Quick Business Survey
Operations Test 2
1220-0192
March 2019
Second Operations Test of the Proposed Quick Business Survey
Part A. Justification
1. Necessity of the Information Collection
This request is for the approval of a second nationwide, operations test for a Quick Business Survey (QBS). The Bureau of Labor Statistics (BLS) will build on the results of a prior test to further evaluate QBS survey processes and operations in a possible production environment.
If successful in the long term, a QBS would allow the BLS to collect information about the US economy in a more efficient manner than is currently possible. This would allow the BLS and data users to be able to understand the impact of specific events (e.g., policy changes, strikes, resource shortages, health epidemics, terrorist attacks, and severe weather) on the economy in a timely manner, allowing policy makers to be able to make informed decisions. Relevancy is one of the main aspects of the BLS mission, and a QBS would allow BLS to provide information in a more timely way that would be extremely relevant to data users.
The Quarterly Census of Employment and Wages (QCEW) program produces a comprehensive tabulation of employment and wages for workers covered by state Unemployment Insurance (UI) laws and Federal workers covered by the Unemployment Compensation for Federal Employees (UCFE) program. The UI administrative records underlie the QCEW. While based on the UI administrative data, it is the detailed review and editing that yield the QCEW and the additional detail and information collected and updated via the Multiple Worksite Report (MWR) and Annual Refiling Survey (ARS) that allows the QCEW to serve as the sampling frame for BLS establishment surveys. The information collected, edited, and maintained by the QCEW program allows BLS to publish information reflective of the entire U.S. economy. QCEW currently is the most frequent, comprehensive, accurate, and timely publication of information available at the maximum industry and geographical detail on the US economy. Additionally, the QCEW is the only source of monthly employment for the entire US economy.
Each year, the QCEW program conducts the ARS by reaching out to approximately 1.2 million establishments requesting verification of their main business activity and their mailing and physical location addresses. Over the last five years, BLS has successfully transitioned the ARS to a fully on-line survey and accelerated the timeframe for collection. This fully web-based ARS introduces a low-cost platform for conducting Quick Business Surveys. The QBSs accompanying the ARS would have little data collection overhead, leveraging the address refinement, printing, and mailing efforts that are undertaken as part of the production ARS. Respondents that are already logged into the ARS secure website could then be directed to a QBS and asked to answer a very limited number of additional survey questions after completing the ARS.
In addition to using the data collection platform, a QBS would allow BLS to leverage the multitude of information already known about the sample units to allow for targeted sampling. Samples could be selected based on monthly employment, quarterly wages, detailed industry codes, date of business birth, specific geocodes along with county, MSA, state designations, linkage to a larger EIN-based firm, non-profit vs for-profit status, etc. Thus, BLS can target in a QBS only the units meeting the specific set of characteristics desired allowing BLS to delve into specific areas of economic interest without burdening establishments which do not meet the specific targeted features. For example, if a policy regarding changing the minimum wage is enacted, a QBS can target establishments in industries that are more likely to be affected, such as Accommodation and Food Services. A QBS can ask questions regarding how business or hiring practices are changing now that the minimum wage has changed. Using information available about ARS respondents, BLS has a rich sampling frame in the QCEW from which to measure economic characteristics and behaviors in an efficient way that places a low burden on US establishments.
In order to develop a QBS, BLS is seeking to conduct a second operations test. There were three goals of the first test that will be further evaluated:
To develop and evaluate a QBS system. The information from this test will be used for internal purposes, to evaluate timelines and costs associated with development and fielding a QBS.
To establish the extent to which ARS respondents have access to different types of information, providing parameters for future QBS surveys.
To understand how ARS respondents will react to receiving additional questions after they’ve submitted their ARS. Being able to estimate response rates, using two sample designs, will allow BLS to understand the potential that the QBS has moving forward.
Additionally, this second test seeks to evaluate revised materials and procedures from the first test. Specifically, we are looking to:
Implement a refined sampling methodology for small units to avoid unnecessary oversampling.
Improve the quota sample methodology by adding randomization.
Improve response rates, both overall and for units not traditionally in the ARS.
Develop a detailed cost model for use in future implementation.
