1205-0431_Part_A_Justification(20110513)

1205-0431 JUSTIFICATION PART A

SUPPORTING STATEMENT FOR REQUEST FOR OMB APPROVAL

UNDER THE PAPERWORK REDUCTION ACT

PART A. JUSTIFICATION.

This is a justification for the Department of Labor's request for approval to extend the program that assesses the accuracy of reports data which States submit to the Department monthly, quarterly or annually pursuant to the Department’s authority in Section 303(a)(6) of the Social Security Act. This program is called the Unemployment Insurance Data Validation (UI DV) program. UI DV is operated pursuant to authority granted under OMB number 1205-0431; collection authority expires on July 31, 2011.

The UI DV program validates 332 report elements from 114 UI required reports to ensure that these key data, used for performance management purposes, are reliable. Appendix A-1 contains the relevant sections of the Social Security Act on which the reporting and validation activity is based.

A-1. Reasons for Data Collection

Section 303(a)(6) of the Social Security Act specifies that the Secretary of Labor will not certify State unemployment compensation programs to receive administrative grants unless the State’s law includes provision for: “...making of such reports...as the Secretary of Labor may from time to time require, and compliance with such provisions as the Secretary may from time to time find necessary to assure the correctness and verification of such reports.”

In the 1970s the Department discovered that the UI required reports data it was using to determine States’ administrative budget allocations were often reported inaccurately. In response, it developed the Workload Validation (WV) program to assess the accuracy of eleven key workload items used to allocate funds for UI administration among States. Under the WV program, States were required to validate 29 elements from four UI required reports. This validation activity was conducted using the instructions in the UI Workload Validation Handbook No. 361 approved for use through 12/31/2000 (OMB Number 1205-0055).

The 29 elements validated through the WV program constituted slightly over 1% of the approximately 2,400 elements States report on 47 UI required reports and only a few Government Performance and Results Act (GPRA) elements. In light of this, the General Accounting Office and the Department of Labor’s Office of Inspector General (OIG) recommended that the program be revised to cover more data elements associated with performance measures.

In the early 1990s as part of a project to develop and test new measures for the timeliness and quality of benefit payment operations, called the Performance Measurement Review (PMR), the Department asked the technical support contractor, Mathematica Policy Research (MPR), to develop a method for validating benefits data. MPR developed an approach based on WV concepts but more amenable to automation. It included nearly 1,200 data elements from 13 benefits reports. The methodology was subsequently extended to include the data reported on the primary tax performance report, ETA 581, “Contribution Operations” (OMB Control No. 1205-0178, expires 10/31/2011). After the method was tested in three states with satisfactory results, the Department had MPR develop software that all states could use to perform validation operations. States installed this software on individual state PC or Sequel Server computers. After completing data validation, results could be exported from the software in different formats (e.g. text, Excel, Word, PDF files) and sent to the National Office via email.

Although the MPR software represented an improvement from the original DV methodology, DOL recognized that it could be enhanced, and as a response, DOL began developing its own version in 2003. The DOL software, released in 2005, is the software that is currently used to validate reported data. It has the advantage that it resides on the Sun-Unix servers that states use for UI reporting operations, which facilitates the deployment of new software versions, allows transmission of validation results directly to the National Office UI database, and allows automatic retrieval of reported counts from the UI database instead of requiring validators to enter those data as in the previous software. The latest version, Version 4.1, was released in July 2010.

The Department required states to implement UI DV. It set July 31, 2003 as the target for states to install the common DV software and develop the extracts of benefits and tax transactions the software processes; the states were to submit reports containing DV results by September 30, 2003. As of June 2010, reports had been received from 48 of the 53 states and territories that operate UI programs. It is expected that the extent of validation will be higher by the next reporting deadline, June 10, 2011.

Unemployment Insurance Program Letter (UIPL) 22-05, available on the DV Web site at http://ows.doleta.gov/dv/, summarizes the UI DV policy framework. Data Validation is administered on a “Validation Year” basis. The Validation Year (VY) coincides with the State Quality Service Plan (SQSP) performance year, which runs from April 1 through the following March 31. For example, a validation of any report for UI activity that occurs between April 1, 2010 and March 31, 2011 is part of VY 2011 and is to be submitted by June 10, 2011. The timing allows validation results to be part of the SQSP process.

