OMB Supporting Statement Part B 3.2.18_Census response

U.S. Department of Commerce

U.S. Census Bureau

OMB Information Collection Request

Annual Business Survey

OMB Control Number 0607-XXXX

Supporting Statement Part B. – Collections of Information Employing Statistical Methods

1. Universe and Respondent Selection

The Annual Business Survey (ABS) universe for this submission will be constructed using a combination of 2015 and 2016 Internal Revenue Service (IRS) tax return data for the 2017 respondent selection; 2016 and 2017 IRS tax return data for the 2018 respondent selection; and so forth, for the 2019-2021 respondent selections ¹. For reference year 2017, the final universe of 5.3 million employer businesses will be those reporting activity on any one of the following 2016 IRS tax forms: 1040 (Schedule C), “Profit or Loss from Business” (Sole Proprietorship); 1065, “U.S. Return of Partnership Income”; 941, “Employer’s Quarterly Federal Tax Return”; 944, “Employer’s Annual Federal Tax Return”; or any one of the 1120 corporate tax forms. For reference years 2018-2021, the final universe of businesses will be those reporting activity on tax forms 1040, 1065, 941, 944, or any one of the 1120 corporate tax forms using the previous year’s tax filings (i.e. 2018 will use 2017 tax forms, 2019 will use 2018 tax forms, etc). Current plans will only request responses from businesses filing the 941, 944, or 1120 tax forms.

The ABS plans to sample approximately 850,000 employer businesses in the benchmark survey year 2017. The large sample sized is needed to produce detailed statistics by owner demographics. Results from the large sample size should allow the Census Bureau to produce tabulations by detailed NAICS (2-6 digit), detailed geography (US, state, metropolitan statistical area (msa), county, and economic place), and detailed race (i.e. Chinese, Filipino, Japanese, Samoan, etc.). In the continuing annual years 2018-2021, the sample size will be approximately 300,000 employer businesses to reduce the burden on the respondents. The smaller sample size will provide summary statistics by owner demographics. Results from the smaller sample size should allow the Census Bureau to produce more frequent data, however the data provided will not be as detailed. For 2018-2021 tabulations plans include data by 2 digit NAICS, US, state, and msa, and will not include detailed race groups.

The Census Bureau uses administrative data to estimate the probability that a firm is minority- or women-owned. Each firm is then placed in one of nine frames for sampling. The sampling frames are; American Indian, Asian, Black or African American, Hispanic, Non-Hispanic White Men, Native Hawaiian and Other Pacific Islander, Some Other Race, Publicly Owned, and Women. The sample is stratified by state, industry, and frame. A standard type of estimation for stratified systematic sampling will be used. (Refer to Sampling Techniques by W.G. Cochran.) Title 13 United States Code, Sections 8(b), 131, and 182, Title 42 United States Code, Section 1861-76 (National Science Foundation Act of 1950, as amended), and Section 505 within the America COMPETES Reauthorization Act of 2010 authorize this collection. Sections 224 and 225 of Title 13 United States Code requires response from sampled firms. . The expected response for the ABS is estimated to be 70 percent. The expected response rates are based on results from the 2012 SBO, and previous Annual Survey of Entrepreneurs (ASE) and BRDI-M data collections. In the 2012 SBO, 69.4% of firms returned a questionnaire or submitted an online response. For the 2015 ASE 69.1% of firms submitted an online response (the only form of response available). The criteria that defines a firm as a respondent was more restrictive for SBO and ASE than will be used for ABS so the above percentages are good proxies to determine expected response for ASE. The Census Bureau will use an electronic reporting system to collect the data. No questionnaires will be mailed to respondents.

Businesses will be asked questions about the gender, ethnicity, race, and veteran status for up to four persons owning the majority of rights, equity, or interest in the business. Businesses will also be asked about characteristics of their businesses and the business owners.

The target relative standard error for the total number of women-, minority-, and veteran-owned businesses will be less than 10 percent. This figure is based on the 2012 SBO, and the ASE.

