Supporting Statement Part B

Page 3 of 3

Department of Commerce

U.S. Census Bureau

OMB Information Collection Request

Business and Professional Classification Report

OMB Control Number 0607-0189

PART B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL

METHODS

1. Universe and Respondent Selection

The universe of births consists of newly opened or newly acquired employer businesses. These new businesses are identified by an IRS-assigned Employer Identification Number (EIN). The universe of births averages 480,000 EIN units annually, usually between 100,000 and 150,000 EIN units per quarter. Sample selection, mailing, and processing are completed once a quarter, on approximately one-fourth of the annual EIN universe of births. The Census Bureau obtains information on these EIN units from the IRS and the SSA. The process is a two-phase sampling procedure. From the list of EIN units, a first phase sample of approximately 14,250 EIN units are selected for each quarterly mailing, making the total potential number of annual responses 57,000. This number varies from quarter to quarter, but is typically ranges between 10- 12 percent of births for that quarter. It can also vary between sample redesigns due to changes in sampling parameters, industry coverage, and other sampling requirements. The entire mailing for the survey is divided into four components over the year for processing purposes. An EIN unit is assigned to a first phase sampling size stratum based on its NAICS code and quarterly payroll. These EIN units are sampled with an equal probability systematic sampling scheme, within each NAICS code by size stratum.  The SQ-CLASS report requests firms to designate their type of business activity, two months of sales or receipts, principle lines of merchandise, whether the firm is owned or owns another establishment, not-for-profit status, wholesale type of operation, inventories, and method of selling for retail and wholesale firms. The information collected from the SQ-CLASS report is then used to determine the EIN's eligibility to continue on in the second phase of the birth sampling process. Those second phase eligible EINs are assigned to a second phase sampling size stratum, based on their reported NAICS code and sales or receipts. A second phase sample of approximately 1,600 EIN units quarterly, or equivalently, 6,400 EIN units annually, is then selected. The second phase selected EINs are then added to the current surveys' ongoing samples for data collection and processing.

Assigning proper classification to these new EIN units is crucial for both the five year Economic Census and current business surveys. We conduct mail reminder follow-ups, followed by telephone follow-ups for non-response cases, in an attempt to maximize response rates. For those cases that we are unable to find a telephone number for, we mail another follow-up letter. The average response rate is approximately 68 %.

2. Procedures for Collecting Information

Each quarter, the first-phase selected EIN units are mailed an SQ-CLASS letter that requests respondents to report via the Internet using a secure respondent portal and report two recent months of sales or receipts. This information is then converted to a measure of size used for second phase sampling. The second phase sampling utilizes a probability proportional to size systematic sampling method, and is conducted quarterly. Following the second phase sampling, the newly selected EIN units are added to the appropriate survey.

3. Methods to Maximize Response

We have taken the following actions to maximize response rates:

• Offer customized mailing material for selected establishments

• Provide three mailing attempts to contact establishments

• Offer delinquent establishments a toll-free number to call and report

• Offer Internet reporting

• Plan telephone follow-up to contact delinquent respondents

• Provide a Web site with responses to Frequently Asked Questions

• Provide a facsimile option to submit form

• Provide a toll-free number that companies can call for assistance

In August 2016, the Census Bureau tested a different size and type of envelope (pressure sealed envelope) that was sent to respondents to see if that affected response. Based on the results of this test, there was no statistical difference between the response rate for the pressured sealed envelope and the traditional envelope. Historically, the Census Bureau has mailed respondents an initial letter, followed by two additional letters for delinquent cases and has conducted telephone follow-up during the last six weeks of quarterly processing. In an attempt to increase response rates in May 2017, a test reminder letter was mailed encouraging respondents to call the toll free telephone number to report. Initially, respondents that received this test letter called to report but ultimately the overall response rate remained unchanged. Additionally, in an attempt to make it easier for respondents to identify their type of business and reduce the burden of having to describe their business activity, the Census Bureau has updated the wording of the question on firm’s primary business activity to reflect descriptions received from prior respondents. We have also attempted to improve response rates by reducing the amount of time before follow-up letters to delinquent cases are mailed. We now try to mail follow-up letters within one week of the survey due date indicated on letter however, there hasn’t been sufficient time to measure the impact on the overall response rates.

4. Testing of Procedures

The current procedures reflect a progressive improvement over many years. We systematically monitor data collection procedures in order to identify ways to reduce burden and streamline processing. We plan to continue researching the accuracy of the NAICS codes used in our first phase sampling. Results of this research may allow us to reduce the number of cases mailed in the future.

5. Contact for Statistical Aspects and Data Collection

Direct questions regarding the sample design and statistical methodology used for this survey to Katrina Washington, Chief, Service Surveys Statistical Methods Branch, 301-763-7212. The assignment of NAICS classification codes under the direction of Scott Handmaker, Chief, Classification Processing Branch, 301-763-7107. Planning, implementation, and information analyses of the current business surveys are under the direction of Mary Potter, Assistant Division Chief, Quarterly and Manufacturing Indicator Programs, 301-763-4207; Andrew Baer, Assistant Division Chief of Services Sectors, 301-763-4802; Scott Scheleur, Assistant Division Chief of Retail and Wholesale Indicator Programs, 301-763-7128; and Ian Thomas, Assistant Division Chief of Retail and Wholesale Sectors, 301-763-7122.

Attachments

1. Initial letter - SQ-CLASS_L1

2. Reminder letter - SQ-CLASS_L1R

3. Follow-up letter for cases delinquent cases - SQ-CLASS-L2

4. Screenshots of SQ-CLASS questions

5. Copy of SQ-CLASS instructions

6. Bureau of Economic Analysis (BEA) letter of support