The American Statistical Association is the largest community of statisticians in the world. Members use, develop, and share statistical practices from their roles in industry, business, government, and academia. Recognizing statistical practice includes activities such as designing the collection of, summarizing, processing, analyzing, interpreting, or presenting data and model or algorithm development and deployment, the ASA Ethical Guidelines for Statistical Practice aim to promote accountability by informing those who rely on any aspects of the statistical practice of the standards they should expect.

Given the critical role of statistics across many diverse disciplines and professions, the ASA’s mission—Promoting the Practice and Profession of Statistics—has an important advocacy element in the ASA Ethical Guidelines for Statistical Practice. These guidelines “are intended to help statistical practitioners make decisions ethically.”

The first-ever revision of the original 1994 guidelines took place in 2014–2015, with the ASA Board approving the new version in 2016. The board also requested the guidelines be updated every five years. In 2018, the ASA Board requested an interim review to add language that sexual harassment, bullying, and other forms of intimidation are prohibited. In 2021, the Committee on Professional Ethics developed a working group and process to update the guidelines for release in 2022. The ASA Board approved the revised version in February 2022.

The 2016 version broadly defined a statistician as anyone who uses statistics, irrespective of job title and training, but retained the title statistician. The 2021 revisions went further, replacing the title statistician with the more general term statistical practitioner. This change broadened the reach of the guidelines to all who use statistical practices in any domain or role, including data science and analytics. The 2022 version also specifies how the guidelines are relevant to organizations that engage in statistical practice.

Structure of the revised version compared to the earlier guidelines, with the revised parts highlighted in red

2022 Revision Process

The Committee on Professional Ethics consists of nine members, serving (at most two) three-year terms. The ASA seeks to balance membership among industry, government, and academia. In 2018, the committee instituted an online input collection mechanism so all ASA Community members could comment on and make suggestions for changes to the Ethical Guidelines for Statistical Practice. The committee members reached out to sections, committees, and others to encourage engagement with the guidelines revision process.

A working group of four committee members, plus the committee chair, was formed in 2020 in preparation for the work of review and revision to come in 2021. Additionally, Rochelle Tractenberg, the former committee chair and friend of the committee after serving two terms, who had led the previous two revision efforts, co-chaired the working group, which included the following:

• Jing Cao (co-chair)
• Rochelle Tractenberg (co-chair)
• Matthew Rotelli (vice chair)
• Jason Gillikin
• Marcia Levenstein
• Donna LaLonde (ASA liaison)

During 2021, the working group met weekly, moving from the title of the document (determining if the title should change from “statistical practice” to “the practice of statistics and data science”), to the preamble and through each principle in the order of the 2018 document. The member responsible for each principle under discussion made a preliminary pass through the relevant text, led that discussion, and maintained notes on suggested changes and discussions. All working group members contributed to all changes, and the co-chairs maintained one running document with all final changes tracked.

The working group’s progress was presented to the Committee on Professional Ethics at its monthly meetings. Other committee members and friends of the committee could comment on changes, make suggestions for revisions, and participate in the working group meetings. These committee members and friends included Harold Gomes, Nilupa Gunaratna, Andrew Hartley, Stephanie Shipp, Millennia Young, and Michael Hawes. The committee unanimously approved changes as the working group completed its work throughout 2021.

Overview of the Revisions

The guidelines have historically been based on transparency, fairness, accountability, and respect, and the 2021 revisions maintained this emphasis. The committee’s goal was to update, clarify, and make the guidelines inclusive, relating them to all statistical practices in all disciplines, jobs, and roles. Given the core roles of statistics in a wide range of fields, the working group agreed the guidelines would be recognizable and applicable to anyone practicing statistics, including statisticians, data scientists, and those without these job titles but who engaged in any and all aspects of statistical practice in government, academia (including students), industry, and consulting. The guidelines defined statistical practices in the preamble to include “activities such as designing the collection of, summarizing, processing, analyzing, interpreting, or presenting data and model or algorithm development and deployment.”

This definition was part of the inclusivity aimed at ensuring individuals—irrespective of their workplace, job title, and level of training—recognize the relevance of the guidelines and ethical statistical practice.

To clarify the 2018 guidelines, the 2022 version clearly separated the obligations of employers and organizations/institutions from those of supervisors, which were made more explicit. The principle relating to allegations of misconduct was moved to Principle H and made less specific to “scientific misconduct,” which is a construct most specific to federally (US) funded research. Now Principle H relates to scientific, statistical, and professional misconduct (as part of the inclusivity and clarification aims).

Describing the Revisions

The preamble was revised to state the following purpose of the guidelines more clearly:

The preamble was also clarified in terms of the need to balance competing interests and the role of the guidelines in that balance:

The 2022 guidelines retain eight principles and an appendix, comprising 72 elements.

