QBS_Supporting_Statement_Part_B_2024

QCEW Business Supplement

OMB Control Number: 1220-0198

OMB Expiration Date: 07/31/2024

SUPPORTING STATEMENT FOR

Quarterly Census of Employment and Wages

Business Supplement

OMB Control No. 1220-0198

B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS

1. Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection methods to be used. Data on the number of entities (e.g., establishments, State and local government units, households, or persons) in the universe covered by the collection and in the corresponding sample are to be provided in tabular form for the universe as a whole and for each of the strata in the proposed sample. Indicate expected response rates for the collection as a whole. If the collection had been conducted previously, include the actual response rate achieved during the last collection.

1a. Universe

The Bureau of Labor Statistics (BLS) conducts surveys under the Quarterly Census of Employment and Wages Business Supplement (QBS) on a sample of establishments included in the BLS Quarterly Census of Employment and Wages (QCEW). The universe of respondents to the QCEW are the 50 States, the District of Columbia, Puerto Rico, and the U.S. Virgin Islands. The primary source of data for these 53 entities are the Quarterly Contribution Reports (QCRs) submitted to State Workforce Agencies (SWAs) by employers subject to state Unemployment Insurance (UI) laws. The QCEW data, which are compiled for each calendar quarter, provide a comprehensive business name and address file with employment and wage information by industry, at the six-digit North American Industry Classification System (NAICS) level, and at the national, state, Metropolitan Statistical Area (MSA), and county levels for employers subject to state UI laws. Similar data for Federal Government employees covered by the Unemployment Compensation for Federal Employees program (UCFE) also are included. Each year, the QCEW program contacts approximately one-third of the universe units that meet eligibility criteria for the Annual Refiling Survey (ARS) to verify general business information including address, county, and main business activity (see OMB 1220-0032).

The 2024 QBS universe of inference will be based on the most recent quarter of QCEW data that are available at the time of sample selection. This will cover about 11 million establishments with about 2 million being out of scope based on the following constraints:

• Private-sector establishments only; not public administration

• Average monthly employment > 0 (average taken over the past year)

• NAICS 814110 (private households) and 491110 (postal service) are excluded

• Establishments with NAICS 624120 (services for the elderly and persons with disabilities) with an average monthly employment < 2 are excluded

• NAICS 999999 (NAICS unassigned/unidentified in QCEW) are excluded

1b. Sample size

For the 2024 QBS survey, BLS plans to select a sample of approximately 320,000 establishments. The objective of the large sample is to produce statistics at detailed levels including by size class, state, industry, and some state-industry and state-size combinations. Production of statistics at any detailed levels will depend on the ability to pass disclosure requirements to ensure confidentiality of the establishments responding to the survey.

	Sample	Estimated Responses (25%)
2024 QBS Sample	320,000	80,000

The response rate for the 2022 QBS survey, which had a virtually identical questionnaire, was 27 percent.

2. Describe the procedures for the collection of information including:

• Statistical methodology for stratification and sample selection,

• Estimation procedure,

• Degree of accuracy needed for the purpose described in the justification,

• Unusual problems requiring specialized sampling procedures, and

• Any use of periodic (less frequent than annual) data collection cycles to reduce burden.

2a. Sample Design and Selection Procedures

The sample design and selection process involves the following steps:

• Determine the 2024 QBS universe stratification

• Identify the research goals

• Calculate sample sufficiency and stratum sample sizes

• Design sample to achieve goals and sufficiency, subject to constraints

• Select sample

Stratification

Generally, the QBS universe is stratified by geography, industry type, and industry size. The 2024 QBS universe will be stratified by state, two-digit NAICS groupings, and industry size classes.

For the 2024 QBS survey, the geographic stratification component consists of 53 localities, including the fifty states plus the District of Columbia, Puerto Rico, and the U.S. Virgin Islands.

For the 2024 QBS survey, the industry type stratification component consists of the two-digit NAICS sector groups listed in the table below. For the purpose of defining research goals later, the table categorizes each two-digit NAICS sector group as goods-producing or services-producing. Note that NAICS sector 92 (Public Administration) establishments are excluded from the 2024 QBS survey.

