UIPL announcing cessation of statistical methodology step in calculations

ADVISORY: UNEMPLOYMENT INSURANCE PROGRAM LETTER NO.

TO: STATE WORKFORCE AGENCIES

FROM: JANE OATES

Assistant Secretary

SUBJECT: Elimination of Sampling in the Preparation of the ETA 203 Report; Characteristics of the Insured Unemployed

1. Purpose. To advise states of the Department of Labor’s (Department’s) elimination of statistical sampling as a method for preparation of the ETA 203 report.

2. References. ET Handbook No. 401, Unemployment Insurance Reports Handbook; OMB collection number 1205-0009.

3. Background. The ETA 203 report provides information, by state and for the Nation, about the characteristics of Unemployment Insurance (UI) claimants. This data is useful in describing the population of claimants and determining how that population changes over time and under various conditions. It can also be compared with characteristic figures of the total unemployed as published by the Bureau of Labor Statistics (BLS).

4. Additional Guidance. The survey week for this report coincides with the survey week for the Bureau of Labor Statistics' Current Population Survey (CPS) week. This usually represents weeks claimed which are filed during the week containing the 19th of the month. The week containing the 19th of the month is assumed to reflect unemployment experienced during the prior week containing the 12th of the month, the CPS survey week.

States have been able to produce data suitable for submission of the ETA 203 report either by (1) summarizing all weeks claimed which are filed during the week containing the 19th of the month, or (2) by taking a sample of weeks claimed from the week containing the 19th of the month and inflating the resulting numbers back up to the population level. Moving forward, states may no longer use the sampling methodology for this report and must provide categorization of the entire universe of available data.

State UI systems operate with a high level of automation. The difficulties that arise in assuring statistically representative sampling and the possibility of additional error introduced through sampling are problems that are substantially greater than any potential benefit from sampling for the purposes of this report. In addition, sampling represents an additional step where states could easily compute summary statistics without the possibility of additional error.

5. Action Required. States should no longer produce their ETA 203 report by extracting a sample of their regular program continuing claims, analyzing that sample of data and then weighting the results back up to the level of the population. States should prepare the report using all weeks claimed from the week containing the 19th of the month

The sum of any given category, such as age groups, should generally be identical with the total for insured unemployed as reported on the ETA 539 report for state UI adjusted continued weeks claimed (CW) for the same week. Continued weeks claimed under the Extended Benefit (EB) program, the Emergency Unemployment Compensation program (EUC), State Additional Benefits Program (AB), or any other extended compensation program should not be included on the ETA 203.

6. Inquiries. All inquiries should be directed to the appropriate Regional Office.

7. Attachments. None.