Skip to main content
Graduate School of Library and Information Studies

Research Ethics

Research ethics, like professional and academic ethics, help balance often competing rights and interests. In the case of research ethics, one important conflict is between our human need for knowledge on the one hand, and the needs and rights of potential research subjects on the other.

The most urgent issue for LIS researchers is the protection of human subjects. Before undertaking any research with human subjects, URI investigators must be certified by the Research Office and must have their research proposals reviewed and approved by the Institutional Review Board. (See http://www.uri.edu/research/tro/about/IRB/index.html.)

Not all GSLIS students will be directly involved in original research with human subjects, but all will benefit by it, because the knowledge base of our profession incorporates the findings of research studies. Therefore, incoming GSLIS students will be required to demonstrate knowledge of human subject protections (see below).

"Research" here means systematic inquiry for the purpose of discovering new information or testing the truth of information that has already been disseminated, thus adding to the sum of human knowledge. This is the specialized meaning of "research" that you will use in LSC 557 (Research and Evaluation), and not the looser, more general "looking things up in the library or online" meaning that we often attach to the word in everyday speech.

Typical LIS research methods include:

  • Survey

  • Bibliometrics

  • Interview

  • Experiment

  • Participant observation

Some of our methods, like survey, bibliometrics, and experiment, are quantitative, and have a basis in logical positivist philosophy, while others, like interview and participant observation, are qualitative, and are based in newer philosophies, such as constructivism. Some, like survey, allow the research "subjects" to be anonymous even to the researcher; in others, like interview, the researcher must meet the "subject" face to face, and preservation of confidentiality becomes more complex. In still others, such as participant observation, the "subjects" may not even know that they are being studied. Both because of their philosophical underpinnings and because of the relationships in which they place researchers and subjects, different research methods raise different ethical issues.

University of Rhode Island requirements

Research can be unethical in several ways -- data can be falsified or simply invented, for instance. A "psychologist, the eminent Sir Cyril Burt, was posthumously accused of inventing data in his studies of intelligence in monozygotic twins reared separately" (Carlin, 2003). Did Burt invent data, twin sets, and even research assistants? There have been attempts to restore his reputation, but for now, his memory stands as a warning against fudging the facts. The URI University Manual explicitly forbids this, in section 8.27: http://www.uri.edu/facsen/8.20-8.27.html.)

But within a university, the phrase "research ethics" most often refers to the ethical treatment of human and animal subjects. Universities establish Institutional Review Boards to monitor compliance with federal regulations designed to protect human subjects from harmful or undignified treatment, and any faculty or student research involving human subjects must be submitted for IRB approval before it can be implemented. All principal investigators are also required to complete training on the Responsible Conduct of Research; completion of this training must be certified by the URI Research Office.

For a concise summary of core ethical issues in the treatment of human subjects, see  Trochim's "Ethics in Research." He looks briefly at how abuses in World War II concentration camps and the Tuskegee Syphilis Study led to the formulation of protective principles like voluntary participation, informed consent, and measures to reduce risk -- and how more recent medical studies have raised questions about the right to service. These concepts are natural corollaries of the principles of respect for persons, beneficence, and justice, laid out in the Belmont Report -- a foundational document for ethical research (you can see it at the website of the Office of Human Subjects Research, which is part of the National Institute of Health). Excerpts:

  • "Respect for persons incorporates at least two ethical convictions: first, that individuals should be treated as autonomous agents, and second, that persons with diminished autonomy are entitled to protection. The principle of respect for persons thus divides into two separate moral requirements: the requirement to acknowledge autonomy and the requirement to protect those with diminished autonomy." Therefore, researchers should be open about their studies, giving recruits enough information to make an intelligent choice about whether to participate. Researchers should also be careful not to exploit power relationships that would make it difficult for recruits to refuse consent -- participation should be completely voluntary. 

  • "Two general rules have been formulated as complementary expressions of beneficent actions . . . : (1) do not harm and (2) maximize possible benefits and minimize possible harms."

