic information — authors, title, journal, volume, pages, year, and DOI (Digital Object Identifier) — for every cited work.
Reading and Evaluating Scientific Papers
Reading a scientific paper requires a different approach than reading a newspaper article or a textbook chapter. Scientific papers are written for specialists, and their density of technical detail can be overwhelming for students encountering a subfield for the first time. Several strategies make the process more manageable.
Start with the abstract, which provides a compressed summary of the paper’s purpose, methods, key findings, and conclusions. If the abstract suggests the paper is relevant, read the introduction (which explains the research question and its context) and the discussion/conclusion (which interprets the findings and explains their significance) before tackling the methods and results sections. This “hourglass” reading strategy — abstract, introduction, discussion, then methods and results — gives the student the big picture before the technical details.
When reading the results section, focus on what the data shows rather than on the statistical details. Tables and figures often communicate results more clearly than the text that describes them. For each major finding, ask: What was measured? What was found? How strong was the effect? Was it statistically significant? Pay attention to the confidence intervals and effect sizes, not just the p-values — a statistically significant result with a tiny effect size may not be practically meaningful, while a large effect with a wide confidence interval may reflect imprecise measurement rather than a real phenomenon.
The limitations section (usually part of the discussion) is one of the most important parts of a scientific paper for research-paper purposes, because it identifies what the study did not or could not demonstrate. A study’s acknowledged limitations may include small sample size, lack of a control group, self-report bias, limited generalizability to other populations, or inability to establish causation. Students should report these limitations when citing the study, rather than presenting the findings as more definitive than the authors themselves claim.
Evaluating a scientific paper’s quality involves assessing the study design (Was it well-controlled? Was the sample appropriate?), the methodology (Were the measurements valid? Was the analysis appropriate?), the reporting (Are the results presented clearly and completely?), and the interpretation (Do the conclusions follow from the data, or do the authors overstate their findings?). The CONSORT guidelines (for randomized trials), the STROBE guidelines (for observational studies), and the PRISMA guidelines (for systematic reviews and meta-analyses) provide checklists that students can use to assess reporting quality.
Science, Technology, and Society (STS)
Not all research papers about science and technology are scientific papers in the IMRAD sense. Some examine science and technology as social, political, or historical phenomena — how scientific knowledge is produced, how technology shapes society, how scientific institutions operate, and how science policy is made. This interdisciplinary field, known as Science, Technology, and Society (STS) or Science and Technology Studies, draws on the methods of history, sociology, philosophy, and political science rather than the experimental methods of the natural sciences.
STS research papers may use databases from multiple disciplines. Web of Science’s Social Sciences Citation Index covers STS journals. JSTOR archives major STS journals including Science, Technology, & Human Values, Social Studies of Science, and Technology and Culture. Sociological Abstracts covers the sociology of science and technology. Philosopher’s Index covers philosophy of science, including debates about scientific realism, the demarcation problem, and the structure of scientific revolutions.
Key journals in STS include Science, Technology, & Human Values, Social Studies of Science, Technology and Culture (published by the Society for the History of Technology), Isis (published by the History of Science Society), and Minerva. Students writing STS papers should use the citation style appropriate to their analytical approach — typically Chicago for historical and humanistic analysis, or APA for sociological analysis.
Students approaching science and technology from an STS perspective should be clear about whether their paper is a scientific paper (reporting or analyzing empirical findings using natural science methods) or an STS paper (analyzing science and technology as social and historical phenomena using humanities and social science methods), because the two types require different databases, methods, and citation styles.
Key Journals in Science and Technology
The two most prestigious general-science journals are Science (published by the American Association for the Advancement of Science) and Nature (Springer Nature). Both publish research across all scientific disciplines and are among the most highly cited journals in the world. Proceedings of the National Academy of Sciences (PNAS) publishes research across the sciences with a U.S. institutional focus. PLOS ONE publishes open-access research across all scientific disciplines.
Within disciplines, the leading journals include Cell and Molecular Cell (cell and molecular biology), Journal of the American Chemical Society (JACS, chemistry), Physical Review Letters (physics), The Lancet and New England Journal of Medicine (medicine), Angewandte Chemie (applied chemistry), Nature Communications and Science Advances (broad multidisciplinary), and IEEE Transactions series (engineering). Computer science is distinctive in that top-tier conference proceedings (NeurIPS, ICML, CVPR, ACL, SIGCOMM) carry as much or more weight than journal publications — students researching computer science topics should search conference proceedings through IEEE and the ACM Digital Library in addition to journal databases.
Professional Associations
Major professional associations publish leading journals, host conferences, and maintain resources for researchers and students. The American Association for the Advancement of Science (AAAS) publishes Science and advocates for science policy. The American Chemical Society (ACS) publishes more than 80 journals and maintains the CAS databases. The American Physical Society (APS) publishes the Physical Review family of journals. The American Institute of Biological Sciences (AIBS) and the Federation of American Societies for Experimental Biology (FASEB) serve the biological sciences. The Institute of Electrical and Electronics Engineers (IEEE) and the Association for Computing Machinery (ACM) serve engineering and computer science.
Most of these associations offer student membership at reduced rates, including journal access and conference discounts. Many also maintain career resources, mentoring programs, and undergraduate research opportunities. The National Science Foundation (NSF) funds Research Experiences for Undergraduates (REU) programs at universities across the United States — intensive summer research opportunities that provide students with hands-on experience in scientific research.
Frequently Asked Questions
What is the best database for science research?
There is no single best database because the sciences span many disciplines. Web of Science and Scopus provide the broadest multidisciplinary coverage and citation tracking. Discipline-specific databases — PubMed for biomedical sciences, CAS/SciFinder for chemistry, IEEE digital library for engineering, ACM Digital Library for computer science — provide deeper coverage within their fields. Google Scholar provides useful broad searching and citation tracking across all disciplines.
How do I find datasets for my research paper?
Start with the data repository most relevant to your discipline: GenBank for genetic sequences, PDB for protein structures, NASA Earthdata for satellite data, Zenodo or Dryad for general scientific data. The Registry of Research Data Repositories (re3data) catalogs thousands of repositories searchable by discipline. Many published papers include links to their underlying datasets in the supplementary materials section or in a data availability statement.
How important are conference papers in science?
The importance of conference papers varies by discipline. In computer science and some areas of engineering, top-tier conference proceedings (published through IEEE or ACM) are considered equivalent to or more prestigious than journal publications, and students should search conference proceedings as part of their literature review. In most natural sciences, conference presentations are considered preliminary communications, and the definitive version of the research appears later in a peer-reviewed journal article. Students should ask their instructor whether conference papers are appropriate sources for their specific assignment.
Can I cite Wikipedia for scientific facts?
Wikipedia should not be cited in a scientific research paper because its content is collaboratively edited, lacks consistent expert authorship, and can be modified at any time. However, Wikipedia’s references sections often point toward peer-reviewed articles, government reports, and textbooks that can be cited. The strategy is to use Wikipedia for orientation and then follow its citations to the primary scientific literature.
Related Pages
- How to Do Research — Overview of research methods and discipline-specific guides
- How to Do Research on Mathematics — MathSciNet, arXiv, and mathematical writing
- How to Do Research on the Environment — EPA, NOAA, USGS, and environmental data
- How to Do Research on Health and Medicine — PubMed, CDC, and medical research
- How to Write a Research Paper — Complete step-by-step guide