Laws, regulations, court opinions, government reports, census data, agency statistics, and legislative records — and how to find, read, and evaluate each with care
Imagine you are writing a research paper on minimum wage policy. You find two sources: a blog post from a political advocacy group arguing that minimum wage increases destroy jobs, and a report from the U.S. Bureau of Labor Statistics presenting employment figures by industry and wage category over a ten-year period. Both sources are real. Both may be useful. But they are not the same kind of source, and they do not carry the same kind of authority.
Official sources — government reports, legal texts, census data, agency statistics, court opinions, policy documents, and public records — occupy a distinctive place in research. They are often the closest available thing to primary evidence on questions of law, policy, population, and public action. A researcher asking what the law says should look at the law. A researcher asking what a government agency found should look at the agency's published findings. A researcher asking how many people live in a region, what they earn, or how they are distributed across industries should look at census data and official statistics.
This does not mean official sources are always correct, always complete, or always free of political influence. Government agencies can suppress data. Official statistics can be collected with flawed methods. Legal documents can be misread or taken out of context. Policy documents can advocate positions rather than present balanced evidence. Learning to work with official sources — to find them, read them carefully, and evaluate their authority and limits — is one of the most important practical skills a researcher can develop.
Suppose you encounter a claim that begins, "According to government data..." How would you trace that claim back to its actual source? What would you need to know — about the agency, the specific publication, the year, and the definitions used — before you could decide whether the claim is accurate? By the end of this chapter, you will have a systematic method for answering exactly that kind of question.
This chapter teaches students how to locate, read, and evaluate government, legal, and statistical sources. By the end of this chapter, you should be able to identify when an official source is appropriate, navigate the major categories of government and legal documentation, understand how census data and agency statistics are collected and used, and apply the MGU Source Evaluation Model to official sources with the same critical care you would apply to any other source type.
Official sources are used in law, public health, policy analysis, journalism, economics, history, sociology, education, and almost every other field that touches public life. Students who learn to find them, read them, and evaluate them well have a significant research advantage — and a defense against the misuse of official-sounding claims that circulate without proper attribution.
When researchers need to know what the law actually says, what a government agency actually reported, how many people actually live in a place, or what policies a legislative body actually passed, there is no substitute for going to the official record. Opinion pieces, news articles, and secondary analyses can all describe or interpret these things — but they interpret official sources. The researcher who goes directly to those sources works from a stronger foundation.
Democratic governments in most countries maintain official records of their activities. These records exist because accountability and transparency are foundational values in democratic governance. Laws are published so citizens can know what they are required to do. Court opinions are published so the reasoning behind legal decisions can be reviewed and contested. Census data is collected and published so policymakers, researchers, and the public can understand the population they serve. Agency reports document how government programs work, what they have found, and what they recommend.
These records are generally public because they are produced by public institutions using public resources. Many are freely available online. Many others can be accessed through libraries. Some are restricted by security classification or by privacy protections on individual-level data, but the vast majority of government-produced information is available to anyone who knows how to find it.
Official sources serve several purposes that other source types cannot easily replicate.
| Purpose | What It Means |
|---|---|
| Primary evidence of law and policy | If your question involves what the law requires or what a court decided, you need the legal documents themselves, not descriptions of them |
| Primary evidence of government action | Reports, hearings, and agency publications document what public institutions have actually done, decided, funded, or recommended |
| Primary statistical evidence | Census data and agency statistics come from large-scale systematic collection that individual researchers typically cannot replicate |
| Authoritative reference points | Going to the original lets you verify whether a statistic or legal description has been cited correctly |
Official sources are not automatically correct, objective, or complete. Several important cautions apply: political influence on data collection and framing; methodological limits such as sampling error and definitional choices; selective or incomplete reporting; outdated information as laws and data age; and access barriers behind paywalls or in physical archives.
Government agencies operate within political environments. Data collection priorities, reporting practices, and the framing of findings can all be influenced by the political context in which an agency operates. Treating "official" as synonymous with "objective" is one of the most common errors researchers make with this source type.
With these cautions in mind, official sources remain among the most important categories of evidence in research. The task is not to trust them uncritically but to use them intelligently.
Legal documents form the foundation of research on law, public policy, rights, obligations, compliance, and governance. Understanding the basic categories of legal sources and how to find them is essential for any researcher working on questions that intersect with law.
