ANN ARBOR, Mich.—Cognitive psychologist Frederick Conrad has been appointed director of the University of Michigan Program in Survey Methodology, based at the U-M Institute for Social Research (ISR). The Program is one of only three in the U.S. to provide graduate-level training in how to conduct scientifically sound polls and surveys.
“I am delighted that Fred Conrad will be heading this important graduate program,” says ISR Director James S. Jackson. “His energy and deep commitment to the highest standards of survey research are crucial to the continuing success and growth of this program. The field of survey research is going through major changes, from incorporating new types of data to using new types of media for collecting data, and Fred’s leadership in understanding the challenges and opportunities facing the next generation of survey researchers is invaluable.”
Conrad succeeds the Program’s founding director, James Lepkowski, a professor of biostatistics and public health who will now serve as associate director of the Program.
“I am very grateful to Jim for his steady and thoughtful leadership of the Program, and for his continuing involvement as associate director,” says Conrad. “He was instrumental in establishing the Program here at Michigan, where scientific survey research has a long and venerable history.
“I am also deeply honored to be heading the Program in Survey Methodology at ISR, which is known world-wide for its expertise in this area. Our program greatly benefits from the involvement of ISR faculty, many of whom are affiliated with Michigan’s social science departments, which are among the best in the world.”
Jackson notes that the program also benefits from the involvement of companies that help to sponsor academic programs through internships and other kinds of support.
Conrad will also serve as director of the Joint Program in Survey Methodology, a long-standing collaboration between U-M, Westat, and the University of Maryland.
Conrad’s appointment as director of both Programs marks an important step in their eventual consolidation. “I hope to advance the continued integration of these programs, by working to coordinate admissions and faculty recruitment,” says Conrad. “Eventually, our goal is to have a single program at two sites, offering degrees issued jointly by Michigan and Maryland.”
The Michigan Program in Survey Methodology was established in 2001, and offers certificates in survey methodology as well as master’s and Ph.D. degrees. Graduates of the program are employed by the U.S. federal statistical system, commercial survey and market research firms, the academic community, and non-profit organizations. “Our graduates are doing very well,” says Conrad. “Even when the economy is weak, survey researchers are in demand. I’m not aware of a single graduate who wants to work in the field who has not been able to find a satisfying position.”
Conrad’s recent research in survey methodology focuses on new data collection methods such as interactive web surveys and virtual interviewers, on interviewer-respondent interactions, and on the effectiveness of pretesting techniques. He has also recently conducted research on the usability of electronic voting systems, and in the role of public events in personal memory.
He is currently the co-principal investigator, with Michael Schober of the New School for Social Research, on a National Science Foundation grant investigating responses to surveys on mobile, multimodal devices. “We’re looking at the impact on survey data quality of collecting information using speech and SMS text modes on mobile devices, when either a human or automated interviewer asks the questions. We’re focusing on iPhones so that the user interface is the same for everyone,” he explains.
“One of the issues we’re examining is whether people are more or less willing to disclose confidential information in some of these conditions than others. It is well known that people are usually more willing to disclose confidential information if the interviewer is automated, but what if questions are asked via a text message so there are relatively few cues about whether the interviewer is a person or a computer? Another issue we’re examining is how these modes affect respondent ‘satisficing’ – the tendency to take shortcuts when answering rather than giving full thought to the answer. And, we are looking at the effect of allowing people to choose one of these modes: for example, will they choose a visual mode like SMS in a noisy environment?”
Conrad received a B.A. from Hampshire College in 1977 and a Ph.D. in cognitive psychology from the University of Chicago in 1986. He worked at Carnegie-Mellon University as a post-doctoral research associate in the Psychology Department then joined the Artificial Intelligence Research Group of Digital Equipment Corporation as principal software engineer. In 1991 he joined the U.S. Bureau of Labor Statistics as a senior research psychologist, and in 2002 he joined both ISR and the Joint Program in Survey Methodology at the University of Maryland as an associate research scientist. He is now research professor at both institutions.
Conrad is the author or co-author of scores of articles and book chapters on questionnaire design, survey interviewing methods, web surveys, and related topics. He is the co-editor of Envisioning the Survey Interview of the Future, and Intersections in Basic and Applied Memory Research, and the co-author of Voting Technology: The Not-So-Simple Act of Casting a Ballot.