The DV program operates on a three-year cycle. All but three groups of transactions—called populations--that pass validation must be validated every third year. The other three populations—which validate report cells from which GPRA indicators are formulated--must be validated annually; and populations that fail validation must be revalidated the next year. UI DV has also adopted the WV parameters for validity: if report counts are within ± 2% of reconstructed (validation) counts, and the random samples show that no more than 5% of transactions in the extract file universe contain invalid data elements, the reported counts are considered valid. (The GPRA report counts are held to a ± 1% standard instead of ± 2%.) The rules for passing validation are also contained in UIPL 22-05. About half of all validation items due as of June 10, 2010, passed validation; about a third that were due were not submitted; and about one sixth were validated but failed validation. Nearly three quarters of items that states submitted passed validation.

A-2. Users, Purposes, and Consequences of Failure to Collect the Information

a. Users, Uses, and Purposes. Accurately reported data are essential for properly assessing State performance and for ensuring that States are treated equitably when administrative funds are distributed. The Department will use UI DV results to evaluate the accuracy of the data States report on selected UI required reports. This will enable it to assure its customers, partners and stakeholders of the accuracy of the performance measures that form the basis of the UI Performs system, allow the Congress to determine the extent the UI system is accomplishing the goals set for the GPRA Strategic and Annual Performance plans and also to ensure that UI administrative funds are allocated properly to the 53 State UI programs.

The Department will use the UI DV information as a performance measure in the State Quality Service Plan (SQSP) process. SQSP is a permanent framework for the annual planning and budget process in each State UI agency. As part of the SQSP process, states that inaccurately submit data on validated reports—from which, among other things, key UI Performs or GPRA measures are based--will be required to take corrective action to ensure accurate reporting.

Because the performance measures are intended to guide and assist States in improving their performance and service to their ultimate customers (workers and employers), States also have an interest in ensuring that the measures are correct. The DV system includes detailed guidance to State programmers for correctly constructing various report elements. This not only ensures they will construct the DV extract files correctly, but also will simplify their tasks when new tax or benefit system software is installed.

b. Consequences of Failure to Collect Data. As noted in A-1 above, the Department has been criticized for failing to validate the data on which key measures are based and for failing to increase the extent of the validation of the reports data it receives from States. The UI DV system allows the Department to respond positively to those criticisms and to ensure the accuracy of the reports data on which it relies to discharge its oversight responsibilities.

A-3. Technology and Obstacles Affecting Reporting Burden

In compliance with the Government Paperwork Elimination Act, beginning on April 1, 2005, the UI DV program has had States submit UI DV findings via the system now used for electronic transmission of UI reports.

The Department has provided each State UI agency with an automated menu-driven system to input, store, and transmit reports data. The UI DV results are entered electronically as a report and stored in the UI Data Base.

The UI DV system is highly automated. The State-specific UI DV handbooks give States detailed specifications for developing data extract files, based on their own State management information systems. The States also enter or download these extract files to the automated system where a software module manipulates the input files for validation by producing validation displays, sorts, reconstruction counts, and duplicate detection tests, and drawing various samples. The software module extracts from the state’s report database the report elements being validated so that validators do not need to enter those counts. The software also provides input interfaces where validators enter the findings of their reviews of quality samples. The results of completed validations are reported electronically to the Department using the overnight pick-up mechanism used to transmit other UI reports data.

In the last year the Department has increased DV automation. DV summary results from the DOL database are available to Department and state users through a Web-based system. A separate Web-based system allows state users access to a database to maintain Module 3 of the benefits and tax handbooks. Module 3 is the “map” that enables the states to relate their key data to Federal reporting definitions. It gives Federal definitions from reporting and validation requirements; and based on those requirements, states note where these elements are to be found in their management information systems. State programmers use Module 3 to find the data elements needed to build DV extract files; validators use it to test the extract files when they examine sampled records during the data element validation phase.

The Department knows of no technical obstacles to implementing and operating UI DV.

A-4. Duplication

The proposed UI DV program does not duplicate any existing Department program. The Department believes that there is no satisfactory alternative to the UI DV methodology for this purpose.

A-5. Burden on Small Business or Other Small Entities

There will be no impact on small businesses or any entities other than State UI agencies.

A-6. Consequences of Less Frequent Data Collection

The Department requires that States conduct a complete validation every third year and a partial validation every year for those items determined to be reported inaccurately by a previous year’s validation or that are used for GPRA indicators. A validation is required within a year after states change State reports-generating software, such as a new state management information system or the Department changes reporting requirements.

A-7. Special Circumstances Involved in Collection of UI DV Data

This request is consistent with 5 CFR 1320.5.

A-8. Preclearance Notice and Responses

In accordance with the Paperwork Reduction Act of 1995, the public was given 60 days to review and comment through a Federal Register notice published on March 2, 2011 (Vol. 76, p 11514). No comments were received.

A-9. Payments to Respondents

This program does not involve payments per se to respondents. However, because UI DV

is a mandated data collection, the Department provides funding to the participating States, who are technically listed as the "respondents" for purposes of the Paperwork Reduction Act. States undertake validation activities out of their UI administrative grant funds, the same as they did for workload validation.

A-10. Confidentiality

Confidentiality is not an issue with this program, which simply involves verifying the accuracy of aggregated data counts, submitted earlier on required reports. States submit no individual records with personal (e.g. Social Security Numbers) or business (e.g., Employer Account Number) identifiers.

A-11. Questions of a Sensitive Nature

The data collection includes no questions of a sensitive nature.

A-12. Respondents' Burden and Cost of Collecting Information

1. Annual Hour Burden: 29,150 hours

• Number of respondents: 53 State Employment Security Agencies

• Burden for a full validation: 900 hours

  • Based on estimates received from four states that have conducted validations; they assumed they were conducting repeat validations using the new validation software.

• Annual Estimate Assumes:

  • States conduct a full validation in year 1;

  • In year 2, they validate half of report cells (i.e., GPRA elements and some failures from full validation, based on initial validation results that show at least half of report items failed);

  • In year 3, they validate one third of report cells in third year (i.e., the GPRA populations and some continued failures from validation in year 2);

  • These assumptions may be conservative; degree of improvement from one validation to another is not known. Burden is expected to decline in the future as report validity improves due to fewer report cells to be revalidated and more random samples passing at the first stage.

• Calculation of Average Annual Burden

  • Full validation = 53*900 = 47,700 hours (Year 1)

  • Half validation = 53*900/2 = 23,850 (Year 2)

  • Third validation = 53*900/3 = 15,900 (Year 3)

  • 3-Year Average = (47,700 + 23,850 + 15,900)/3 = 29,150

2. Cost of Annual Hour Burden to Respondents: $1,189,612

• Annual Burden = 29,150 hours

• Average Wage Rate for State UI agencies: $40.81 = average salary used for UI budget formulation purposes for FY 2011 = $69,785; average yearly work hours used is 1,710)

• Annual Cost = 29,150 hours * $40.81 /hour = $1,189,612

A-13. Annual Cost to Respondents

The States should not incur any capital or start-up costs to be added to the staff costs reported in A-12. The Department provides States with the computer equipment necessary for retrieving, manipulating, storing, and reporting the validation results. We assume that nearly all states will have implemented DV by 6/10/2011

A-14. Federal Cost

Federal costs are the staff and contractor costs required to maintain and manage UI DV. These include the costs of providing technical assistance to states; monitoring the validation process in the states; maintaining the DV handbooks and the validation software to reflect changes in reporting; improving the functionality of the UI DV state software and programming the Federal software used to produce reports on DV results; and analyzing DV results and using those results to assess state reporting operations and making sure that states continue to improve reporting accuracy. These costs are expected to cost average approximately $380,000 per year over the 2011-2014 period. Federal allocations of funds for State UI administration will also cover the costs in A-12.

Category	Estimated Costs
Staff:
DOL ETA National Office GS-12 Step 5 (2.0 SY)	$136,620
DOL ETA Regional Office GS-12 Step 5 (1.2 SY)	$81,972
Programming/ADP Support	$120,000
Contractor Technical Validation Support (3 Webinars at $3,000)	$9,000
Total Costs	$380,682

A-15. Reasons for Program Change and Change in Burden

This submission assumes that there is no change in burden.

A-16. Publication Information

The Department publishes results of validations on its Data Validation Web site at http://ows.doleta.gov/dv/.

A-17. Reasons for Not Displaying Date OMB Approval Expires

The Department will display approval information in the ETA Handbook 361.

A-18. Exceptions to Certification

There are no exceptions to the certification statement in 83-I.