In the United States and in most constitutional democracies, legal authority is structured in layers. Understanding this structure helps you identify which type of legal document you need and which has greater authority.
| Layer | Produced By | Function |
|---|---|---|
| Constitutional provisions | Constitutional conventions / amendment processes | Establish the fundamental structure of government and protect individual rights |
| Statutes | Legislatures (Congress, state legislatures, local councils) | Laws that must be consistent with constitutional requirements; collected in codes such as the U.S. Code |
| Regulations | Executive branch agencies | Rules issued under authority granted by statute; published in the Federal Register and codified in the CFR |
| Court opinions | Federal and state courts | Interpret statutes, regulations, and constitutional provisions; become binding precedent |
| Local ordinances | Cities, counties, school districts | Laws passed by local governments, consistent with state and federal law |
Federal statutes are published in the United States Code (U.S.C.), organized by subject area into titles. Title 26, for example, is the Internal Revenue Code; Title 42 covers public health and welfare.
The U.S. Code is freely searchable at uscode.house.gov (maintained by the House of Representatives Office of Law Revision Counsel) and at govinfo.gov (maintained by the Government Publishing Office). Both are official government sources. The Cornell Legal Information Institute at law.cornell.edu also provides a freely accessible, searchable version of the U.S. Code, along with helpful explanatory material.
When you find a statute, note the title and section numbers, the official name of the act, and the date of enactment or most recent amendment.
Regulations in effect are found in the Code of Federal Regulations (CFR), organized by title corresponding to different areas of federal regulation. The CFR is freely available at ecfr.gov and govinfo.gov.
New and proposed regulations are published in the Federal Register, the official daily journal of the federal government, searchable at federalregister.gov and govinfo.gov. Researchers studying regulatory history, the notice-and-comment process, or proposed but not-yet-finalized rules should use the Federal Register.
State statutes and regulations are available through state government websites. The structure varies by state, but most states maintain searchable online codes and registers. The Cornell LII's state pages provide links to official state code repositories for all fifty states.
Legal documents use technical language that takes time to learn. Several practical strategies help.
Read the full provision before drawing conclusions. Legal meaning depends heavily on context. A single sentence may be qualified by definitions elsewhere in the code or by exceptions in adjacent sections.
Look for definitions sections. Most statutes and regulations specify how key terms are used throughout, and those meanings often differ from everyday usage.
Check for amendments. Laws are frequently amended; an older version may not reflect current law.
Use explanatory resources cautiously. News articles and agency summaries interpret rather than reproduce the law. Orient yourself, then go to the original.
Know your limits. Legal research is a professional skill. Cite the primary documents you have read, note any interpretive resources you consulted, and be appropriately modest about complex legal questions.
Court opinions are one of the most important and distinctive source types in legal and policy research. They do not merely state what the law is; they explain — sometimes at length — how legal principles apply to specific facts.
The U.S. court system has two parallel structures: the federal courts and the state courts. Federal courts handle cases involving federal law, constitutional questions, and disputes between states, with three levels: district courts (trial courts), circuit courts of appeals, and the Supreme Court of the United States. State courts handle most civil and criminal matters arising under state law, typically including trial courts, intermediate appellate courts, and a court of last resort.
Court opinions from higher courts are binding on lower courts within their jurisdiction — a principle called stare decisis (standing by decisions). Supreme Court opinions on federal law bind all U.S. courts. Circuit court opinions bind district courts within that circuit. State supreme court opinions bind lower courts within that state.
| Opinion Type | What It Is |
|---|---|
| Majority opinion | The official decision of the court; carries the force of law as precedent and explains the court's reasoning |
| Concurring opinion | Written by justices who agree with the outcome but for different reasons; influences but does not bind |
| Dissenting opinion | Written by justices who disagree with the majority; carries no precedential authority but may shape future reconsideration |
Supreme Court opinions are published in the United States Reports. Free access is available at supremecourt.gov, through the Cornell LII, and through Justia (justia.com).
Federal circuit court opinions are published in the Federal Reporter. Many are freely available through court websites and through PACER, the federal courts' electronic filing system, though PACER charges per-page fees. Justia and CourtListener (courtlistener.com) provide free access to many federal appellate opinions.
Commercial legal databases such as Westlaw and LexisNexis provide comprehensive coverage with sophisticated search tools and editorial annotations. These services are expensive but often available through university libraries.
State court opinions vary in accessibility; many state court systems post opinions on official websites, and the Cornell LII links to state databases.
Court opinions follow a conventional structure: a caption and docket number identifying the parties and court; a syllabus summarizing the holding (helpful but not citable as authority); the majority opinion itself, with facts, procedural history, the legal question, the holding, and the reasoning; and any concurrences and dissents, separately labeled and attributed.
