Thomas W. Watson
Armstrong Laboratory

Surveys and interviews may be intrusive when conducted for purposes other than occupational analysis. Attitudes, beliefs, interests, values, behaviors and background data are requested which may be personal and possibly controversial. If some of this information were linked to individuals, it could put them in jeopardy. Thus, survey and interview professionals must protect each participant’s well being to prevent harm and to get accurate information. This paper examines survey and interview ethics and is based on the more detailed information in Watson’s (1997a, 1997b) ethics chapters and appendices.

The ethical principles which guide data gathering are rooted in two inalienable human rights: free speech and privacy. Survey and interview participants have the right to: 1) speak freely, without constraint, even if others may not like what they say, and 2) remain silent, or if they speak, to set limits on the personal information they divulge, and have what they say as individuals remain confidential (unless they consent to disclosure).

Free Speech. The right to speak freely is protected by the First Amendment. In the historical defense of free speech, Supreme Court Justices Holmes and Brandeis have argued that people should be free to discuss and debate ideas without the censorship of prevailing doctrine and that in the free market place of ideas, good ideas remedy bad ones (Alderman & Kennedy, 1991).

Permitting free speech does not guarantee it will occur. Speech and other forms of expression are usually constrained. Some laws constrain speech and expression; organizational, family, group and societal norms constrain speech to that which is accepted or expected in certain contexts; fear of harassment or punishment also constrains speech. Thus, much goes unsaid--information that could be helpful to parents, teachers, health and social service agencies, politicians, manufacturers, employers, and others who wish to understand and deal effectively with organizational, societal and consumer issues. Under the right ethical conditions, surveys and interviews create a "safe haven" in which unconstrained speech is facilitated.

Privacy. People have as much right not to speak as to speak. Our founding fathers were sensitive to the right to be left alone, since they experienced oppression from a distant government and many early settlers came to America to escape religious persecution. Although the word "privacy" is not explicitly mentioned in our Constitution’s Bill of Rights (i.e., the first ten amendments), protection of the right to privacy has a long history in the First, Third, Fourth, Fifth, Ninth and Fourteenth Amendments. In the "penumbra" of the First Amendment is the right to privacy of association and belief. The Third Amendment prevents the quartering of soldiers in private homes. The Fourth Amendment protects against unreasonable search. The Fifth Amendment protects a person from being "a witness against himself." The Ninth Amendment, pertaining to other unenumerated rights being retained by the people, is the most frequently mentioned amendment protecting privacy interests. The Fourteenth Amendment prevents the states from denying constitutionally protected rights. Various state constitutions also have privacy provisions.

Privacy interests are also protected in American common law. Privacy was first championed by Warren and Brandeis (1890) in a famous Harvard Law Review article before Justice Brandeis served on the Supreme Court. Brandeis expanded on this position in a 1928 dissenting opinion by arguing that the makers of our Constitution:

The views expressed are those of the author. They do not necessarily reflect the views of the IMTA, the Armstrong Laboratory, the US Air Force, or the United States Government.

"sought to protect Americans in their beliefs, their thought, their emotions and their sensations. They conferred, as against the Government, the right to be let alone--the most comprehensive of rights and the right most valued by civilized man. To protect that right, every unjustifiable intrusion by the Government upon the privacy of the individual, whatever the means employed, must be deemed a violation of the Fourth Amendment" (Alderman & Kennedy, 1991, p.136).

Almost a half century later, the Supreme Court adopted this interpretation of the Fourth Amendment. This amendment was interpreted to protect people, not just places, from intrusion and established that an individual’s "reasonable expectation of privacy" should be respected.

Ethical codes guiding professional behavior also provide ethical parameters for conducting surveys or interviews. Guidelines across different professionals are usually similar. The predominant messages are clear, as relevant themes from the American Psychological Association’s (APA’s) Ethics Code (APA, 1992) makes clear: respect peoples’ rights and dignity, help rather than harm or harass your clients, maintain confidentiality and minimize intrusions on privacy, disclose confidential information only as mandated by law and explain any constraints on confidentiality to clients, disclose confidential data only with your client’s consent and maintain the confidentiality of databases, minimize invasiveness of persons or their milieu, and articulate your objections and behave in accordance with the code if laws or organizational expectations conflict with professional ethical practices. These professional ethical guidelines are based in part on two key legal and ethical concepts: privilege and informed consent. Privilege refers to the confidentiality of communications between certain professionals (lawyers, doctors, clerics, mental health professionals) and their clients and the right of testimonial privilege in courts of law. Informed consent refers to the necessity of providing clients with an understanding--in clear language--that participation is voluntary, informing them of the anticipated consequences of participation or nonparticipation and the factors which could reasonably influence their decision, protecting prospective participants from harm if they decline to participate and explaining other aspects about which they may inquire. It also refers to the need to obtain consent from the person involved before divulging confidential information.

