Impact of Internet-only Reporting on Response in the Business R&D and Innovation Survey

• Activity Number: 536
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Government Statistics Section
• Author(s): Richard Hough* and Brandon Shackelford
• Companies: U.S. Census Bureau and Twin Ravens Consulting
• Keywords: Company survey ; Survey modes ; R&D ; Internet reporting mode ; Response rates

Abstract

The 2013 Business R&D and Innovation Survey (BRDIS) included an experimental design to test the impact of removing the questionnaire booklet from the initial mail packages on response. The letter/experiment group received a letter with instructions for completing the survey online and a questionnaire booklet was not included in the mail package. The questionnaire/control group received a questionnaire booklet in the mail package that included instructions for reporting online consistent with past BRDIS cycles. At the end of the experiment, the weighted check-in rate for the letter/experiment group was 58.9% compared to 63.8% for the questionnaire/control group. The difference in check-in rates decreased from 4.9% to 2.3% after the 2nd nonresponse follow-up in which both groups were sent questionnaire booklets. Large companies(defined as having at least 500 domestic employees) exhibited higher weighted check-in rates in the letter/experiment group after all processing was complete. Companies reporting online reported significantly larger amounts of R&D expenditures (the key variable of interest to BRDIS) than those reporting on the questionnaire booklet.