During the QBS Operations Test, data collection will run for one period of approximately nine months. The test is estimated to run between July 2019 and June 2020. A full assessment of the information obtained during the test will be conducted in addition to a careful evaluation of the processes, response rates, response times, and other test features.
2. Uses of Information
The data collected will be used primarily to evaluate the QBS, and the resources required in terms of response rates and potential nonresponse bias. The data collected will be used for internal research purposes to gain insight into current BLS data and capacities and to allow for informed decisions to be made about future implementations of the QBS.
Collecting information about the relationship respondents have with the company they’re reporting for, the ownership of that company, as well as what type of information they have access to about that company, will provide an idea of what topics can be reliably covered in a QBS.
Another goal of the second test is to prove whether estimates can be successfully generated using a stratified quota sample outlined in Part B. This technique, aimed at reducing cost to the government, reducing respondent burden, and accelerating data collection is new to BLS and we need to compare it to our traditional method to ensure that statistically sound, non-biased estimates can be produced using this method.
Besides comparing response rates to the previous test of the QBS and other surveys conducted by BLS, a question is added to this second test that is explicitly chosen to test for biases that may result from the new sampling strategy. The added question about employment in the establishment being surveyed is intended to be identical to the question asked in the BLS Current Employment Statistics (CES) survey. The goal of asking this question is that it allows analysis comparing employment numbers from this test run, to employment numbers reported in the CES which uses a random sample, and also information about employment in the QCEW which is administrative data. Comparing responses to these other data sources gives a measure of how successful this particular sample design is. This additional question is necessary to be able to test for substantive biases in responses from the new sample design. Lastly, the previous test had a median response time of less than 2 minutes, so this additional question should not have a meaningful impact on respondent burden.
3. Electronic Collection Methods
The data collection will use available information and technology to reduce costs to the government and the burden on the respondents. The QBS will be an online survey. There will be one method of solicitation for the QBS. Solicitation will be a request to answer additional questions upon completion of the ARS using the ARSWeb online application. Follow-up with non-respondents will be conducted via the normal non-response follow up conducted for the ARS in accordance with O.M.B. No. 1220-0032. This includes solicitation via email when an email address is available and via one-page letters.
4. Efforts to Identify Duplication
For the QBS Operations Test, the data collected will be used for internal purposes only, to evaluate survey processes and the feasibility of developing and fielding a QBS in a production environment. Therefore, duplication of data is not an issue. In addition, for most of these questions, no similar information of this type is available to the BLS.
5. Impact on Small Businesses
The QBS Operations Test is designed to provide information that are representative of establishments nationwide and will be conducted for a limited time only. Respondent burden will be minimized through the use of online reporting and by keeping the questionnaires short, restricting questions to generally available information and specifically developed for simplicity of understanding and ease of completion. Surveying smaller institutions will be necessary to fulfill information needs; however, information already known about the sample units will allow BLS to selectively target establishments. Targeted sampling will allow BLS to collect the information needed to evaluate the QBS process without over-burdening establishments.
6. Consequences of Not Collecting the Data or Less Frequent Data Collection
The goal of the QBS Operations Test is to evaluate the feasibility of developing and fielding a QBS in a production environment. Without conducting these tests, BLS would be unable to pursue further development of a QBS.
7. Special Circumstances
The QBS Operations Test will be conducted over a nine month period. Data will be collected in a manner consistent with the guidelines in 5 CFR 1320.5 with one exception. Responses are requested in fewer than 30 days. Upon completion of the ARS, respondents will be requested to immediately complete the QBS. Due to the nature of the QBS, data are intended to be collected quickly in order to inform the public in as timely a manner as possible.
8. Federal Register Notice/Outside Consultation
No public comments were received as a result of the Federal Register notice published in the Federal Register, 83 FR 67360 on December 28, 2018, for the Quick Business Survey Operations Test (OMB No. 1220-0192).
Outside Consultation
The BLS is in constant contact with the State Workforce Agencies (SWAs) on the administration of the QCEW program and will continue to partner with these parties on the QBS, such as:
Daily contact between BLS Regional Offices and SWA officials.
Field visits by BLS Washington and Regional Office staffs to SWA offices.
Guidance from the QCEW Policy Council established by BLS in 2000. The Council is composed of State and BLS staff members that provide ongoing technical expertise and research as well as assist in the prioritization of program enhancements.
