What AAPOR standards require of US survey research teams

What to take away

• Survey research standards in the United States center on the American Association for Public Opinion Research (AAPOR) code and its transparency disclosure requirements.
• AAPOR expects teams to publish question wording, sampling frame, mode, fieldwork dates, and weighting details alongside results.
• Response rate reporting follows a defined formula, and nonresponse must be described rather than hidden.
• Pew Research Center studies, including work from the American Trends Panel, show what full disclosure looks like in practice.
• The Federal Trade Commission's substantiation standard overlaps with AAPOR duties when survey results support advertising claims.
• A written disclosure checklist keeps a research team consistent across vendors, panels, and client reporting.

What AAPOR standards cover for US survey teams

The American Association for Public Opinion Research is the main professional body setting survey research standards in the United States. Its code of conduct and transparency initiative apply to anyone who designs, fields, analyzes, or reports survey data. The rules reach commercial firms, universities, nonprofits, and government contractors alike.

The code addresses several duties at once. Researchers must treat respondents fairly, protect confidentiality, avoid conflicts of interest, and report methods honestly. Disclosure is the part clients feel most directly, because it shapes what appears in a published report.

AAPOR's transparency disclosure requirements ask for the information a reader needs to judge a result. That includes who sponsored the study, who conducted it, the population studied, the sampling frame, and the mode of interview. Fieldwork dates, sample size, and exact question wording belong there too.

For teams running market research surveys for corporate clients, this changes the deliverable. A slide deck with a topline is not enough. The report needs a methods statement that a skeptical reader can check.

AAPOR is not a regulator. It cannot fine a firm or revoke a license. Its power comes from professional norms, journal requirements, and the fact that courts and regulators sometimes look at whether a study followed accepted practice.

Other bodies overlap with this work. The Insights Association, formed from CASRO and the Marketing Research Association, publishes its own standards for US firms. ESOMAR has US members and a global code. Many teams align with AAPOR first because its transparency rules are the most explicit about published reporting.

Transparency disclosures required in published reports

A published report should let a reader reconstruct the study. AAPOR's transparency initiative lists the elements that belong in that record, and the list is longer than most marketing decks assume.

Start with identification. Name the sponsor and the research organization. If a client commissioned the work, say so. If a vendor subcontracted fieldwork, that belongs in the record too.

The population and sampling frame come next. State who was eligible, how they were selected, and what list or panel supplied the sample. A probability sample from an address-based frame is a different object from a convenience panel, and the report should say which one was used.

Mode and fieldwork matter. Was the survey conducted online, by phone, by mail, or in person? What were the first and last dates of data collection? A result collected over three days in July is not the same as one collected across six weeks.

Question wording is a disclosure item, not an appendix afterthought. Report the exact text of every question used for a published figure, including response options and whether the order was randomized.

Weighting belongs in the methods statement. Say what variables were used, what benchmarks they came from, and whether the weights were trimmed. If the study used a model-based adjustment, describe it.

Finally, report the margin of sampling error and the design effect where they apply. These figures tell a reader how much precision the estimate carries. Many teams now publish a short methods page alongside the topline, which is the pattern Our Methods | Pew Research Center uses across its work.

Question wording and pre-testing expectations

AAPOR treats question wording as a source of error, not a matter of style. A leading question produces a leading result, and the code expects researchers to test wording before fielding it at scale.

Pre-testing takes several forms. Cognitive interviews ask a small number of respondents to explain what a question means to them. Expert review catches double-barreled items and unclear terms. A pilot fielding checks skip logic, timing, and refusal patterns.

Small wording changes move answers. Asking whether someone supports a policy versus whether they would accept it can shift the distribution. AAPOR's disclosure rules do not ban such wording, but they require that the exact text be published so readers can judge it.

Order effects also belong in the record. If a battery of items was rotated, say so. If a question followed a prompt that primed a topic, disclose the order.

The same discipline applies to open-ended coding. Report who coded responses, how many coders worked independently, and how disagreements were resolved. This is basic common market research strategy questions territory for teams that field qualitative and quantitative work together.

Translation adds another layer. For Spanish-language instruments in the United States, describe the translation and back-translation process. A question that works in English may not carry the same meaning after translation.

Response rate and nonresponse reporting

Response rate reporting follows a defined formula under AAPOR conventions. The rate is the number of complete interviews divided by the number of eligible sample units, with partial interviews and unknown eligibility handled by stated rules.

AAPOR publishes several response rate definitions, from RR1 to RR6, which differ in how they treat partials and cases of unknown eligibility. A report should name which definition it used. Saying only that the response rate was low tells a reader nothing.

Nonresponse is the harder disclosure. A low response rate does not automatically mean a biased estimate, but it raises the risk. The report should describe what is known about nonrespondents and whether weighting corrected for the difference.

Panel surveys face a related issue. Attrition removes people over time, and the remaining sample may differ from the original. Disclosure should state the retention rate and any refreshment sampling used to offset it.

For business-to-business studies, the eligible population is often uncertain, which makes the denominator hard to define. State the assumption you used. A stated assumption is honest. A missing denominator is not.

Teams that keep a diy versus agency consumer research cost record of every fielding will find this section easy to write, because the numbers already exist in the fieldwork system.

Panel and weighting disclosure in practice

Panel-based research dominates US commercial survey work, and it carries its own disclosure duties. A report should name the panel, describe how members were recruited, and state whether the sample was drawn from the full panel or a targeted subgroup.

Incentives belong in the record. If respondents received points, cash, or a chance at a prize, say so. Incentive structure can affect who joins and who stays.

