• Poster

Trust and Completion of Surveys in the US Navy

Citation

Olmsted, M. (2004, May). Trust and Completion of Surveys in the US Navy. Presented at American Association for Public Opinion Research 59th Annual Conference, Phoenix, AZ.

Abstract

Over the past several years, the U.S. Navy has been engaged in the process of transitioning much of the survey research enterprise to the Internet. This change of format has occurred as the everyday use of the Internet has increased in the life of the average Sailor. At the present, over 90 percent of Sailors have regular access to the Internet. A number of major initiatives such as the Navy/Marine Corps Intranet and the Information Technology for the 21st Century (IT-21) seek to provide universal access to for all personnel over the next few years.

The Institute for Organizational Assessment at the Navy Personnel Research, Studies, and Technology Department (NPRST) is actively pursuing a variety of research on the validity and reliability of large-scale Internet-based personnel surveys. When transitioning surveys to the Internet, changes introduced by the new method may cause respondents to act differently than they would to traditional paper surveys. An area of particular interest is how these changes affect the trust of respondents in the organization. A number of previously published studies have shown potential respondents tend to not participate when they have trust concerns about how an organization will handle their data, the safety of their data, or whether they believe that data will be used effectively to affect change.

To evaluate this issue, this presentation will compare the findings from two Navy surveys that have fielded parallel paper and Internet versions. These surveys include two editions of the Navy-wide Personnel Survey which both included Internet survey options; one edition included this option as an experiment, the other as a choice response option. The presentation will cover issues such as response rate, completion patterns, response to personal identifiers, and differences in item/scale responses between the paper and Internet versions of the surveys.