CAMEO | Ask A Librarian |  Site Search | Home

Information Ethics Problem of the Month is a feature on the Carnegie Mellon University Libraries Web site. Librarians, computing professionals, faculty, and students will be invited to contribute short essays on questions of special interest to the university community. If you would like to comment on an essay, suggest a topic, or volunteer to write a guest column, please contact the feature editor, Jean Alexander, jeana@andrew.cmu.edu.

The Ethics of Data Collection and Usage
by
Paul Neuhaus
Social Sciences Librarian
Hunt Library
neuhaus@andrew.cmu.edu

Two issues that have received increasing attention over the last century are the ethical treatment of human subjects in research and the importance of protecting the privacy and confidentiality of individuals. These two concerns are often conjoined in research endeavors. Revelations about Nazi medical experiments spurred efforts to ensure the ethical treatment of research subjects. The Nuremberg Code of 1947 laid down ten stipulations that researchers must follow to guarantee that human subjects are treated ethically. The first clause states that the "voluntary consent of the human subject is absolutely essential." A copy of the Nuremberg Code is available at http://ohrp.osophs.dhhs.gov/irb/irb_appendices.htm#j5. Secondly, concerns about privacy and confidentiality intensified as data collection and retention efforts increased along with the development of increasingly powerful computers to analyze the data.

Researchers utilize a variety of data gathering and analysis methodologies. These range from experiments in which each research subject gives written consent to unobtrusive methodologies in which subjects may never know what data is being collected about them or who has access to the data. Whatever the methodology employed, researchers must be certain that their data collection and usage efforts meet ethical standards. Thomas Peters notes that data collection can be unethical in three broad ways. The data itself may be private or confidential so that the mere collection of the data represents an invasion of privacy. Secondly, the method used to collect the data may be unethical. Lastly, use of the data may be unethical. Peters argues that people focus too much upon the ethics of data usage and not enough upon the ethics of data collection.[1]

Fortunately, Carnegie Mellon researchers have various sources for guidance on these issues. The Office for Human Research Protections (http://ohrp.osophs.dhhs.gov/index.html) offers policy guidance, compliance guidelines, and sponsors a series of workshops on the ethics of research. Secondly, the Federal Policy for the Protection of Human Subjects addresses issues of informed consent. The relevant regulations are found in Title 45, Section 46 of the Code of Federal Regulations (45 CFR 46). The regulations can be viewed at the National Archives and Records Administration web site, http://www.access.gpo.gov/nara/cfr/cfr-table-search.html.

On the Carnegie Mellon University level, an Institutional Review Board (IRB) works with researchers to guarantee that human subjects receive ethical treatment. Most academic institutions that conduct research with human subjects have IRBs. These Boards are responsible for reviewing specific research proposals involving human subjects prior to the initiation of the research procedure. Not all IRBs interpret the regulations in the same way. There are inconsistencies between campuses on how the regulations are applied and how much time and effort must be devoted to obtaining approval. If you are working with researchers at other institutions or gathering data at more than one institution, you may need to obtain IRB approval at multiple institutions. Information and forms for Carnegie Mellon researchers are available at http://www.andrew.cmu.edu/~alo/IRBfinal/index.html.

Not all researchers gather their own data. Sometimes researchers utilize data collected by other researchers for the purposes of secondary analysis. Researchers cannot assume that datasets they employ for secondary analysis are free from ethical or confidentiality considerations. For example, one issue that is becoming more pronounced in secondary analysis is whether or not researchers can breach confidentiality when combining multiple datasets. Individual datasets may protect confidentiality absolutely but a researcher who combines tax, health, and social security records for senior citizens in Pittsburgh increases her/his chances of identifying individuals.

Members of the Carnegie Mellon IRB examined this issue of secondary analysis last year. They determined that certain organizations have sufficient privacy and confidentiality safeguards in place so that Carnegie Mellon researchers can utilize the datasets without prior IRB approval. These exemptions are:

• Inter-University Consortium for Political and Social Research (ICPSR)
• U.S. Bureau of the Census
• National Center for Health Statistics
• National Center for Education Statistics
• National Election Studies

Researchers using other data for secondary analysis should consult Carnegie Mellon's Institutional Review Board. Additional information concerning privacy issues and datasets is available from ICPSR (http://www.icpsr.umich.edu/irb/index.html) and the Carnegie Mellon Privacy Technology Center (http://privacy.cs.cmu.edu/center/index.html)

Not all research poses the same level of ethical challenges. The primary question a researcher must ask is "How might research subjects be harmed by this research or by the disclosure of individual findings?" Researchers using human subjects in medical research may face a much stricter ethical challenge than librarians using library data or business researchers examining data on consumers' preferences for toothpaste. Because data collection can be just as unethical as data usage, it is imperative that researchers go through the IRB process. Review by a third party helps to ensure that research is conducted in an ethical manner. This protects both the human subject and the researcher.