Standards & Ethics

Home
About AAPOR
Standards & Ethics

The AAPOR Code of Professional Ethics and Practices expresses the principles that we, as an association, believe support sound and ethical practice in the conduct and use of public opinion and survey research. Article IX of the AAPOR Bylaws requires the Standards committee to review the Code at least every five years to keep it current with changing environmental circumstances.

In May 2009, Council authorized the development of an ad hoc committee to review the current Code and make recommendations to the Standards committee regarding additions and revisions that are deemed necessary to update and strengthen the Code. The review committee included: Sandra Bauman, Jill Darling, Jennifer Franz, Tom Guterbock, Deborah Jay, Gary Langer, Ron Langley, Jay Leve, Mary Losch (chair), and Tom Smith. Individually and collectively, they brought a wealth of expertise and admirable professional commitment to this project, and AAPOR owes a debt of gratitude to them for their tireless efforts in this revision.

The Standards committee reviewed the ad hoc committee’s recommendations and offered several revisions of its own. Subsequently, Council accepted the majority of the recommendations, revised some and ultimately crafted the version that has been posted online.

Summary of Major Revisions and Rationale

The obligations of integrity and transparency extend not just to AAPOR members. The applicability of these principles to all research professionals has been made clearer in the preamble (3rd paragraph).

The Code is a set of principles. The preamble (4th paragraph) now specifies that judgments on the merits of specific research methods will be communicated elsewhere.

The Code was reordered to improve its organization, to emphasize the importance of our professional responsibility toward respondents and prospective respondents by placing those principles first, and to place the standards for disclosure immediately following the principle calling for accurate and detailed disclosure.

Respect for respondents’ concerns about privacy was expanded to clarify the definition of privacy as the desire not to provide personal information. The principle of integrity also calls for full disclosure when potential respondents ask how their contact information was obtained (I-A3).

The need for full disclosure regarding the survey’s sponsorship and purpose has also been added, with an exception when such knowledge could lead to bias or endangerment (I-A4).

New technologies and existing databases increase the risk of disclosure of identities from otherwise deidentified survey data. The principles regarding respondents’ confidentiality were expanded to include the need to protect against deductive disclosure (I-A5).

The revised Code makes clear that nondisclosure agreements with private clients should not be used to prevent disclosure of the conduct and findings of the research if the private client makes factual misrepresentations or distortions of the data in a public forum, or if the work becomes the subject of a formal confidential investigation of an alleged violation of the Code (I-B1).

Transparency regarding errors, factual misrepresentations, and distortions is given added emphasis in the revised Code (I-C3).

AAPOR has an ongoing interest in expanding the number of researchers and research organizations that agree to the principles of integrity and transparency, but like membership in the organization, adherence to the Code should not be cited as evidence of professional competence (I-D2).

Inappropriate use of statistics such as margin of sampling error is one way that interpretations of data can be accorded greater confidence than is warranted. The Code was modified to specifically note that such claims of precision should take into account the methods used (II-A4).

Prohibitions on fabrication and falsification are now included in the Code (II-A5), consistent with standards codes from other social science associations.

One of AAPOR’s goals is to ensure proper use of public opinion and survey research. This goal calls on us to accurately describe public opinion and survey research conducted not only by ourselves, but also by others (II-A6).

A renewed commitment to transparency calls on us to maximize disclosure regarding how research was conducted rather than limit disclosure to the minimum necessary. With that interest in mind, the Code no longer refers to “minimal disclosure.” The list of items that compose full and complete disclosure of essential information was expanded to include: who funded the survey (III-A1), the geographic location of the population under study (III-A3), visual or sensory exhibits (III-B2), estimates of the sampling frame’s coverage of the target population (III-B3), compensation or incentives offered (III-B5), details regarding the design effects and the development of sampling weights (III-B7), procedures to verify data (III-B8), and methods of interviewer training and monitoring (III-B8). Response rates are not part of required disclosure (III-C), but disclosure of summaries of the disposition of sample records remains essential.

The list of items that compose full and complete disclosure has also been expanded to account for issues that have accompanied the growth of online survey research, especially with pre-recruited panels or pools of respondents. Regardless of how data were collected, researchers should disclose the names of any sampling frame suppliers (III-A4), the methods used to recruit the panel if a panel was used (III-B4), and sufficient detail to determine whether the data were obtained from a probability or nonprobability sample (III-A4).

The survey research field has seen increased use of multiple frames and multiple methods within the same project. Transparency regarding each frame and each method is required (III-D).

Finally, the items that compose full and complete disclosure were divided into two tiers. Ideally, full disclosure of all required information would occur at the same time as the release of findings. Yet, some required disclosure elements may not be available (or available in an accessible form) at the time that the research results are released. The use of tiers permits different timeliness standards regarding how quickly disclosure must occur. Different timeliness standards for each tier are consistent with the Principles of Disclosure from the National Council on Public Polls, though the specific elements in each tier differ slightly between the Code and those Principles.

On March 12, 2010, Council agreed on the draft of the revised Code and posted it online. Interested members were invited to submit written comments on or before Friday, April 2, 2010.

The Standards Committee and Council reviewed the comments and a final revised Code, as agreed to by Council, was presented to the membership for its approval by way of a ballot. The revised Code was approved overwhelmingly by AAPOR members and announced at the association’s 65^th Annual Conference in Chicago.

Read the approved, revised Code.