Like physicians and lawyers, survey and interview professionals protect individual data from disclosure. They do so by protecting the identity of individual participants and by sharing aggregated (grouped) data. However, such professionals do not have automatic testimonial privilege. Subpoenas can be used to force disclosure of confidential research information. In 1988 Congress provided some relief in a law allowing researchers to seek a certificate of confidentiality to protect their data from a subpoena (Bradburn & Sudman, 1988; Sieber, 1992). As Bradburn and Sudman (1988) and Fowler (1993) indicate, researchers have sent their data abroad to prevent breaches of confidentiality in United States courts. The APA encourages members to familiarize themselves with steps to protect data from disclosure short of noncompliance with the law and has prepared a document on strategies for coping with subpoenas or compelled testimony (APA, 1995, 1996).

Ensure Voluntary Participation. The right to privacy and informed consent usually give people the right to refuse to be surveyed or interviewed, or to not answer specific content. Bradburn and Sudman (1988) state that all professional civilian survey organizations recognize this right. The author strongly recommends that voluntary participation be ensured. In the military, however, some surveys are still mandatory. Switching from mandatory to voluntary participation is not only the ethical thing to do from a rights perspective--it is the smart thing to do. You may decrease response rate somewhat but you are almost certain to get more accurate information, especially if you are asking about threatening or very intrusive topics. Of course, you must also maintain confidentiality.

Maintain Confidentiality and Non-Attribution. As Seiber (1992) and Salant and Dillman (1994) point out, there is a distinction between anonymity and confidentiality. Anonymity is best, but cannot always be promised, and may not be believed. There is a subtle "can not" versus "will not" distinction concerning the identity of respondents. Interviews are almost always not anonymous since they are typically face to face. If a survey is anonymous, no information is collected which could link data to individual respondents. If it is confidential, researchers can link data back to individual respondents or small groups, but agree not to. The linkage could be direct via an assigned code, a social security number or due to face-to-face interaction. It could be indirect via demographic information. Since some people don’t even trust anonymous surveys, it has been the author’s approach to ignore the anonymity/confidentiality distinction to avoid arguing whether or not respondents can be identified. Rather, he stresses confidentiality and promises that individual data, except for verbatim comments (from which individual identifying information has been removed), will not be released.

Survey and interview professionals have an obligation not to harm respondents or diminish their dignity. Improperly protected data could do both. In the wrong hands, survey or interview data, especially data expressing proscribed views or behaviors, would be damaging if linked to individuals. Researchers using surveys and interviews determine the incidence of attitudes or behaviors in groups. They examine aggregate rather than individual data. This information is used to gain greater understanding of issues, to improve policies and practices and to help people rather than harm them. If you know of institutional or legal constraints on confidentiality (e.g., company or court directives forcing disclosure), mention these to participants so they can make an informed decision about participation.

To gain cooperation, explain in a cover letter, or in introductory comments, how survey or interview data will be used, the benefits to respondents (e.g., being consulted, positive change) and how privacy will be safeguarded. Stress that data are shared with management or other clients in aggregate form. In an organizational survey, indicate that individual responses will not be shared with supervisors or anyone else in the chain of command, except individual comments which may be transcribed verbatim. Explain that even when individual comments are shared (e.g., in verbatim quotes), by-name references are removed except when the reference is obvious, such as to a commander.

Avoid asking for name or social security number. If such information must be collected (e.g., to conduct longitudinal research or match with other data files), assign random codes and substitute them for identifying information in your data file. Share the correspondence between people and codes with as few people on the survey team as possible. Limit access to your raw data. Stress with team members that individual responses and by-name comment data must not be divulged. If clients pressure you to reveal the responses or statements of specific individuals, refuse such requests.