Government agencies at the federal, state, and local levels produce a vast range of reports: research studies, program evaluations, audits, investigative reports, annual reports, environmental impact statements, budget documents, and more. These reports are among the most useful sources in policy research, public health research, environmental research, education research, and many other fields.
Research and evaluation reports are produced when agencies commission or conduct studies — the NIH funding biomedical research, the Department of Education studying interventions, the Department of Labor evaluating workforce programs.
Audit and investigative reports come primarily from the Government Accountability Office (GAO) and agency Inspectors General, evaluating whether programs work as intended and whether funds are used appropriately. GAO reports are notable for methodological rigor and political independence.
Annual and performance reports document agency activities over a reporting period, often required by Congress.
Environmental impact statements are required by the National Environmental Policy Act before major federal actions that may significantly affect the environment.
Budget documents — especially the President's annual budget submission and agency budget justifications — show how government resources are allocated.
Congressional reports accompany legislation, explaining committee intent and findings, important for understanding legislative history.
GovInfo.gov is the official portal for U.S. government publications, providing access to the Congressional Record, the Federal Register, the CFR, Supreme Court opinions, congressional reports, GAO reports, and more.
The GAO website (gao.gov) provides searchable access to GAO reports and testimony, organized by topic area.
Agency websites — the CDC, the BLS, the Census Bureau, the Department of Education's National Center for Education Statistics (NCES), and many others — maintain extensive publication archives.
Government reports vary considerably in quality, independence, and relevance. Ask: Who produced the report — an independent audit office or an agency reporting on its own programs? What was the methodology? What does the report actually claim to show, and over what scope? How current is it? Who funded the research, and are there potential conflicts of interest?
The decennial census of the United States is one of the largest and most important official data collection operations in the world. But the census is only the most prominent of a vast array of official data collection efforts that together constitute the national statistical system.
The U.S. Constitution requires a count of the population every ten years to apportion seats in the House of Representatives. The census collects basic demographic information — name, age, sex, race and ethnicity, relationship to household members, and housing tenure. The data are used for legislative apportionment, distributing federal funds, and as a foundation for other statistical programs.
Finding census data. Census data are available through census.gov and the search tool at data.census.gov, which allows searching by geography, topic, and survey, down to the national, state, county, city, census tract, block group, and block level.
Census limitations. The census aims for a complete count but never achieves one. Some populations — young children, renters, minorities, immigrants, and people experiencing homelessness — are systematically undercounted. The Census Bureau publishes coverage estimates quantifying the likely undercount by group and geography.
The American Community Survey (ACS) is a continuous survey that replaced the decennial census long form. While the decennial census counts the population, the ACS collects detailed social and economic data on an ongoing basis and is now the primary source for detailed demographic, economic, housing, and social data between decennial censuses.
The ACS produces 1-year estimates (for areas with populations of 65,000 or more) and 5-year estimates (for all geographic areas, including small places). The 5-year estimates are more reliable for small areas because they draw on larger samples, though they cover a five-year period rather than a single year.
| Program | Conducted By | Primary Use |
|---|---|---|
| Current Population Survey (CPS) | Census Bureau / BLS | Source of official national unemployment statistics |
| American Housing Survey (AHS) | Census Bureau / HUD | Size, composition, and quality of the U.S. housing stock |
| National Health Interview Survey (NHIS) | CDC / NCHS | Health status, behaviors, and health care access |
| NCES surveys | National Center for Education Statistics | Educational attainment, enrollment, performance, finances |
| Consumer Expenditure Survey (CEX) | BLS | Household spending patterns |
| American Time Use Survey (ATUS) | BLS | How people spend their time |
Researchers studying other countries or global trends should know the major international statistical sources: the World Bank DataBank for global development data; the United Nations Statistics Division for population, national accounts, energy, and trade; the OECD StatExtracts for member-country data; and Eurostat for European Union statistics.
Official statistics are numbers with stories behind them. Responsible use requires attention to several issues:
| Issue | Why It Matters |
|---|---|
| Unit of analysis | What is being counted — individuals, households, businesses? Median household income is not per-capita income. |
| Definitions | Official terms often have technical meanings that differ from everyday usage — for example, how "unemployed" is officially defined. |
| Margins of error | Survey estimates carry uncertainty, especially for small geographic areas or small subgroups. |
| Base rates and percentages | Percentages are only meaningful relative to their base — a large percentage change in a rare event may matter less than a small change in a common one. |
| Temporal comparability | Definitions and methods can change over time, complicating year-to-year comparisons. |
Beyond census and survey data, federal and state agencies collect and publish statistics as part of their regulatory, enforcement, and program administration activities. These administrative data sources are often the best available evidence for specific research questions.
