U.S. Department of Commerce

Research Reports

You are here: Census.govSubjects A to ZResearch Reports Sorted by Year › Abstract of SM95/21
Skip top of page navigation

Evaluation Results from a Pilot Test of a Computerized Self-Administered Questionnaire (CSAQ) for the 1994 Industrial Research and Development (R&D) Survey

Elizabeth Sweet and Magdalena Ramos

KEY WORDS: Electronic Questionnaire


We performed this evaluation to determine whether economic respondents could quickly, easily and accurately provide data using a CSAQ. We conclude from our evaluation that, although both the Bureau of the Census and R&D Survey respondents consider CSAQ and electronic reporting the way of the future, the PEDRO/CASES CSAQ system used for the pilot test is not appropriate for implementation at the Census Bureau. We did receive user feedback and identification of improvements necessary for a CSAQ system to be a viable data collection tool and consequently, we recommend continued research into CSAQ authoring and communications alternatives. Our conclusion is based on an analysis of several factors. An analysis comparing the response rate of companies that received the CSAQ to companies that received the paper questionnaire found that the CSAQ companies had a significantly lower response rate. Some companies that received a CSAQ did not respond using a CSAQ because they either had problems with the CSAQ system or perceived that this type of reporting would be more difficult. This problem occurred even after the test universe was restricted to cases interested in CSAQ reporting and that according to responses to a screener questionnaire, appeared to have the hardware/software necessary to use the system we were testing. The CSAQ had edits built into the instrument and thus allowed respondents to correct their data prior to sending us the completed instrument. This is seen as an advantage for us (better data quality and less review) and as an advantage for the respondent (fewer follow-up calls concerning edit failures). We re-edited the data and found the CSAQ panel edit failure rate was significantly lower than the control panel rate. The cost of the mailout package for the CSAQ (postage, material, diskettes, and envelopes) was approximately nine times higher than the mailout cost of a paper questionnaire. In the future, there might be cost savings associated with a CSAQ instrument when compared to the mailout cost for surveys where boxes of questionnaires are mailed to companies that report for all of their establishments.

CITATION: 1995, "Evaluation Results from a Pilot Test of a Computerized Self- Administered Questionnaire (CSAQ) for the 1994 Industrial Research and Development (R&D) Survey." US Bureau of the Census. Economic Statistical Methods Report Series-9503.

Source: U.S. Census Bureau | Statistical Research Division | (301) 763-3215 (or chad.eric.russell@census.gov) |   Last Revised: October 08, 2010