Limit Objectionable or Very Intrusive Content. The author recommends that you limit, but not necessarily eliminate, potentially objectionable, threatening, or very intrusive questioning. Such content may be legitimate since is it important to study controversial issues to deal with them effectively. Arvey and Sackett (1993) caution that requesting information which invades privacy will be perceived as less fair in a selection situation (e.g., on an application form) than information which is not so invasive. If such items are included, emphasize your commitment to confidentiality and do not betray that pledge. Do what you can to develop trust. You can explain in an introductory statement that your intent is not to judge the correctness of expressed attitudes or the morality/legality of reported behaviors. Explain that you are trying to determine the prevalence of attitudes or behaviors and have no interest in linking them to individuals for punitive purposes. If participation is voluntary, as recommended, respondents can skip objectionable items or refuse to answer questions they consider too intrusive.

Avoid Illegal Content. If a legitimate need exists to include very intrusive content, researchers can do so and encourage honest responses by adhering to ethical principles. Illegal items or questions are a different story--such content should be avoided. These are discussed in two contexts: on an employment or other evaluation application or interview and on any survey when job irrelevant organizational affiliations are requested.

Employment or Other Evaluation Situation. Examples of items/questions or behaviors that may be illegal or constrained (depending on jurisdiction) on an evaluation application or in an evaluation interview are:

age, ancestry, arrest record, birth control usage, birthplace, citizenship in a specific country (other than United States; naturalization papers; parents’ citizenship), disability not related to work, ethnic group, family, spouse, children, relatives (specific information about family members or relatives), gender, native language or how a foreign language was learned, marital status, mental illness or treatment by a mental health professional, military experience outside the United States, miscellaneous: anything not related to work or employment eligibility, name (original or maiden), national origin, organizational membership (not considered relevant by the applicant to job performance; or related to race, religion, national origin, etc.), pregnancy or plans to have children, physical limitations not relevant to work, race, religion or religious beliefs, sexual orientation views or behaviors, sexual request.

It is also sometimes unlawful to ask questions that would indirectly reveal these facts. In some instances, sensitive questions are unlawful if asked directly. For example, "What is your age" is prohibited. However, it is all right to inquire more broadly, for instance, if a person is "between 18 and 65 years of age." Likewise, you cannot ask about an arrest record, but you can ask about convictions if there is legitimate "business necessity" involved. However a conviction is not usually grounds for being denied employment, unless directly tied to the job or if it would put an employer at considerable risk (Medley, 1984; Smart, 1983).

Organizational Membership. Asking for information about job irrelevant organizational memberships may be illegal on job applications. Such questioning may also be unlawful on constitutional grounds. Bond (1994) wrote about two cases demonstrating the legal precedent prohibiting forced disclosure of membership in private groups. These cases involved the First Amendment rights of free speech and assembly. They also pertain to privacy rights (see Shattuck, 1977, Chapter 2, for a discussion of the privacy of association and belief and several more case examples). Both cases involved groups protecting their membership lists from disclosure to government entities. The first case involved the National Association for the Advancement of Colored People (NAACP) versus the State of Alabama in 1958. The second case occurred in 1994 and involved the Texas Knights of the Ku Klux Klan versus the Texas Commission on Human Rights. Organizational membership lists could be amassed from survey or interview data. Given the legal precedents mentioned and the historic and recent polarization of American society (see Bari, 1994; Anderson, 1995) the author recommends that requests for specific information about job-irrelevant organizational affiliations be avoided.

Ethical implications have been discussed from the perspective of the obligations of survey or interview data gatherers. Focusing now on prospective participants, the author contends they have the right to decline to participate by not responding to the survey or interview or to specific items or questions, if they believe any of the following conditions are true:

• The survey or interview is not voluntary: respondents are told they must respond and they desire not to, or other forms of coercion are used, such as the threat of punishment.
• Insufficient information is given upon which participants could come to a decision based on informed consent.
• The survey or interview is requested as a part of one’s job, but some or all content is not job relevant, including questions about job irrelevant organizational memberships or privileged information the respondent desires not to share.
• Privacy may not be protected: the survey is not anonymous, confidentiality may be breached, analysis and reporting of non-narrative data may occur at the level of the individual respondent or at a level of disaggregation which might permit disclosure of identity, or personal identifying information may not be excised from narrative data, except where the identity of a prominent individual (e.g., a CEO or Commander) would be difficult to conceal.
• The requested information may be used to discriminate unfairly against respondents, harass them or harm them in other ways.
• The items or questions are needlessly intrusive, or they are too offensive.

American citizens and employees have become more aware and protective recently of their rights to free speech, free association and privacy. Former President Kennedy’s daughter has co-authored two popular books on these and other "Bill or Rights" issues (Alderman and Kennedy, 1991, 1995). Privacy-rights litigation has increased dramatically in recent years (see Alderman and Kennedy, 1995, pp. 273-320, Privacy in the Workplace). Lawyers have invoked common law principles and federal and state constitutional provisions to successfully represent plaintiffs. Between 1985 and mid-1987 there were nearly 100 privacy verdicts against employers and the average workplace-privacy jury award during this time was $316,000 (Hendricks, Hayden and Novik, 1990; Nachman & Ryan, 1994). Alderman and Kennedy (1995, p. 305) report that "Government workers are protected by the federal and state constitutions and certain statutes which often enable them to be more successful than their private-sector counterparts when suing their employer." Thus, safeguarding privacy is not only an ethical necessity; it also makes good legal and business sense. As Beaney (1990) argues, legal privacy rights are particularly important as individuals and groups experience increasing and more varied assaults on their inner life and thoughts. He includes the wide spread use of surveys (and by extension, interviews) in government and industry as one form of such assault. He suggests that if Congress and government agencies develop a heightened sensitivity to potential threats to dignity and privacy, the need for judicial intervention will decline.

Alderman, E., & Kennedy, C. (1991). In our defense, the bill of rights in action. New York: William Morrow and Company

Alderman, E., & Kennedy, C. (1995). The right to privacy. New York: Knopf.

American Psychological Association. (1992). Ethical principles of psychologists and code of conduct. American Psychologist, 47, 1597-1611.

American Psychological Association (1995). Strategies for coping with subpoenas or compelled testimony for test data. Washington DC: Office of Legal Affairs.

American Psychological Association (1996). Statement on the Disclosure of Test Data. American Psychologist, 51, 644-648.

Anderson, T. H. (1995). The movement and the sixties. New York: Oxford.

Arvey, R. D., & Sackett, P. R. (1993). Fairness in selection: Current developments and perspectives. In N. Schmitt & W. C. Borman (Eds.). Personnel Selection in Organizations. San Francisco: Jossey-Bass.

Bari, J. (1994). Timber wars. Monroe ME: Common Courage.

Beaney, W. M. (1990). The Griswold case and the expanding right to privacy. In P. L. Murphy (Eds.). The right to privacy and the ninth amendment. (Vol. I), pp. 1-17. New York: Garland.

Bond, J. C. (Spring 1994). First amendment shock: An unusual civil liberties story. Civil Liberties, 380, p. 6.

Bradburn, N. M., & Sudman, S. (1988). Polls and surveys: Understanding what they tell us. San Francisco: Jossey-Bass.

Fowler, F. J. (1993). Survey research methods. (2nd ed.) Newbury Park, CA: Sage.

Hendricks, E., Hayden, T., & Novik, J. D. (1990). Your right to privacy: A basic guide to legal rights in an information society. Carbondale and Edwardsville IL: Southern Illinois University Press.

Medley, A. H. (1984). Sweaty palms: The neglected art of being interviewed. Berkeley CA: Ten Speed.

Nachman, S. A., & Ryan, J. S. (1994). Company policy manual: Special report, privacy in the workplace. New York: Panel Publishers (Aspen).

Salant, P., & Dillman, D. A. (1994). How to conduct your own survey. New York: John Wiley & Sons.

Shattuck, J. H. F. (1977). Rights of privacy. Lincolnwood, IL: National Textbook Company.

Sieber, J. (1992). Planning ethically responsible research: A guide for students and internal review boards. Newbury Park, CA: Sage.

Smart, B. D. (1983). Selection interviewing: A management psychologist’s recommended approach. New York: John Wiley & Sons.

Warren, S. D., & Brandeis, L. D. (1890). The right to privacy. Harvard Law Review, (1890-1891), IV, pp. 193-220.

Watson, T. W. (1996a). Guidelines for conducting interviews. (AL/HR TR-1997-0043) Brooks AFB, TX: Cognition and Performance Division, Human Resources Directorate, Armstrong Laboratory.

Watson, T. W. (1996b). Guidelines for conducting surveys. (AL/HR TR-1997-0044) Brooks AFB, TX: Cognition and Performance Division, Human Resources Directorate, Armstrong Laboratory.