Administrative data are records collected as part of government operations rather than for the specific purpose of statistical research — tax records, hospital discharge records, school enrollment records, court filings, and unemployment insurance claims, for example. When agencies compile and publish aggregated statistics based on these records, the result is agency statistical data.
Strengths: administrative data cover entire populations rather than samples, are collected continuously, and often include longitudinal information. Weaknesses: they reflect administrative rather than research needs, participation may be uneven, and record-keeping practices vary across jurisdictions.
| Agency | What It Publishes | Access Point |
|---|---|---|
| Bureau of Labor Statistics (BLS) | Unemployment, wages, inflation (CPI/PPI), workplace injuries | bls.gov |
| Bureau of Economic Analysis (BEA) | GDP, personal income, corporate profits, trade | bea.gov |
| Internal Revenue Service (IRS) | Income distribution and tax data (Statistics of Income) | irs.gov/statistics |
| CDC / National Center for Health Statistics | Birth and death statistics, hospital and outpatient data | cdc.gov/nchs |
| Bureau of Justice Statistics (BJS) | Crime, victimization, courts, corrections | bjs.ojp.gov |
| FBI Crime Data | Uniform Crime Reporting (UCR) and NIBRS data | ucr.fbi.gov |
| Environmental Protection Agency (EPA) | Air quality, emissions, toxic releases, water quality | EPA Envirofacts |
| National Center for Education Statistics (NCES) | Enrollment, graduation rates, test scores, school finances | nces.ed.gov |
When using agency statistical data, identify the source clearly: the agency, the specific program or publication, the year of data collection, and the publication or release date. Note whether the data come from a survey (with sampling) or from administrative records (with coverage limitations), and check for technical documentation explaining methodology and category definitions.
"According to the FBI, violent crime rates have declined" is not a citable claim without specifying which FBI publication, which years, and which definition of violent crime.
Policy researchers, political scientists, historians, and students in many fields need access to the documentary record of how laws are made and how policy decisions are reached. This includes committee hearings, floor debates, conference reports, presidential executive orders, agency rulemaking records, and a wide range of other materials.
The Congressional Record is the official record of the proceedings of the U.S. Congress, published daily when Congress is in session, available at congress.gov and govinfo.gov.
Congress.gov is the primary public portal for legislative information — bill text, bill status, committee information, voting records, and the Congressional Record.
Committee hearings and reports are published as Senate or House hearings on govinfo.gov. Committee reports accompany reported legislation and explain legislative intent.
Conference reports resolve differences between House and Senate versions of a bill and contain the final agreed text, available through govinfo.gov and congress.gov.
Executive orders are directives to federal executive agencies, published in the Federal Register and codified in Title 3 of the CFR, available at federalregister.gov and the American Presidency Project (presidency.ucsb.edu). Proclamations, memoranda, and signing statements are related presidential document types, with signing statements sometimes expressing the President's interpretation of specific provisions.
Agencies issuing regulations must follow the notice-and-comment process required by the Administrative Procedure Act. A Notice of Proposed Rulemaking (NPRM) is published in the Federal Register, explaining the proposed rule and its legal authority. Public comments submitted in response are publicly available through regulations.gov. The final rule, also in the Federal Register, includes a preamble responding to significant comments.
State legislative records are available through state legislature websites with varying accessibility. Local government records — city council minutes, county commission agendas, school board documents, planning commission records — are typically available through local government websites and public records requests.
The Freedom of Information Act (FOIA), 5 U.S.C. § 552, requires federal agencies to make their records available to the public upon request, subject to exemptions for national security, law enforcement, privacy, and other interests. State and local governments have parallel open records laws, sometimes called sunshine laws.
FOIA requests are submitted through agency FOIA portals or by mail; agencies must respond within statutory time limits and may charge fees, with waivers available for educational and news media requesters. The FOIA.gov website, maintained by the Department of Justice, explains how to submit requests and tracks agency processing statistics.
Understanding FOIA is important even for students who do not plan to submit requests themselves, because many documents obtained through FOIA requests are now publicly posted online by requesters, journalists, and advocacy organizations.
Official sources deserve critical evaluation just as much as any other source type. The fact that a source is produced by a government agency does not automatically make it accurate, complete, objective, or appropriate for a particular research purpose. Apply the MGU Source Evaluation Model systematically to every official source you use.