Modified NAICS Sector	Goods- or Services-Producing	Description
11-21	G	Agriculture, Forestry, Fishing, and Hunting and Mining, Quarrying, and Oil and Gas Extraction
22	S	Utilities
23	G	Construction
31-33	G	Manufacturing
42	S	Wholesale Trade
44-45	S	Retail Trade
48-49	S	Transportation and Warehousing
51	S	Information
52-53	S	Finance and Insurance and Real Estate and Rental and Leasing
54-56	S	Professional, Scientific, and Technical Services; Management of Companies and Enterprises; and Administrative and Support and Waste Management and Remediation Services
61	S	Educational Services
62*	S	Health Care and Social Assistance (*Excluding Services for the Elderly and Persons with Disabilities establishments with less than two employees)
72	S	Accommodation and Food Services
81	S	Other Services (except Public Administration)

For the 2024 QBS survey, the industry size stratification component consists of nine categories, where an establishment’s size is based on its number of employees. For the purpose of defining research goals later, these nine narrow categories are then further grouped into four broader size classes. The table below shows both the narrow and broad industry size classes.

Narrow Industry Size Class	Broad Industry Size Class
1-4	1-19
5-9
10-19
20-49	20-99
50-99
100-249	100-499
250-499
500-999	500+
1000+

Research Goals

The sample for the 2024 QBS survey is designed to support publication using the following breakouts:

• State by Industry Goods-Producing/Services-Producing categorization

{53 states * 2 industry categorizations =106 estimation cells}

• State by Broad Industry Size

{53 states * 4 broad industry size categories = 212 estimation cells}

• Modified NAICS Sector by Broad Industry Size Class

{14 modified NAICS sectors * 4 broad industry size categories = 56 estimation cells}

• Narrow Industry Size Class

{9 narrow industry size classes = 9 estimation cells}

Sample Sufficiency and Stratum Sample Sizes

Sufficiency is determined for each estimation cell within each of the four survey analysis breakouts listed in the Research Goals section. Each of these estimation cell level sufficiency counts is proportionately allocated to all strata that map to the estimation cell. This results in four allocated sufficiency counts per stratum. For each stratum, the maximum of the four allocated sufficiency counts is chosen to ensure proper sufficiency across all four research goals. If this value is greater than or equal to three, it is used as the stratum’s final sufficiency count. Otherwise, the stratum’s final sufficiency count is set to three provided the stratum contains enough selectable establishments.

Sufficiency is determined based on estimating proportions to a certain degree of precision, where precision is based on weighing researcher needs versus survey burden and cost. The formula for the sample sufficiency of an estimation cell is based on the deconstruction of the formula for the variance of a proportion (using simple random sampling within the cell):

Once the final sufficiency count is determined for each stratum, the stratum sample size is determined by dividing the sufficiency count by a response rate estimate. If the resulting value is less than or equal to the number of selectable establishments in the stratum, it is set as the final stratum sample size. Otherwise, the final stratum sample size is set equal to the number of selectable establishments in the stratum. If, after this process, the resulting sample size no longer supports sufficiency for an estimation cell, additional sample size will be allocated to strata within the estimation cell to compensate.

Sample Design

The 2024 QBS sample will follow a stratified design, with strata as defined earlier. The survey will be synchronized with the QCEW’s Annual Refiling Survey (ARS). This is an efficient means of QBS data collection. However, it also imposes some complications on the QBS sample design that stem from differences between the two surveys in terms of function and scope.

The main purpose of the ARS is to update QCEW administrative information for a subset of the QCEW population. The ARS is administered to employers with employment greater than three. These establishments are referred to herein as “ARS eligible.” ARS-eligible establishments are administered the ARS on a rotating basis, with most rotating every three years, and some in low-change industries rotating every six years. ARS-eligible establishments that are rotating into a particular year’s ARS are referred to herein as “ARS active” and/or “QBS active.” Conversely, ARS-eligible establishments that are not rotating into a particular year’s ARS are denoted as “ARS inactive” and/or “QBS inactive.” ARS rotation schedules are assigned randomly such that a typical ARS-eligible establishment will be administered the ARS in three-year intervals.

The main purpose of QBS surveys is to make statistical inferences about a population of interest. Typically, the QBS population of interest is not limited to the ARS-eligible population. Specifically, QBS coverage typically includes employers with employment of three or less. Establishments that are in the QBS population of interest, but that are not ARS eligible, are herein referred to as “ARS ineligible.”

When selecting from the ARS-eligible part of the QBS universe, only ARS-active establishments are allowed to be selected. Establishments that meet this criterion are considered “QBS selectable.” This sampling practice reduces QBS cost and respondent burden over time.

