Sampling Procedures 5 8 2012

Sampling Procedures for Emergency Unemployment Compensation (EUC) Claims

Work Search Audits

EUC Payment File

1. Create a file of all EUC claims paid during the week beginning 12:00 midnight on Sunday and ending 11:59 pm Saturday. The minimum data elements in the file are:

• Claimant’s Social Security Number (SSN);

• Amount paid to the claimant (must be >= $1); and

• Week ending date of the week in which the agency issued the payment (not the week ending date of the week claimed). Format is MM/DD/YYYY.

• State may include additional data elements for control or identification (optional).

2. Sort the file created in step 1 by 1) amount paid (ascending), and 2) the last four digits of the claimant’s SSN (ascending).

3. Assign a case number from 1 to the total number of records in the sorted file (N).

Spreadsheet Sampling Tool

4. The Department of Labor has developed a spreadsheet sampling tool. It is designed to be used in conjunction with the file created in steps 1 to 3. The file created in steps 1 to 3 is external to the spreadsheet; that is, the cases are not imported into the spreadsheet.

• Enter the week ending date for the payment file created in steps 1 to 3 using the drop-down menu.

• Enter the number of records in the file created in steps 1 to 3 in column B on the row “Enter Population”. This will equal N (from step 3).

• Enter the sample size between 50 and 1,500 cases. If the number of records in the file created in steps 1 to 3 is less than 50, audit all the records.

5. The spreadsheet will retrieve the random number for the week ending date, calculate the skip interval, and display the case numbers of the EUC payments randomly selected.

6. Cases are selected using systematic selection.

• A skip interval (i) = N / n is calculated, where N is the number of records in the EUC payment file created in step 1 and n is the requested sample size.

• The first case (n₁) is selected by multiplying the skip interval (i) by the random number (r); the result is rounded to the nearest integer.

• The next (n-1) cases are selected by: [n₁ + (j x i)], where j = 1, 2, …, (n-1); the results are rounded to the nearest integer.

7. Query the file created in steps 1 to 3 and select the records corresponding to the case numbers selected by the spreadsheet in step 5 (in the column labeled “Case Number”). These are the cases you will audit.