Periodic meetings with members of the National Association of State Workforce Agencies (NASWA).
Consultation with members of the Federal Data Users Advisory Committee (DUAC).
BLS staff present papers and participate in the International Conference on Establishment Statistics and American Statistical Association conferences where emerging technologies for collecting data from employers and lowering respondent burden are discussed in great detail. BLS staff also present papers and participate in the Joint Statistical Meetings (JSM), which is the largest gathering of statisticians is held in North America. In addition, BLS staff present and participate in FedCASIC, which is a series of annual meetings sponsored by the Census Bureau and BLS to include all forms of computer assisted survey information collection (CASIC).
9. Gifts or Payments to Respondents
There will be no gifts or payments to respondents.
10. BLS Confidentiality Policy
The Confidential Information Protection and Statistical Efficiency Act (CIPSEA) safeguards the confidentiality of individually identifiable information acquired under a pledge of confidentiality for exclusively statistical purposes by controlling access to, and uses made of, such information. CIPSEA includes fines and penalties for any knowing and willful disclosure of individually identifiable information by an officer, employee, or agent of the BLS.
The majority of BLS surveys protect the confidentiality of respondent provided data based on CIPSEA and in those surveys BLS provides respondents with a confidentiality (or CIPSEA) pledge which states that the data will be protected from unauthorized disclosure and used for statistical purposes only. ARS respondents are not provided with a confidentiality pledge. This is because the ARS data are used by the States for Unemployment Insurance purposes. However, when respondents are directed to the QBS upon completion of the ARS, they will be provided with a confidentiality pledge. The standard BLS confidentiality pledge/informed consent statement will appear on the first page of the QBS survey, following submission of their ARS data:
The Bureau of Labor Statistics, its employees, agents, and partner statistical agencies, will use the information you provide for statistical purposes only and will hold the information in confidence to the full extent permitted by law. In accordance with the Confidential Information Protection and Statistical Efficiency Act (44 USC 3572) and other applicable Federal laws, your responses will not be disclosed in identifiable form without your informed consent. Per the Federal Cybersecurity Enhancement Act of 2015, Federal information systems are protected from malicious activities through cybersecurity screening of transmitted data.
BLS policy on confidentiality states: “Respondent Identifiable Information acquired or maintained by the BLS for exclusively statistical purposes and under a pledge of confidentiality shall be treated in a manner that ensures the information will be used only for statistical purposes and will be accessible only to authorized individuals with a need-to-know.”
11. Sensitive Questions
No sensitive questions will be asked during the survey.
12. Estimated Reporting Burden
Number of respondents: 26,286
Frequency of response: One Time
Estimated time per response 3 minutes
Annual burden hours: (26,286 x 3)/60 = 1,314
Estimated cost to respondents $26.40 x 1,314 = $34,689.60
The estimated time per response is a conservative estimate based on responses to the first QBS test. The cost to respondents is based on employer costs for wages, salaries, and benefits for Office and Administrative Support occupations was obtained from the BLS National Compensation Survey, Employer Costs for Employee Compensation program. This rate was $26.40/hour based on September 2018 data (https://www.bls.gov/news.release/pdf/ecec.pdf).
13. Capital/Start-up costs
There are no capital/start-up costs or operation and maintenance and purchase of service costs resulting from the collection of this information.
14. Cost to the Federal Government
The estimated cost to the Federal government for the QBS Operations Test survey is $45,000.
15. Program Changes or Adjustments
The estimated respondent burden for the second QBS Operations Test survey is higher than the estimated burden the first operations test due to the increased sample sizes needed to test refined sampling methodologies.
16. Plans for Publication
The information collected during this research project will not be released for publication. The information gathered will be used to evaluate a new information collection and the development of a data collection system that will be put forth for clearance in a separate package.
The information collected during this research project will be used for publication of research-oriented papers about the process of developing this new collection and collection system. It will not focus on the content of the data collected, but rather on the response rates and overall effectiveness and viability of the QBS.
17. Approval to not Display the OMB Expiration Date
Approval to not display the expiration date for OMB approval is not being sought.
18. Exceptions to the Certification Statement
There are no exceptions to the certification statement.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Rowan, Carol - BLS |
| File Modified | 0000-00-00 |
| File Created | 2021-01-15 |