Relevance. Authority. Evidence. Accuracy. Purpose. Currency. Context.
| Element | Applied to Official Sources |
|---|---|
| Relevance | Does this source match your specific question in content, geography, time period, and population? |
| Authority | Who produced this, and do they have legitimate mandate to do so? A BLS report and a local task force report do not carry equal institutional weight. |
| Evidence | What data were collected and how were they analyzed? Do not treat an executive summary as equivalent to the full evidence base. |
| Accuracy | Can you verify figures against other sources? Is the legal text current? |
| Purpose | Why was this document produced — independent audit, agency self-report, regulatory justification, opposition comment? |
| Currency | Has the law changed? Is more recent data available? |
| Context | What political environment or methodological choice shaped this document? Has anyone contested its findings? |
Treating official as synonymous with unbiased. Official does not mean objective. Citing outdated data as current. Always check for the most recent release. Misreading legal provisions. Missing a qualifying clause or exception can seriously misrepresent the law. Overgeneralizing from official data. Conclusions should not exceed the actual scope of the data. Ignoring methodological documentation. Technical notes explain what the categories actually mean. Failing to cite precisely. "According to government data" is not a citation.
Identify a federal law relevant to one of the following topics: environmental protection, workplace safety, civil rights, disability rights, or food safety. Use congress.gov or uscode.house.gov to find the official text.
Go to data.census.gov and find data for your home county (or any county you choose) on at least two of the following: median household income, educational attainment, population by age group, housing tenure, or health insurance coverage.
Go to gao.gov and search for a GAO report on a topic related to education, health care, veterans, the environment, or housing, published within the last five years.
Source: Bureau of Labor Statistics. The Employment Situation — [Current Month]. U.S. Department of Labor. Released monthly.
Is this source relevant to your research question? Useful for unemployment trends, employment by industry, or labor force participation; less directly relevant to housing prices or health outcomes.
BLS is the principal federal statistical agency for U.S. labor market conditions, with a statutory mandate and a long track record of methodological rigor.
The report draws on the Current Population Survey (roughly 60,000 households) for the unemployment rate, and the Current Employment Statistics survey (roughly 145,000 businesses) for payroll employment. Both have well-documented methodologies.
Employment figures are revised the following month and again in annual benchmark revisions. Note whether you are using preliminary or revised data.
A regular statistical release fulfilling BLS's mission of providing accurate, timely labor market information — not an advocacy document.
Released the first Friday of each month; for current conditions, use the most recent release.
The official unemployment rate uses a specific technical definition that excludes people marginally attached to the labor force; BLS also publishes broader underutilization measures (U-1 through U-6).
A high-authority, well-documented source for labor market data. Cite the specific release date, note the definition of unemployment used, and check whether preliminary figures have been revised.
Below is a model MGU Research Note for a government report, following the MGU Research Note Model: Source, Summary, Key evidence, Quotation, Paraphrase, Usefulness, Limits, Citation information.
Three common errors recur with government and legal sources.
A claim presents a simplified, distorted, or selective version of a legal provision — "the law says you can't do X" — when the law actually includes significant exceptions or has been qualified by court interpretation. What to do: go to the primary legal source, read the full provision including definitions and exceptions, and check whether courts have interpreted it.
A widely circulated statistic about poverty, crime, or graduation rates is actually several years old and may have changed significantly. What to do: always identify the year of data collection, and search for the most recent release before citing any figure as current.
A claim attributes a statistic to the wrong agency or program, making it difficult to verify or properly cite. What to do: trace every statistic to its original source rather than relying on a secondary citation.
Who made it? The claim invokes the Census Bureau, but it is unclear who is actually making the claim — the Bureau itself, or someone citing it. What is the claim? That more than half of Americans live in poverty. What evidence is offered? A vague attribution only. What is missing? The specific publication, table, year, and definition of poverty used. Current official figures show poverty rates well below 50%. Can another reliable source confirm it? No — official poverty data at census.gov contradict the claim. Is the emotional reaction stronger than the evidence? Yes, which is a reason for more scrutiny, not less. Has the source been altered or misrepresented? This appears to be a false or wildly distorted attribution.
Conclusion: The claim is false and should not be repeated or shared.
Different citation styles handle government and legal sources differently, but the key elements to record remain the same.
Omitting the year of data collection (different from the year accessed); citing "the Census Bureau" generally without specifying the survey and year; citing a case name without the reporter citation and year; citing a statute without the U.S. Code title and section number; attributing an institutional report to an individual author.
Chapter 15 examined the official record of law, policy, and population. Chapter 16 broadens the picture of evidence further, moving beyond written text into data sets, white papers, images, maps, audio, video, interviews, archives, letters, diaries, and artifacts. You will learn how to handle nontraditional source types carefully, applying the same evaluative habits introduced in this chapter to evidence that does not arrive as a sentence on a page.