Weighting is where panel studies earn or lose credibility. A typical approach uses raking to align the sample with known population benchmarks on age, sex, education, race, and region. The report should name the benchmarks, which often come from U.S. Census Bureau data or the Bureau of Labor Statistics.

Design weights handle unequal selection probabilities. Post-stratification weights correct for differential nonresponse. Both should be described, even briefly, and the report should note whether weights were trimmed to reduce variance.

Pew Research Center's American Trends Panel weighting is a useful reference point. The panel draws a nationally representative sample and applies weights that account for selection and nonresponse, with the details published for each study.

Teams writing their own methods statements can follow the structure described in the American Trends Panel (ATP) surveys included in the report | Pew Research Center.

For state-level work in California, New York, Texas, Illinois, and Florida, the benchmarks get thinner. A state estimate may need more aggressive weighting or a larger sample, and the report should say which tradeoff was made.

Examples from published studies that meet the code

Published studies from Pew Research Center show what full disclosure looks like when it is done consistently. The organization maintains methods pages for each report, and those pages carry the sampling frame, mode, fieldwork dates, sample size, and weighting description.

Its U.S. surveys documentation describes how samples are drawn and weighted, including the use of address-based sampling and online panels. That level of detail lets a reader evaluate a specific estimate rather than accepting a topline, and it is set out in U.S. Surveys | Pew Research Center.

The American Trends Panel methodology is published in full, covering recruitment, survey administration, and the weighting approach applied to each wave. A team building its own panel can use that document as a template for its methods statement.

The broader methods library covers question testing, mode effects, and nonresponse research, which is useful when a client asks why a wording change matters. Browsing the Methodological Research - Research and data from Pew Research Center is one of the cheaper ways to raise a team's disclosure quality.

Not every commercial study needs the same depth. A short client survey with a convenience sample can disclose its limits in a paragraph. The point is that the limits are stated, not omitted.

When a result travels beyond the client, the disclosure travels with it. Journalists, regulators, and opposing counsel all read methods sections. A study that meets the code is harder to dismiss. For teams choosing which questions to field, the same discipline applies to market research strategy examples, where wording and purpose are set before fieldwork begins.

Where AAPOR standards intersect with FTC substantiation

The Federal Trade Commission requires advertisers to have a reasonable basis for claims before they run. For survey-based claims, that basis often means the study itself must be sound enough to support the number quoted.

This is where AAPOR standards and FTC substantiation overlap. A claim that "most consumers prefer" a product needs a study with a defined population, a disclosed question, and a sample that can bear the weight of the claim. A convenience poll of a few hundred panelists may not.

The FTC looks at the claim as consumers would read it. If the ad says a national survey found something, the study should actually be national. If it says the result is representative, the weighting should support that word.

Disclosure helps here. A methods statement that names the sample, the question, and the margin of error gives the advertiser a documented basis. Without it, the claim rests on a number no one can check.

Endorsement and testimonial rules add a related layer. If a survey result is used in advertising alongside a testimonial, the disclosure requirements stack. The advertiser must substantiate the claim and disclose any material connection.

State consumer protection laws in California, New York, Texas, and elsewhere often mirror the FTC standard, so a weak study creates exposure beyond the federal agency. Research managers should treat substantiation as a reason to document methods, not an afterthought.

A disclosure checklist for survey research standards

Use this list before a report leaves the building. It maps to the transparency requirements and to what a reader, regulator, or opposing expert will look for.

• Sponsor and research organization named
• Population, sampling frame, and panel source stated
• Mode of interview and fieldwork dates reported
• Exact question wording and response options published
• Sample size, margin of error, and design effect given
• Response rate definition named and nonresponse described
• Weighting variables, benchmarks, and trimming explained

A worked example shows how the pieces fit. Suppose a team fields a national online survey of 2,000 adults on a consumer topic.

The methods statement would name the sponsor, identify the panel, state that fieldwork ran for two weeks, and publish the question text. It would report the response rate under a named AAPOR definition, describe the raking weights against Census benchmarks, and give the margin of error.

If the result later appears in an advertisement, the same statement becomes the substantiation file. That is the practical payoff of survey research standards: one document serves the code, the client, and the regulator.

Teams that build this into their template do it once and reuse it. Teams that write methods from scratch for each report tend to skip items under deadline. The template is the cheaper path, and it aligns with wider conjoint analysis consumer research that treat documentation as part of the analysis rather than an extra step.

Common questions

Does AAPOR have legal authority to enforce its code? No. AAPOR is a professional association, not a regulator. Its standards carry weight through professional norms, journal requirements, and the fact that courts and the FTC may consider whether a study followed accepted practice.

What response rate definition should a report name? Name the specific AAPOR definition you used, from RR1 through RR6. The definitions differ in how they treat partial interviews and cases of unknown eligibility, so the label matters as much as the number.

Is a low response rate automatically a problem? Not by itself. What matters is whether nonresponse biased the estimate and whether weighting corrected for it. The report should describe what is known about nonrespondents rather than hide the rate.

Do panel surveys need different disclosure than probability samples? Yes. Panel studies should name the panel, describe recruitment and incentives, report attrition, and explain the weighting model. Probability samples still need the frame and selection method described.

How does FTC substantiation relate to AAPOR disclosure? The FTC requires a reasonable basis for advertising claims before they run. A survey used to support a claim needs disclosed methods so the advertiser can show the study actually supports the wording.

Where can a team see full disclosure done well? Pew Research Center publishes methods pages with each report, including sampling, fieldwork, and weighting details, as shown in Pew Research Center methods. Those pages are a practical template for a team's own methods statement.