2. Procedures for Collecting Information

The ABS is an electronic-only collection. An initial letter that requires the respondents’ completion of the online survey and provides survey access instructions will be mailed from the Census Bureau’s National Processing Center in Jeffersonville, Indiana. There will be 850,000 letters mailed to employer businesses to collect data on business operations during 2017; and 300,000 letters mailed for each of the 2018-2021 data collections. The 2017 initial mail is scheduled for June 2018, with the 2018-2021 initial mails scheduled for June in 2019-2022. Responses will be due approximately 40 days from initial mail. All respondents will receive a due date reminder approximately one week before responses are due. Additionally, there will be two follow-up letter mailings to nonrespondents after the due date. Select nonrespondents may receive a certified mailing for the second follow-up if needed. Closeout of mail operations is scheduled for December. Upon the close of the collection periods, the response data will be processed, edited, reviewed, tabulated, and released.

3. Methods to Maximize Response

This information collection will maximize response through the following means:

• Mailing materials that emphasize the mandatory and confidential nature of census reports, as provided by Title 13, United States Code;

• Designing effective electronic reporting instruments and instructions;

• Offering a toll-free telephone number for companies that have questions or need assistance in completing the electronic survey;

• Conducting systematic mail follow-ups to nonrespondents;

• Reducing mailings of similar content to similar businesses by combing several surveys into the ABS.

• Conducting nonresponse bias analysis if unit response rate falls below 80 percent.

During collections, response rates will be monitored regularly. The overall response goal is 70 percent per collection. If response rates are below 36 percent prior to the start of the first follow-up, Census Bureau analysts will conduct address research on cases that flag as undeliverable. If response rates are below 56 percent prior to the second follow-up, the Census Bureau will conduct the follow-up using targeted certified mailings.

Post collection, unit response rates will be evaluated. The Census Bureau will conduct a nonresponse bias analysis if unit response rates fall below 80 percent. The nonresponse bias analysis will examine unit response rates and total quantity response rates to determine whether there is evidence of nonresponse bias to the key estimates of the survey. The analysis will focus on the differences of response rate by owner characteristics (by gender, ethnicity, race, and veteran status), industry, state-level geography, and legal form of organization. The analysis will also use available administrative receipts data to examine if there is evidence of nonresponse bias in the estimates of the size of firms and evaluate donor imputation methods. If required, the nonresponse bias analysis will be conducted following the end of the current data collection period and prior to the start of the collection period for the next survey year.

4. Tests of Procedures or Methods

The information collections will use procedures that are based on a considerable body of experience with economic censuses and surveys, such as cognitive testing, expert panels, and evaluation of data results. For the 2017 ABS, the Census Bureau conducted nineteen interviews to evaluate several new questions. The Census Bureau will also conduct usability testing on the collection instrument. Previous economic censuses also have been the subject of evaluation studies that have examined methodology and conceptual issues. As a result, the procedures used by the Annual Business Survey are very well tested.

The survey questions have been successfully tested through cognitive interviews (see attachments E and F).

5. Contacts for Statistical Aspects and Data Collection

Persons responsible for directing the sampling, estimation, nonresponse adjustment and disclosure avoidance procedures:

Xijian Liu, Methodology Director

Retail, Wholesale, & Services Programs

Economic Statistical Methods Division

U.S. Census Bureau

Washington, DC 20233

301-763-4272

[email protected]

James W. Hunt, Chief

Financial and Business Owners Surveys Statistical Methods Branch

Economic Statistical Methods Division

U.S. Census Bureau

Washington, DC 20233

301-763-6599

[email protected]

Persons responsible for directing the survey design, content development, processing, analysis and data publication:

Aneta Erdie, Assistant Division Chief

Governments and Business Owners Programs

Economic Reimbursable Surveys Division

U.S. Census Bureau

Washington, DC 20233

301-763-4841

[email protected]

Patrice C. Norman, Chief

Business Owners Branch

Economic Reimbursable Surveys Division

U.S. Census Bureau

Washington, DC 20233

301-763-7198

[email protected]

Attachments

1. Initial Mail and Follow-up Letters

2. ABS – 1

3. ABS – 2

4. Electronic instrument screens showing the burden statement, collection authority, and purpose and uses statements available to respondents.

5. ABS Cognitive Testing Results

6. ABS Innovation Cognitive Testing Results

7. Presubmission Notice Comments

8. 30 – Day Notice Comments

1 This clearance package will only cover 3 of the 5 year survey. The Census Bureau will prepare a new package for the remaining survey years at the appropriate time.

3