  • Justice requires, among other things, that researchers should use as subjects people who are likely to benefit from the research -- not people who are just easy to round up. "For example, the selection of research subjects needs to be scrutinized in order to determine whether some classes (e.g., welfare patients, particular racial and ethnic minorities, or persons confined to institutions) are being systematically selected simply because of their easy availability, their compromised position, or their manipulability, rather than for reasons directly related to the problem being studied. Finally, whenever research supported by public funds leads to the development of therapeutic devices and procedures, justice demands both that these not provide advantages only to those who can afford them and that such research should not unduly involve persons from groups unlikely to be among the beneficiaries of subsequent applications of the research."

Ethics in Social Science Research

In both the Belmont Report and Trochim's summary, you'll notice that many examples of shaky research ethics are drawn from biomedical research, and principles were developed to prevent the scandalous use of prisoners and other vulnerable populations as guinea pigs. All this may seem irrelevant to research in Library and Information Science (LIS), which is more closely related to psychology, sociology, and education than to medicine. Research in the social sciences generally poses little risk of physical harm, but it has sometimes been psychologically dangerous. Consider how the Belmont principles of respect for persons, beneficence, and justice apply to some of these famous cases, summarized by Carlin (2003):

  • Milgram's (1963) study of "obedience and responses to authority" involved deceit. Subjects were recruited for a study of "memory and learning," and led to believe that they were chosen by lot to be "learners" or "teachers." In fact, the selection process was rigged: the "learners" (or victims) were members of the research team, and the subjects were always "teachers." They were told to administer shocks of increasing magnitude to the "learners" for each wrong answer -- and they continued to do so, even after the voltage reached 300,  the unseen learners pounded on the walls, and answers stopped, suggesting the learners might be unconscious or dead. In reality, the "learners" received no shocks. The real potential for harm in this study was to the "teachers," who had not given their informed consent (since they were misinformed about the purpose of the study) and who were confronted with the realization that they, like Hitler's soldiers, might just follow harmful and immoral orders. Related links:
    An abridged version of Milgram's "The Perils of Obedience": http://www.age-of-the-sage.org/psychology/milgram_perils_authority_1974.html
    Discussion of Milgram's work by Thomas Blass: http://www.stanleymilgram.com/

  • At Stanford, Zimbardo et al. (1973) recruited a group of students for a study of why prisoners and guards behave as they do. The students, randomly assigned to enact the roles of "prisoners" or "warders," were supposed to simulate a prison environment for two weeks -- but the researchers called a halt after six days, as "the warders became increasingly aggressive, whereas the prisoners became increasingly submissive." Related links:
    The Homepage of Professor Philip G. Zimbardo: http://www.zimbardo.com
    The Stanford Prison Experiment: http://www.prisonexp.org/ 
    The role of Zimbardo's ex-wife, psychologist Christina Maslach, in stopping the experiment: http://news-service.stanford.edu/pr/97/970108prisonexp.html

  • Rosenhan (1973) deliberately misled the doctors and staffs of psychiatric hospitals when he had "pseudopatients" present themselves with simulated indicators of mental illness. After admission, the pseudopatients acted normal, but their sanity was not noticed by the psychiatrists. In a follow-up, Rosenhan informed a psychiatric establishment that more pseudopatients might be sent. They were not; but the hospital suspected 41 of the next 193 applicants for admission of being sane. Related sites: 
    David Rosenhan, "On Being Sane in Insane Places": http://psychrights.org/articles/rosenham.htm
    Summary of experiment: http://www.holah.karoo.net/rosenhan.htm

Ethics in LIS Research

So what does all this have to do with research in library and information science?

First, we have a responsibility to design our research carefully, using reliable methods to gather and analyze data, so that we can have reasonable confidence in the validity of our findings. Carlin (2003) suggests further that, as an interdisciplinary field, we must treat imported theories, concepts, and methods ethically, maintaining a level of intellectual rigor that respects their origins. That would be consistent with our careful use of cited material, although it goes well beyond mere quotation: it means we have a responsibility to understand the methods we use, and not just apply them mechanically.

Second, we have a responsibility to treat our human subjects with respect, beneficence, and justice. LIS research is unlikely to pose physical danger or risk grave psychological harm to participants.

