Commodity Flow Survey Pretesting Plan

The Census Bureau plans to conduct additional research under the generic clearance for questionnaire pretesting research (OMB number 0607-0725). We plan to conduct research, using respondent debriefings on the response process of this survey for the 2012 data collection cycle. The Commodity Flow Survey (CFS) is a quarterly panel survey conducted during the Economic Census year. Respondents to the survey provide information on select shipments that occur within the designated quarter.

In the 2012 survey, respondents had the option to submit their responses either through the Web survey or on the paper form. For the current research, CFS staff is interested in finding out more information about the overall response process to the survey, in particular, for respondents who chose the Web. This research will include the analysis of production data from all four quarters of the 2012 CFS to identify problems with the response process, as well as brief phone interviews with key groups of respondents to the 2012 survey.

CFS has outlined the following research questions they wish to answer in this evaluation:

1. Did Web respondents encounter any problems with the Web interface?

Respondents had the option to upload a spreadsheet containing their shipment record information. Production data indicates that respondents may have had trouble uploading the spreadsheets. To identify those problems, we will conduct 6-8 brief phone interviews with respondents who had difficulty uploading spreadsheets that focus on their experience with the upload process.

2. Why did respondents choose a particular method for entering their shipments information?

The Web interface offered respondents two options for entering their shipments: entering them screen-by-screen with the interface or by downloading a spreadsheet, filling it out, and uploading it back into the Web interface. We will conduct interviews with approximately 6-8 respondents and ask them about their choice for entering the shipment information.

3. Why did respondents switch response modes?

We will interview a small group of 6-8 mode –switching respondents to find out why they chose to switch response modes (from Web to paper, or from paper to Web) during the reporting quarters. In particular, we are interested in finding out if there were problems with the Web interface that led respondents to switch to the paper mode.

4. Why did some respondents send their shipment records to the National Processing Center?

A subset of respondents did not fill out either the paper form or use the Web. Instead, they sent shipment records to the National Processing Center. We will talk to approximately 6-8 of these respondents to find out more about why those chose this response method.

5. How do companies with multiple establishments handle the response process?

Currently, some businesses with multiple locations receive separate survey forms for each of their locations, requiring them to log in and/or fill out multiple forms. We would like to know more about how to simplify this process. We will conduct interviews with approximately 6-8 of these multi-establishments businesses to find out more about how to simply their response process.

From July through September of 2013, staff from the Response Improvement Research Staff (RIRS) within the Census Bureau’s Office of Economic Planning and Innovation will conduct a maximum of 40 brief phone interviews across the U.S. with these five groups of participants in the 2012 CFS. We will use the results from the analysis of production data to identify the respondents who fall into the five groups outlined above and recruit from those lists. Respondents in each group will answer only a few focused questions meant to shed light on their experiences with the response process. A copy of the proposed questions for each of these groups of respondents is enclosed.

Before beginning the interviews, we will inform participants that their response is voluntary. We also will inform the participants that the information they provide is confidential. We will not be providing respondents with monetary incentives to participate in this study.

We estimate that it will be necessary to interview only one participant at each business. We estimate the length of the interviews will average 15 minutes. Therefore, the maximum estimated burden for this research is approximately 10 hours (15 minutes x 40 interviews = 10 hours).

The contact person for questions regarding this research is:

Jennifer Beck

Response Improvement Research Staff

U.S. Census Bureau

Washington, D.C. 20233

(301) 763-1736

[email protected]