In previous QBS surveys, when selecting from the ARS-ineligible part of the QBS universe, no restrictions were placed on the ARS ineligible sub-population, meaning that each ARS-ineligible establishment could be randomly selected each time QBS samples were drawn. In other words, all ARS-ineligible establishments were QBS selectable for each previous QBS survey. Starting with the 2024 QBS survey, two new sampling flexibilities will be implemented that will result in some ARS-ineligible establishments being QBS selectable, while others will be categorized as QBS not selectable. One of these flexibilities addresses respondent burden and one addresses survey cost, as detailed below.

The first new QBS sampling flexibility that will be applied to ARS-ineligible establishments involves putting them on a random three-year rotation in the same way that ARS-eligible establishments are. This will result in some ARS-ineligible establishments being categorized as QBS active and others being categorized as QBS inactive. ARS-ineligible establishments that are QBS inactive will not be selectable for the current QBS survey. This sampling practice reduces respondent burden over time for ARS-ineligible establishments that would otherwise tend to be selected in QBS samples more frequently than once every three years. Use of this practice will be monitored to make sure it does not overly constrain selection to the point of sample degradation.

The second new QBS sampling flexibility that will be applied involves implementing the ability to oversample ARS-ineligible establishments for which there is access to usable contact email addresses. When this flexibility is used, sampling weights will be adjusted accordingly. This sampling practice will reduce QBS costs by reducing the cost to contact sampled ARS-ineligible establishments. This practice will also be monitored for its effects on the sample.

Given the differences between the functions and scope of the QBS versus the ARS, the various sampling constraints and flexibilities discussed above, and to facilitate the description of sample selection practices, QBS strata and sub-strata are classified into the structure below:

1. Strata with Only ARS-Eligible Establishments

1. ARS/QBS Active (QBS Selectable)

2. ARS/QBS Inactive (QBS Not Selectable)

2. Strata with Only ARS-Ineligible Establishments

1. QBS Active (QBS Selectable)

1. Usable Email Address (Flexibility to Oversample)

2. No Usable Email Address (Flexibility to Undersample)

2. QBS Inactive (QBS Not Selectable)

3. Mixed Strata Containing Both ARS-Eligible and ARS-Ineligible Establishments

1. ARS-Eligible

1. ARS/QBS Active (QBS Selectable)

2. ARS/QBS Inactive (QBS Not Selectable)

2. ARS-Ineligible

1. QBS Active

1. Usable Email Address (Flexibility to Oversample)

2. No Usable Email Address (Flexibility to Undersample)

2. QBS Inactive (QBS Not Selectable)

As noted earlier, the ARS is administered to roughly one-third of the ARS-eligible establishments on a rotating schedule. The ARS primarily uses the 7^th-8^th digits of the EIN to subset the QCEW universe into thirds for data collection purposes. Based on previous evaluation, the 7^th-8^th digits of the EIN were determined to be a suitably random method of selection. For QBS purposes, the same random mechanism will be applied to ARS-ineligible establishments to assign QBS active versus QBS inactive status.

Sample Selection

Sample selection for the 2024 QBS survey will be conducted slightly differently depending on the type of stratum, where stratum types are described in the outline above.

• Sample Selection in Strata with Only ARS-Eligible Establishments [Stratum Type I]

Strata with only ARS-eligible establishments are segmented into ARS-active and ARS-inactive subsets. Within each of these strata, a random sample will be drawn from the ARS-active subset only.

• Sample Selection in Strata with Only ARS-Ineligible Establishments [Stratum Type II]

Strata with only ARS-ineligible establishments are segmented into QBS-active and QBS-inactive subsets. Within each stratum of this type, establishments will be sampled from only the QBS-active subset. Within the QBS-active subset, two options exist. The first option involves the random selection of all establishments regardless of email address versus no email address distinction. The second option involves segmenting the QBS-active subset into email address and no email address groups, then allocating a disproportionately higher fraction of the stratum sample size to the email address group as compared to the no email address group, adjusting sampling weights accordingly. After allocating the stratum sample size to the email/no email groups, random samples will be selected from within each group. This second option would be enacted as a cost savings measure, if needed.

• Mixed Strata Containing Both ARS-Eligible and ARS-Ineligible Establishments [Stratum Type III]

Strata with both ARS-eligible and ARS-ineligible establishments are first segmented into ARS-eligible and ARS-ineligible subsets. Next, within each stratum of this type, the stratum sample size is split up and allocated to the two stratum subsets. The allocation will be done proportionately unless a cost-based need to oversample from ARS-eligible strata subsets arises. Once the stratum sample size is parsed, sample selection for the ARS-eligible subset echoes the sample selection description for Stratum Type I given earlier. Similarly, sample selection for the ARS-ineligible subset echoes the sample selection description for Stratum Type II given above.