The gravest risks of research in our field to human subjects, however, are to the dignity of participants, not their physical well-being. Therefore, in designing studies that respect the rights of human subjects, LIS researchers must pay special attention to the principles of anonymity and confidentiality: data are handled carefully, so that information about individual people and even institutions cannot be used in ways that would harm or embarrass them. LIS researchers must also concern themselves with the ownership of information and ideas, making careful attribution when drawing from the intellectual capital of others.

So what if an anonymous participant in your study gives you a brilliant idea? How do you preserve confidentiality while citing your source? Can you find an ethical solution to this dilemma? Remember, you need to protect both the integrity of the research and the rights and dignity of your informant. Is the solution the same for all research methods?  

If you are doing a survey, your subjects may be anonymous even to you. If you are using a qualitative research method like in-depth interview or participant observation, you will spend significant amounts of time with your subjects; you can ask probe questions if their responses aren't clear, and if appropriate you can even allow them to respond to drafts of your report. This kind of research has parallels in disciplines like linguistics, and the ethical considerations are similar. For instance, see a lesson on ethical considerations in orthographic transcription at Lancaster University's site (Baker, et al.): "Ethics is partly a matter of fairness to people whose language you are studying; not exposing them to risk or ridicule, and not abusing their goodwill in participating in your study in the first place."

One research paradigm in LIS is the unobtrusive evaluation of reference service. Researchers select typical reference questions to which the answers are known in advance, and send trained surrogates (often LIS students) to ask the questions, posing as regular library users. These studies have found that on average only 55% or 60% of ready reference questions asked at public and academic libraries get complete, correct answers -- but there are ways to increase the score. Some have objected to the studies as unethical, because they could affect the employment of working librarians who are unaware that they are being observed. What are the ethical issues here, and how should they be handled? 

Do concerns about ethics sometimes prevent us from doing potentially useful research? Under-researched areas in our field may include discipline, or the handling of problem patrons. Some patrons (including some teenagers and some of the mentally ill) become "problems" by disrupting the smooth routine of library service; our obligations to respect them as individuals and to meet their information needs come into conflict with our responsibilities to other patrons, who may express discomfort or annoyance at their behavior. Most of the literature we have on these issues is anecdotal, based on practitioner experience and opinions. There is some legal research, analyzing case law. But it is harder to find research-based answers to questions about, for instance, what librarian behaviors work best to ensure a harmonious environment that meets the needs of "problem" and "non-problem" patrons alike. Some possible barriers to the production of such research:

  • IRB review is more stringent for any study involving mental patients, minors, or other vulnerable populations. To survey or study minors, the researcher needs informed consent from both parent and child -- conceivably easier to get in the case of teens who are on speaking terms with their parents, a condition that might screen out a disproportionate number of the "problem" patrons. Thus, the process used to protect research subjects from harm might bias the outcome of the research, limiting its usefulness.
  • To ascertain what librarian behaviors help create a positive, disciplined, environment, a researcher might want to observe both disciplined and undisciplined libraries -- and less effective disciplinarians might feel threatened by the study. To avoid conflict with practitioner colleagues, the researcher would want to maintain a delicate balance between tact and accuracy in collecting and reporting results. 

Such under-researched areas are among the things librarians often wish they'd been taught more about in library school. To design research strategies that are both ethical and effective can be difficult, but worthwhile. Do new technologies offer a solution?

With the advent of electronic communications technologies and especially the Internet, new observational techniques have become possible. For instance, a researcher who wanted to know more about the information needs of fibromyalgia sufferers or parents of children with ADHD could join online discussion lists and do content analysis of the messages. But is this ethical?

Ethical research in cyberspace

Jones (1994) offers a thoughtful exploration of how guidelines developed for biomedical research can be adapted to the kind of behavioral research we're more likely to do. He discusses the importance and difficulties of public/private distinctions (can individuals reasonably expect privacy on the Internet?) and of informed consent (how can social scientists provide full disclosure and obtain voluntary consent from all individuals in a constantly changing virtual environment? will the informed consent process compromise the research by changing the very behavior it seeks to observe?). He suggests possible solutions, based partly on a realistic appraisal of what "undisclosed risks to subjects that are more than minimal" could arise from social science research.

More on the ethics of research in cyberspace:

More on the ethics of research with human subjects: