U.S. DEPARTMENT OF LABOR
Employment and Training Administration
Washington, D. C. 20210
April 13, 1990
April 30, 1991
UNEMPLOYMENT INSURANCE PROGRAM LETTER NO. 24-90
ALL STATE EMPLOYMENT SECURITY AGENCIES
DONALD J. KULICK
Procedures for Release of Unemployment Insurance Benefits Quality Control (QC) Data
Purpose. To provide State Employment Security Agencies SESAs) with guidelines for the annual release of Unemployment Insurance (UI) Benefits Quality Control (BQC) program data for Calendar Year (CY) 1989.
References. 52 FR 33520, September 3, 1987; and Benefits Quality Control State Operations Handbook, ET Handbook 395.
Background. The final rule to establish the Unemployment Insurance Quality Control program at 20 CFR Part 602 was published in the Federal Register at 52 FR 33520 on September 3, 1987. Effective October 5, 1987, QC programs in the 50 States, the District of Columbia, and Puerto Rico selected weekly random samples of UI benefit payments for investigation under the mandatory program.
Section 602.21(g) of 20 CPR provides that each State shall:
Release the results of the QC program at the same time each year, providing calendar year results using a standardized format to present the data as prescribed by the Department; States will have the opportunity to release this information prior to any release by. the Department.
The notice of procedures for the release of benefits Quality Control data for the Unemployment Insurance program was published in the Federal Register on March 26, 1990, at 54 FR 11112.
Date of Record. The Department has established c.o.b. April 27, 1990, as the date of record for State databases for computation of the error rates for the annual report of QC data. All information in the annual report will be based on CY 1989 QC cases that have been signed off by the QC supervisor by c.o.b. April 27, 1990. States must not update any CY 1989 QC cases after this date until they have reviewed the computations of the error rates and have reconciled any differences. Any data analysis supporting the annual report must be run on the database as it exists at c.o.b. on April 27. If States make any further changes, the data would differ from the data that the National Office (NO) will use to compute final results for CY 1989 and any comparisons cannot be accomplished.
Annual Report Software. The current software for the annual report will not change in regard to the computation of rates. The only changes are minor wording changes to the footnotes. States may run the rates as soon as all cases are completed. The date of record will he c.o.b. April 27, 1990.
The NO will produce an annual report for each State, based on the QC cases as of c.o.b. April 27, 1990.
Annual Report Format. The annual report format is shown in Attachment 1 It is divided into three main sections.
Quality Control Data
Total Dollars Paid in Population
Total UI benefits paid to the population of UI claimants who comprise the sampling frames for all weeks in CY 1989 for which the State pulled a QC sample.
Total UI payments selected during CY 1989 (QC batches 8901 through 8952) and completed (QC supervisor sign-off) by c.o.b. April 27, 1990. This is the number of QC sample cases from which the rates and sampling errors (confidence intervals) are estimated.
The weighted ratio estimate of total dollars properly paid to total dollars paid expressed as a percentage.
The weighted ratio estimate of total dollars overpaid to total dollars paid expressed as a percentage.
The weighted ratio estimate of total dollars underpaid to tota dollars paid expressed as a percentage.
95% Confidence Interval
A confidence interval, expressed as +/- a percentage, will be constructed for each of the three estimated rates. The actual rate is expected to lie within 95 percent of the intervals constructed from repeated samples of the same size and selected in the same manner as the QC sample.
The following footnotes will appear on State QC annual reports to describe certain conditions that affect the data:
Claimants failing to conduct required work search were given formal warnings and no overpayment was established. The proper payment rate would be lower and the overpayment rate would be higher if these cases were counted as erroneous payments.
Condition: Any case with key week action code '14' in fields hi2, hj2, or hk2 of the CY 1989 QC database.
Percentages apply to less than a complete year of UI payments due to the State not pulling a sample for xx weeks.
Condition: There is no comparison report for one or more QC batches 8901 through 8952. (The CY 1989 summary of State QC sample selection is provided in Attachment 2.)
The precision of the percentage estimates is reduced because the State selected samples that were below the minimum prescribed levels for xx weeks.
Condition: State selected one or more QC weekly samples below the minimum level prescribed in Benefits Quality Control State Operations Handbook, ET Handbook 395, chapter III, p. 2.7. (See CY 1989 summary of State QC sample selection in Attachment 2.)
Percentages based on data collection procedures that were not completely in accordance with QC methodology prescribed in ET Handbook No. 395.
Condition: Based on QC monitoring conducted by the Department of Labor Regional Offices and. National Office.
The State completed xx% of the cases within 90 days. The QC program standard is 95% completed within 90 days.
Condition: State failed to meet case completion objectives established in Benefits Quality Control State Operations Handbook, ET Handbook 395, chapter VI, p. 6.
xx.x% of the sample cases were not completed when this report was prepared. The QC program requirement is that no more than 2% of the cases for the year remain incomplete.
Condition: Based on the number of QC cases that were not completed (QC supervisor sign-off) by c.o.b. April 27, 1990, divided by the number of cases selected for QC weekly samples during CY 1989 (QC batches 8901 through 8953).
States are invited to supply a narrative analysis that will accompany their annual report. The Department will include this narrative in the digest of State annual reports that it will prepare (see section 10, below).
The narrative may include, but is not limited to, a discussion of State laws and how they affect error rates, an explanation of circumstances that led to some degree of inaccuracy in the reported error rates, and proposed or actual corrective action strategies.