2b. Estimation Procedure

The primary measure of interest will be an estimated proportion possessing an attribute being assessed by a survey question. Strata estimates will be calculated within each stratum, provided the stratum has usable survey responses. Composite estimates will be constructed for analysis cells that comprise multiple strata. For example, survey estimates for Virginia establishments of broad industry size 20-99 (i.e., VA/20-99) will be built as a weighted composite of twenty-eight strata estimates – one stratum estimate for each of the fourteen two-digit NAICS sector groups for VA/20-49 and one stratum estimate for each of the fourteen two-digit NAICS sector groups for VA/50-99.

More specifically, proportions will be estimated for each of the analysis breakout cells described in the Research Goals section in 2a. Composite estimates for additional analysis breakouts can and will be produced, since the formulas below can be generalized to composite estimators over various strata groupings. Typically, estimates are also produced at the state level, the two-digit NAICS sector group level, and the narrow industry size level, among others.

It should be noted that the sample was not designed with the intention of being sufficient for analyses other than the four analysis breakouts identified in the research goals. However, sufficiency will still be achieved for analysis breakouts that are broader than corresponding analysis breakouts in the Research Goals section. For example, the sample was designed to achieve sufficiency for State*Goods/Services breakouts. Therefore, sufficiency should be achieved at the broader State breakout.

The proportion estimate formula is:

Each stratum weight is the population proportion of each stratum relative to the composite population of interest:

The formula for the estimated sample proportion for some stratum (h), generalized for non-response adjustment, is:

The proportion of non-responders possessing the attribute of interest is generally unknowable. Therefore, the assumption is made that, within each stratum, responders and non-responders possess the attribute of interest in the same proportion, and therefore the formula reduces as follows:

At a later date, survey sponsors may decide to request composite estimates for new analysis breakouts that were not determined ahead of time. The definitions above can be generalized to composite estimators over various strata groupings by redefining the universe to the one of interest. However, the sample was not designed with the intention of being sufficient for these new analyses.

Strata estimates will be calculated within each stratum, provided the stratum has usable survey responses. Composite estimates will be constructed for analysis cells that comprise multiple strata. For example, survey estimates for Virginia establishments of broad industry size 20-99 (i.e., VA/20-99) will be built as a weighted composite of twenty-eight strata estimates – one stratum estimate for each of the fourteen modified NAICS sectors for VA/20-49 and one stratum estimate for each of the fourteen modified NAICS sectors for VA/50-99.

2c. Reliability

Variance estimation will involve (i) the application of the general formula for the variance of a composite proportion estimator drawn from a stratified random sample and (ii) the application of the basic formula for the variance of a proportion drawn from a simple random sample. Specifically, the variance of the proportion estimator for some particular analysis cell is:

Under the assumption that, within each particular stratum, non-responders possess the attribute of interest in the same proportion as responders, the formula for the within-stratum variance of a proportion calculated from a simple random sample is:

In the formula above, note that is the number of establishments in stratum h that responded to the survey question of interest. It is not the stratum h sample size.

The formulas above can be tailored to the desired composite estimator by applying it across only the set of strata that are relevant to that particular composite.

2d. Data Collection Cycles

Since the 2024 QBS survey will largely be collected directly after respondents submit the ARS, the data collection cycle for the QBS will follow the established pattern of collection used for the ARS as outlined in materials for OMB Control No. 1220-0032.

Overall data collection is expected to be conducted over an eight to twelve-week period starting in late July 2024. This will give time for response rate review and analysis to determine the optimal data collection outreach methods to maximize response rates and reduce burden. BLS will rely on review and analysis tools developed for the ARS and QBS to assist in decision making.

3. Describe methods to maximize response rates and to deal with issues of non-response. The accuracy and reliability of information collected must be shown to be adequate for intended uses. For collections based on sampling, a special justification must be provided for any collection that will not yield "reliable" data that can be generalized to the universe studied.

BLS expects a response rate of 25%. This expectation is based on responses to prior QBS pilot tests and results observed from the QBS in 2020, 2021, and 2022.