Changes in Guidelines

1. Principle A, Professional Integrity and Accountability, expanded from 7 to 11 elements. The principle was revised to be more general so its elements would more clearly instantiate how to exhibit or conform with the high-level principle. Given the increasing importance of data science and prevalence of statistical practice by data scientists, the elements were revised to be more inclusive of nonexperimental data and a wider diversity of roles of the individual who seeks to use statistical methods/practice in an ethical manner.
2. Principle B, Integrity of Data and Methods, decreased from 11 to 7 elements by refining existing elements and ensuring they were relevant for experimental/survey/designed and not-experimental/not-survey/not-designed data collection efforts.
3. Principle C was clarified to Responsibilities to Stakeholders, with the role of stakeholder, rather than a few specific stakeholders, being the focus of articulated responsibilities.
4. Principle D, Responsibilities to Research Subjects, Data Subjects, and Those Directly Affected by Statistical Practices, was expanded to include more than human subjects in designed experiments.
5. Principle E, Responsibilities to Multidisciplinary Team Colleagues, was clarified to highlight multidisciplinary team colleagues, not just single-discipline colleagues, and updated to reflect colleagues in projects including research.
6. Principle F, Responsibilities to Fellow Statistical Practitioners and the Profession, was expanded to include responsibilities for promoting integrity in “the practice and profession.”
7. One original principle was divided for clarity and inclusiveness, but also to update the guidelines, reflecting the critical role of statistical practices in many diverse workplace environments. The original became:
   • Principle G, Responsibilities of Leaders, Supervisors, and Mentors in Statistical Practice, clarifies the specific responsibilities to promote ethical statistical practice that accrues as the practitioner’s role incorporates leadership, instruction, and mentorship. This principle was formulated with a new focus on specific responsibilities for leadership roles. It was created by revising a previous principle relating to employers.
   • An appendix with two parts. Part 1 defines the responsibilities of organizations/institutions employing statistical practice. Part 2 describes the responsibilities of those in leadership, supervisory, or managerial positions who oversee statistical practitioners but may not be practitioners themselves.
8. Principle H, Responsibilities Regarding Potential Misconduct, was moved to the final position, and the applicability beyond scientific misconduct was expanded, updated, and clarified.

The Committee on Professional Ethics forwarded the recommended revisions to the ASA Board in November 2021 for discussion and comment, and the ASA Board approved the final version in February 2022. Also in 2022, the committee developed and implemented an education program to use statistics to socialize the Ethical Guidelines for Statistical Practice across disciplines.

Committee Outreach: Webinars, Podcasts, and Conference Sessions

To introduce the new guidelines, the Committee on Professional Ethics held two webinars in May and June of 2022. Each webinar covered about half the guidelines. A representative from academia, government, and industry presented the application to specific guideline principles and elements to vignettes from their career experiences. Both webinars covered issues around handling disagreements about methods, analysis, results, and interpretation and authorship and co-authorship.

Selected Vignettes from the May and June 2022 Webinars

During the May webinar, Nancy Potok—former chief statistician of the United States, CEO of NAPx Consulting, and senior fellow at The George Washington University—described the ethical challenges raised when preparing sensitive questions asked of minors in the screener for the National Crime Victimization Survey. She noted Principle D provided the needed guidance to handle this issue.

• D. Responsibilities to Research Subjects, Data Subjects, or Those Directly Affected by Statistical Practices
  • D2. Makes informed recommendations for sample size and statistical practice methodology to avoid the use of excessive or inadequate numbers of subjects and excessive risk to subjects.
  • D4. Protects people’s privacy and the confidentiality of data concerning them, whether obtained from individuals directly, other persons, or existing records. Knows and adheres to applicable rules, consents, and guidelines to protect private information.
  • D6. Considers the impact of statistical practice on society, groups, and individuals. Recognizes that statistical practice could adversely affect groups or the public perception of groups, including marginalized groups. Considers approaches to minimize negative impacts in applications or in framing results in reporting.

  At the June webinar, Jennifer Ortman—principal demographer in the U.S. Census Bureau Population Division—described her experiences transitioning from peer to supervisor. She connected the relevant principles and elements that guided her:

  • E. Responsibilities to Members of Multidisciplinary Teams
    • E. It is essential to have a respectful exchange of views.
  • F. Responsibilities to Fellow Statistical Practitioners and the Profession
    • F1. Constructive discourse with mutual respect focuses on scientific principles and methodology and not personal attributes.
    • F2. Help strengthen, and do not undermine, the work of others through appropriate peer review or consultation.
  • H. Responsibilities Regarding Potential Misconduct
    • H1. Seek to clarify facts and intent before alleging misconduct by others.

  We encourage you to review the guidelines, webinars, and other materials below to learn more about how to use the ASA Ethical Guidelines in your work. Questions can be sent to Committee on Professional Ethics co-chair, Jing Cao, or committee vice chair, Matthew Rotelli.

  Want to Know More?

  • Committee on Professional Ethics
  • ASA Ethical Guidelines for Statistical Practice
  • Webinar on Principles A, B, C, and D
  • Webinar on Principles E, F, G, H, and the Appendix
  • Practical Significance, Episode 22: What’s in the 2022 Ethical Guidelines for Statistical Practice?
  • Conference on Statistical Practice Presentations, February 2023