The Department prefers that the narrative be submitted on a diskette in WORDPERFECT (versions 4.2 or 5.0) or LEX. If it is not possible to submit a diskette, the narrative may be submitted on paper. If not submitted on a diskette, the narrative must be submitted on 8 ½" x 11" white bond (at least 20# weight), using letter quality printing that can be successfully photocopied. The narrative must not exceed two single-spaced pages. To ensure uniformity in the Federal digest, the following format (see Attachment 3) is required for both diskette and paper submissions:
|Margins||--||one inch all around|
|Title||--||"ANNUAL REPORT NARRATIVE" typed in capital letters, and centered|
|Subtitle||--||"STATE NAME" typed in capital letters, and centered on the line below the title|
|Date||--||"CY 1989" centered on line below the subtitle|
|Paging||--||"STATE POSTAL CODE-Page #" right justified one inch from the bottom of the page (i.e., AK-1)||Spacing||--||The text should follow three lines below the subtitle and end with a maximum length of. three lines above the page number. Use six lines per inch.|
State narratives must: be received in the NO no later than June 8, 1990. Narratives should be sent to the attention of Susan Makara at:
U. S. Department of Labor
200 Constitution Ave., NW
Attn: TEUQS, Room S-4015
Washington, DC 20210
If the State does not. plan to submit a narrative, the State should notify the NO at the above address by June R, 1990 that no narrative will be submitted.
Supplemental Data. In order for the SESAs to interpret the data in the annual report and provide narrative explanations of the data, the Department recognizes the need for additional information on. cause and responsibility. The software that produces the QC annual report will also generate a separate report consisting of up to 16 responsibility categories and up to 6 cause categories. If the categories are equal to "0" the item will not be displayed in the table. Responsibility and cause data will be reported as percentages of UI dollars overpaid. The format for this report is shown in Attachment 4.
In order to provide States with the greatest flexibility in interpreting the data, all possible responsibility codes and broad cause categories can be generated. States can combine categories that contain few sample cases, where appropriate. No sampling errors will be computed because the publication of these percentage rates is not mandatory. States should be aware that many of these percentages are based on small numbers of QC sample cases and may have relatively large sampling errors. The Department may publish such data in the technical appendix, along with additional analysis where appropriate.
Comment Period. QC annual reports. produced by the Department of Labor will be transmitted to the States on May 11, 1990. Each State should review its annual report and if there are any comments submit them to the NO (Attn: TEUQS, Andy Spisak). SESA comments will be discussed and reconciled by June 8, 1990.
Public Release by SESA. As stated in the Federal Register notice of March 26, 1990:
Each State will release the required data annually through established channels for disseminating State performance data. In accordance with established State procedures, the data must be provided to those who normally receive performance/evaluation data and to anyone else who requests it. It is no longer necessary to publish a notice of availability.
States must release their annual reports by June 29, 1990. If a State fails to release data in accordance with these procedures, the Department, in its annual review of State QC operations as specified in 20 CFR, Part 502, Section 602.31, will take appropriate action.
Federal Release of Data. The Department will announce the availability of QC program data for CY 1989 through a notice published in the Federal Register on or about July 31, 1990. The Department's publication will be a digest prefaced by background information on the Quality Control program, and a brief, discussion of how the error rate estimates were computed. Readers will be discouraged from comparing results among States. Each State's section, including its annual report and narrative, will be displayed separately. Further, the report may contain other information that the Department feels is pertinent. State sections will be ordered alphabetically to discourage ranking and comparisons.
If the Department determines that a State has failed. to follow the prescribed QC methodology--in terms of sample selection, data collection methodology, or case completion timeliness--to such an extent that the data are not reliable, the Department will publish an explanation of the major deficiency in lieu of an annual report for that State.
Technical Specifications. The procedures and statistical formulas used to produce the data in the annual report and the supplementary data on responsibility and cause are available from the Department of Labor upon request through the appropriate Regional office.
Key Dates. The following is a summary of key dates for the public release of CY 1989 QC data.
|April 27||Date of record for State QC databases.|
|April 30||Department of Labor provides QC annual report software to States.|
|May 1 - 13||Department of Labor generates annual reports for all States.|
|May 11||Department of Labor sends annual reports t4 the States.|
|May 15 - 31||States review annual report.|
|June 1||States' comments are received by the Department of Labor.|
|June 1 - 8||SESA comments discussed and reconciled.|
|June 8||SESA Narratives due to the Department of Labor; Notification to the Department that no narrative will be submitted.|
|June 29||States release CY 1989 QC annual report.|
|July 31||Department of Labor publishes notice in the Federal Register announcing the availability of, QC-data.|
Action Required. State Administrators are requested to:
(a) provide copies of these guidelines to the appropriate staff; and
(b) ensure that the following information is supplied to ETA Regional Administrators:
(1) the name, location, and telephone number of a contact person who interested parties may write or call for State-specific interpretations by June 1, 1990;
(2) a copy of the State Annual QC data release by June 29, 1990;
(3) copies of any press releases and/or newspaper articles relating to the annual report released by the SESA for CY 1989.
Inquiries. Questions should be directed to the appropriate Regional Office.
Attachments. (1) Annual Report Format; (2) State QC Sample Selection by Batch for CY 1989; (3) Format for QC Annual Report Narrative; and (4) Format for QC Annual Report Supplemental Data