Dealing with data

As an empirical science, psychology relies on data. We collect, analyze and report data—all in the service of drawing informed conclusions about our subject matter. We depend on data to provide the foundation of our discipline.

The activity of data collection in science is accompanied by a host of responsibilities and obligations. Ethical principles abound, especially relating to confidentiality, sharing and reporting. Practical matters need attention, such as the form in which data are saved or archived. Difficult questions must be resolved, including ones about data ownership and control.

To a large extent, data-related practices are governed by a discipline's culture. In the physical sciences, such as astronomy or oceanography, data are collected with shared instrumentation involving large groups of investigators supported by federal research grants. The data are shared among many scientists and typically produce numerous publications.

In the social sciences, such as economics, sociology or political science, coordinated data collection efforts are common. Scientists participate in the design of those activities, and base much of their own research on them. The data are often archived in publicly accessible facilities and available for anyone to scrutinize.

Psychology has its own unique data culture. Because our data often relate to individual human research participants, we spend a great deal of effort to protect the confidentiality and privacy of those participants. This is one of the reasons why psychologists are reluctant to place their data in publicly accessible archives or to share their data with anyone other than other psychologists who are working in closely related areas.

When federal grants support the collection of data, special rules apply as a condition of accepting the funds. For example, the National Science Foundation expects investigators to share with other researchers their data and other materials collected in the course of the work. NSF encourages the sharing of data, software and inventions, but also recognizes that exceptions can be made to protect the rights of individuals and subjects, the validity of results and other legitimate interests of investigators.

The NSF rules offer a relatively narrow set of data sharing obligations and clearly tilt toward protecting the interests and needs of investigators. Our own ethical principles also support the narrowest of data sharing obligations for psychologists. APA Ethical Standard 8.14 stipulates that psychologists do not withhold their data from other competent professionals who seek to verify substantive claims, placing the burden of expenses associated with sharing data on those who seek it.

Neither the NSF rules nor APA's ethical standards place a high priority on making data publicly accessible. A broader set of rules applies, however, in the Office of Management and Budget Circular A-110, which establishes requirements for federal grants and agreements. These are the rules that NIH applies to all of its grants, and forms the basis of such NIH requirements as a formal data sharing plan in all applications seeking $500,000 or more in direct costs.

Even the rules of OMB Circular A-110 allow a narrow data sharing model for psychology. Under the rules, research data must be made publicly available, excluding anything that could compromise privacy (including the identity of a particular person in a research study).

The data culture of psychology is one of limited sharing, and then only among a select few with demonstrated competence and legitimate need. The federal rules right now support this culture, and psychologists cling tightly to it.

Yet, the winds of cultural change are blowing. Increasingly, the government and the public demand greater access to the results of science—to our publications and to our data. A cultural shift for psychology is looming. We should begin now to prepare ourselves.

Resources relating to data sharing and archiving are available on APA's Web site at www.apa.org/science/data/homepage.html.