2007 Nonemployer Frame Cognitive Interviewing Plan

The U.S. Census Bureau plans to conduct additional research under the generic clearance for questionnaire pretesting research (OMB number 0607-0725). We will be conducting cognitive interviews and a field feasibility study with respondent debriefing to test a survey of corporations and partnerships from the 2007 nonemployer frame to validate their status as nonemployer companies. Currently, the Census Bureau does not conduct any survey solely for this purpose. Administrative record sources are used to build its database of businesses, known as the Business Register, including nonemployer businesses. The nonemployer frame, taken from the Business Register, is used to form the estimates for the annual Nonemployer Statistics program. It is assumed that most of the entities on the nonemployer frame with large income amounts do not represent true nonemployers, but they are not routinely surveyed to ensure that is the case. A survey of these cases will help to improve the quality of the Nonemployer Statistics and other economic measures. It will better define the type of entities that should be included in Nonemployer Statistics and produce measures of economic activity from an improved universe of true nonemployers.

This feasibility study will provide needed insights into the detailed requirements necessary for conducting a survey of these companies, which will assist with validating assumptions of business activity and the types of workers that perform those activities. Another area of focus is to investigate whether companies listed on the nonemployer frame are actually employer businesses that report employment under a payroll Employer Identification Number (EIN) that is different from the EIN used to report receipts. The results should lead to improved coverage of a difficult part of our employer and nonemployer frames and would be a stepping-stone to improved future survey data collections.

Two phases of data collection are planned. The first phase of this collection is cognitive and usability testing of the draft instrument, which will be administered using a Web application. One round of testing will be performed in the Washington, DC metropolitan area and possibly one or two other U.S. cities from mid-June through early July 2009. We will attempt to obtain interviews with at most 20 respondents from corporations and partnerships that vary in size and complexity. The purpose of cognitive and usability testing is to ensure that the questions are clear and correctly collect the desired data and that the Web instrument functions as intended. Minor changes to the Web application are anticipated after the cognitive and usability testing and prior to the beginning of the second phase of the study.

After businesses are recruited for the test interviews, participants will receive follow-up reminders about their appointments. Interviews will be conducted at the business site by staff from the Establishment Survey Methods Staff within the Census Bureau’s Office of Statistical Methods and Research for Economic Programs. Subject area specialists from the Census Bureau will accompany the researcher on most, if not all, of the interviews. They will assist with respondent debriefing, especially in cases where additional clarification of the subject matter is required. Interviews will be audio recorded with the participants’ permission to aid researchers in accurately reporting findings and recommendations.

The second phase of the feasibility study will be administered via the Web. It consists of an initial letter mailed to the sample inviting participants to complete the Web questionnaire, along with nonresponse follow-up consisting of a reminder mailing and telephone follow-up.

From early July through the end of September 2009, the Census Bureau will attempt to obtain a total of 500 completed questionnaires from candidate businesses, including those that participated in Phase 1 testing. The objective is to permit adequate evaluation of the questions, the Web application, and the feasibility of collecting this information in this way. The sample will be selected from the 2007 nonemployer frame and stratified using the following stratification criteria:

• Nonemployer EINs that had indications of employer status based on administrative data;

• Nonemployer EINs that were matched by name and address to the 2007 employer frame;

• EINs from the nonemployer frame within selected industries (real estate, finance and insurance, and others) that have large receipts.

Five hundred cases are needed to ensure sufficient variability in industry, size and type of entity, enabling us to adequately assess the collectibility of information needed to verify status and to validate matching criteria.

On July 20, the initial mailing letter will be sent to the entire sample except those few respondents surveyed during cognitive and usability testing. The initial letter explains the purpose of the survey to the respondent, provides the necessary information for logging on to the Web application, and provides contact information for any questions the respondent may have.

The Web application used for this study is password protected and will be administered through Centurion, the Census Bureau’s secure survey administration software. All data submitted via Centurion is completely confidential and securely transmitted to uphold Title 13 of the United States Code.

A second letter will be mailed to nonrespondents on August 3, 2009. This reminder letter again explains the purpose of the survey to the respondent, provides the necessary information for logging on to the Web application, and provides contact information for questions.

Census Bureau staff at the Jeffersonville Telephone Center will perform telephone follow-up of remaining nonrespondents beginning August 12 through September 25, 2009, or until 500 completed cases are obtained. The staff at Jeffersonville will be trained to perform data collection over the telephone. Respondent data will be entered into the Web application and linked to the respondent’s User ID.

At all phases of the data collection, respondents will be informed that their responses are voluntary and that the information they provide is confidential and will be seen only by Census Bureau or special sworn employees involved in the research project. We will not be providing monetary incentives to participants in this study.

Enclosed are copies of the questionnaire that is currently being converted to a Web application, the initial mail-out letter, and the nonresponse follow-up letter. The protocol for conducting the cognitive and usability testing interviews is also enclosed.

We expect only one respondent to be surveyed in each company. We estimate the length of the testing interviews will average 30 minutes, with an overall testing burden of 10 hours. We estimate the data collection via Web application or telephone follow-up will average 10 minutes per respondent, with an overall burden of 80 hours. Thus, the maximum estimated burden for this research is 90 hours {[(20 test interviews X 30 min/each + 480 completed cases X 10 min/each)  60 min/hr] = 90 hours}.

The contact person for questions regarding data collection and statistical aspects of the design of this research is listed below:

Michael E. Kornbau, Branch Chief

Program Research and Methods Branch

Economic Planning and Coordination Division

U.S. Census Bureau,

Washington, D.C. 20233

(301) 763-6779

Michael.E.Kornbau@census.gov