3a. Maximize Response Rates

To maximize response rates, all ARS units selected in the sample will be transitioned to the QBS questions once they complete the ARS. All ARS-ineligible sample units selected for the QBS will be contacted using established ARS contact methods, i.e., email and survey solicitation letters, and directed to a stand-alone collection system that displays the questions identically to the ARSWeb system. Nonresponse follow-up will be conducted per the current ARS procedures as outlined in materials for OMB Control No. 1220-0032.

BLS will rely on established, tested data collection processes to ensure maximum response rates to the QBS. BLS will make use of existing processes from the ARS and prior QBS surveys, along with consulting other established BLS survey programs to maximize efficiency and reduce burden.

All of the data collection will take place online via the BLS Internet Data Collection Facility (IDCF). This method of fully online data collection has been successfully employed and tested with both the ARS and the QBS and has been effective in reducing cost to the government, reducing respondent burden, and maximizing response rates.¹²³

3b. Non-Response Adjustment

BLS expects a response rate of 25 percent. However, this response rate will not be uniform across all strata. Additionally, some strata are smaller than others. Consequently, there will likely be strata with no usable survey responses. When building composite estimates that consist of “empty” strata, the empty strata will be imputed.

For strata imputation, survey strata and question combinations with no usable item response will have their establishment proportions and variances imputed according to an ordered hierarchy of related composite estimates.

Non-response bias adjustment methods may be assessed based on survey results. For example, another method to explore is to impute missing questionnaire responses using the response of the nearest responding neighbor. Final methods will be documented along with all of the other statistical design methods on the public webpage used for dissemination of the results.⁴

3c. Confidentiality

Before estimates are released to the public, they must first be screened to ensure that they do not violate the BLS confidentiality pledge. A promise is made by BLS to each private sector sample unit that its employment data will not be released to the public in a manner that would allow others to identify the unit. If an estimate fails a predetermined primary confidentiality threshold, such as the p% rule, then the cell can be protected. Whether this protection is suppression, rounding or other method is somewhat dependent on the objective of the survey. Secondary confidentiality protection is also implemented to protect respondent information at this level.

4. Describe any tests of procedures or methods to be undertaken. Testing is encouraged as an effective means of refining collections of information to minimize burden and improve utility. Tests must be approved if they call for answers to identical questions from 10 or more respondents. A proposed test or set of tests may be submitted for approval separately or in combination with the main collection of information.

The QBS is built on years of testing for the QBS and the ARS. This testing has informed the platform used, contact strategies, data processing methods, and expected data review time frames. In addition, prior to the fielding of the 2024 QBS survey, BLS will test the instrument to ensure that skip patterns are working as expected and data is being recorded correctly.

The content proposed for inclusion in the 2024 QBS survey was collected in the 2022 QBS.⁵ Two additional “screener” questions have been added to the survey for navigational purposes. Skip patterns are now programmed within the questionnaire, which will minimize respondent confusion by only showing respondents relevant questions based on their answers to prior questions and will improve data collection and editing efficiency by reducing the possibility of respondents providing conflicting responses. These recent updates to the QBS collection system implementing skip pattern navigation and the BBS stand-alone front end were tested in March 2023 under OMB Clearance (1220-0141).

5. Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze person(s) who will actually collect and/or analyze the information for the agency.

Mr. Ed Robison, Division Chief of the Statistical Methods Staff, Office of Employment and Unemployment Statistics, is responsible for the statistical aspects of this survey.

[email protected]

Mr. David Talan, Chief of the Division of Administrative Statistics and Labor Turnover, is responsible for the data collection and publication aspects of this survey.

[email protected]

1 Stang and Thomas “Web Collection in the Quarterly Census of Employment and Wages Program”, ICES-V, 2016. http://ww2.amstat.org/meetings/ices/2016/proceedings/072_ices15Final00299.pdf

2 Stang and Thomas “Email Solicitation for a Business Establishment Survey – Results from the 2015 Annual Refiling Survey”, JSM 2016. http://9004e5e16f4a25df17a0-290e28d0a6d5d71f78b4f59d5f323756.r86.cf1.rackcdn.com/ASA-JSM/pdf/389517.pdf

3 Stang and Thomas “Developing and Testing the Business Research Survey,” JSM 2018. https://ww2.amstat.org/meetings/jsm/2018/onlineprogram/AbstractDetails.cfm?abstractid=328621

4 https://www.bls.gov/brs/methods/2022-technical-notes.htm

5 https://www.bls.gov/brs/about/survey/

13