Supporting Statement Part B (1220-0134)_2019

5

Multiple Worksite Report/Report of Federal Employment and Wages

1220-0134

May 2019

Supporting Statement to Request for OMB Approval

Multiple Worksite Report and the Report of Federal Employment and Wages

B. Collection of Information Employing Statistical Methods

1a. Universe

Multiple Worksite Report - The universe for this program will include the multi-establishment employers subject to Unemployment Insurance coverage in the 50 States, District of Columbia, Puerto Rico, and the Virgin Islands. The employers covered by this census are those having more than one establishment under the same UI account number within the State and the sum of the employment in all of their secondary establishments is 10 or greater. The most current QCEW working file(s) for each State serves as the business list.

Report of Federal Employment and Wages/DOD Collection - The universe for this census will include the Federal employers subject to Unemployment Compensation for Federal Employees coverage in the 50 States, District of Columbia, Puerto Rico, and the Virgin Islands. The most current QCEW working file(s) for each State serves as the business list.

b. Sample Size

Multiple Worksite Report - Since this is a quarterly census of multiple worksite businesses, the sample size will be 185,229, the same as the universe. With a loss of 5% of selected units due to out-of-business, duplicates, or non-mailables, and a 15% non-response, the number of respondents is expected to be 148,200. This number, based on 1^st Quarter of 2018 information, may increase or decrease depending upon changes with the census of multi-establishment employers.

Report of Federal Employment and Wages - Since this is a quarterly census of Federal installations’ multiple worksites, the sample size will be 2,630 non-defense respondents as well as the 5 major civilian (appropriated and non-appropriated) defense respondents, the same as the universe.

2a. Sample Allocation and Selection Procedures

The sampling frame used by each State is the most current QCEW working file(s). All employers that report more than one establishment under the same UI account number within the State and where their employment for secondary establishments is greater than or equal to 10 will be included in the MWR census. All Federal employers within each State will be included in the RFEW census.

b. Estimation Procedure

These data are being collected to ensure the micro data on the longitudinal QCEW data and on the QCEW Report are on an establishment/installation basis instead of some consolidated level consisting of two or more establishments/installations. Therefore, after data are edited and reconciled for apparent inconsistencies and completeness, these data will be the micro data on the longitudinal QCEW data. Estimates of totals at higher tabulation levels for the QCEW Report will be the sum of the appropriate micro level data. Since this is a census, the estimates are not subject to any sampling error. Therefore, no standard errors will be calculated.

c. Accuracy

Because a census is being conducted, the estimates are not subject to any sampling error.

d. Problems

There are no unusual problems requiring specialized sampling procedures.

e. Frequency

This is conducted quarterly.

3a. Response

To maximize the response rate for the MWR and the RFEW, an explanation of the importance of the information, the need for employer cooperation, and the electronic reporting options are emphasized.

Additionally, States conduct follow-up mailings after the initial mailing and contact key non-respondents by telephone.

Employers whose total UI account employment within a State is less than 10 are not subject to this report.

b. Non-response Adjustment

A proration adjustment procedure is used to impute data for missing worksites when data are reported on the current UI contribution report but not the MWR. The data are prorated using the previous quarters’ distribution of employment and wages and the current UI contribution report for the master unit. When data are missing on both the contribution report and the MWR, data for the master unit are imputed using historical data and then prorated to the worksites. Going forward, when contribution report data are reported and worksite data are missing, proration will continue to be used. When data are missing on both the contribution report and the MWR, the cell ratio method is proposed for imputing data for missing worksites as explained in http://www.bls.gov/osmr/pdf/st090090.pdf. Data will then be aggregated from worksites to the master unit.

c. Reliability

Because this study is a census, no sampling errors are calculated.

To control non-sampling errors, quality control procedures were incorporated into the survey's design. These procedures include follow-up of all non-respondents and validation of all edit failures. Additionally, the States and Regions were given training and assistance in conducting this quarterly census. The automation of most survey operations reduced many sources of non-sampling errors. Quality control measures are detailed in http://www.bls.gov/ore/pdf/st050070.pdf.

4. Tests

There are no plans to conduct tests for this ongoing data collection activity.

5. Statistical Responsibility

Mr. Edwin L. Robison, Supervisory Mathematical Statistician, Statistical Methods Staff, Office of Employment and Unemployment Statistics is responsible for the statistical aspects of this survey.