Questions and answers on denied-claim accuracy measurement

U.S. Department of Labor
Employment and Training Administration
Washington, D.C. 20210
Classification: UI
Correspondence symbol: OWS/OCTA/DPM
Date: July 27, 2001

DIRECTIVE: UNEMPLOYMENT INSURANCE PROGRAM LETTER NO. 45-01

TO: STATE EMPLOYMENT SECURITY AGENCIES

FROM: GRACE A. KILBANE
Administrator
Office of Workforce Security

SUBJECT: Questions and Answers for Unemployment Insurance Benefit Accuracy Measurement (BAM) Denied Claims Accuracy (DCA)

RESCISSIONS: None
EXPIRATION DATE: Continuing

1. Purpose. To transmit BAM DCA Questions and Answers to State agencies.

2. References. Unemployment Insurance Program Letter (UIPL) No. 28-01 (April 10, 2001); UIPL No. 34-01 (June 15, 2001); and Benefit Accuracy Measurement State Operations Handbook, ET Handbook No. 395, 3rd edition.

3. Background. UIPL No. 28-01 informed States of the Office of Management and Budget’s approval of the Department of Labor’s request to modify the BAM information collection program to collect data on the accuracy of denied claims for unemployment compensation and to include interstate liable claims in BAM paid claims accuracy (PCA) samples. UIPL No. 34-01 notified States that the implementation of DCA and the inclusion of paid interstate liable claims in PCA samples will be postponed until July 29, 2001 -- BAM batch (sampling week) 200131.

The Benefit Accuracy Measurement State Operations Handbook (ET Handbook No. 395, 3rd edition) is the principal resource for the implementation and operation of BAM DCA. In order to facilitate implementation, the Department of Labor has summarized answers to questions that were raised at the DCA training conferences, which were held in Washington, DC, (April 11-13, 2001) and San Diego, CA, (May 23-25, 2001) or were submitted to the Department of Labor by State agencies.

4. Action Required. State Administrators are requested to provide copies of the BAM DCA Questions and Answers to the appropriate staff.

5. Inquiries. Questions should be directed to the appropriate ETA Regional Office.

6. Attachment. BAM DCA Questions and Answers.

Attachment: BAM Denied Claims Accuracy (DCA) Q’s & A’s

Official attachment: https://www.dol.gov/sites/dolgov/files/ETA/advisories/UIPL/2001/UIPL45-01Attach.pdf

Sampling and Software Issues

1. What is maximum weekly, monthly, and quarterly sample pull? What is the minimum?

A. There is no maximum. BAM supervisors may change the weekly sample sizes to accommodate investigator vacation schedules or other staffing contingencies. However, States are expected to pull at least two (2) cases each week of each sample type.

2. Do States have to pull samples of each type of denial each week? Under what circumstances can States defer sampling?

A. States will normally pull samples of each type of denial each week. However, States may defer sampling of one or more types of denials for a batch due to staffing contingencies, such as unexpected medical leave or staffing vacancies. States that plan to suspend sampling should contact their Regional Office and inform them in advance of such plans, along with the reason for the suspension, length of suspension, and plans to compensate in subsequent sample selections to insure that the minimum number of annual cases (150 of each type of denial) is selected and investigated.

States cannot sample the annual minimum number of cases before the end of the calendar year and then not pull any samples for several weeks.

3. Why is the sampling of monetary denials delayed by two weeks?

A. In the pilot test of DCA, all denials that were issued during the sampling week (batch) were included in the sampling frame. This was consistent with the BAM paid claims methodology and was intended to improve the chances of obtaining information from the denied claimant. However, the results of the DCA pilot indicated that a two-week delay would improve the identification of eligible and denied claims involving UCFE, UCX and out-of-State combined wage claim (CWC) credits. Such claims were often initially denied in the sampling week, but were redetermined to be eligible when the requested wage credits were subsequently received.

The BAM population edit and sample selection COBOL program has been revised to allow extra time to obtain wage credits for monetary denials. The edit in the COBOL program will: 1) check that the date that the denial was issued (transaction date) is no earlier than 14 days prior to the beginning date of the batch and no later than 14 days prior to the ending date of the batch; and 2) check that the claim date is on or before the date that the denial was issued (transaction date).

4. Should denials resulting from an appeals decision that denies benefits be included in the sampling frames?

A. No. The sampling frames do not include any appeals decisions. The sampling frames are constructed earlier in the history of each claim, when the agency initially determines that the claimant is ineligible for unemployment compensation (UC). Sampling frame definitions are included in Unemployment Insurance Program Letter (UIPL) No. 10-01 (November 23, 2000) and UIPL No. 28-01 (April 10, 2001).

An appeals decision denying benefits either reverses a previous finding of eligibility, in which case the claimant was eligible for inclusion in the paid claims sampling frame, or affirms a previous denial, in which case the claimant was eligible for inclusion in the appropriate denied claims sampling frame.

5. Will report-writing software be included in the Denied Claims Accuracy (DCA) software?

A. The initial release of the DCA software will not include report-writing software. However, two Informix ACE report programs will be distributed. These will allow States to produce a DCA accuracy rate report and a case completion / time lapse report. States are also urged to use the Informix SQL and ACE software to produce State-specific reports. Future releases of the DCA software will include generic reports.

6. Are monetary denials as a result of claims being set up under the wrong SSN included in sample?

A. Yes, such claims should be included in the sampling frames and should be investigated if selected in the sample. The DCA audit could uncover information related to fraudulent use of SSNs or inadequate procedures for reporting or recording wage data.

7. The batch sequence numbers assigned in the sample selection process do not distinguish between monetary, separation, and nonseparation denials; each may have sequence numbers of ‘1’, ‘2’, etc. How are denial cases uniquely identified in the database?

A. Queries to the DCA database must include the batch, sample type, and sequence number to uniquely identify denial cases. The serial number and case type used in BAM paid claims is not used in denials.

8. How should a nonmonetary determination that was erroneously included in the sampling frame be processed? (Nonmons that do not meet the definition of a countable nonmonetary determination under ET Handbook 401.)

A. As with any sample case that does not meet the definition for inclusion in the BAM sampling frame, the Program field will be coded ‘9’, to denote a “deleted” record. The State agency will have to contact the Department of Labor National Office via e-mail (Andy Spisak, [email protected], or Julius Green, [email protected]) to request that the record be coded ‘9’ (deleted). States are strongly urged to eliminate such records before the UI transactions file is constructed. Erroneous denials will be credited for the State’s sample allocation; however, if the number of erroneous denials exceeds five percent of the sampled cases, States will be required to oversample in order to insure that an adequate number of valid denials is available for DCA investigation.

9. If a State that considers alternative base periods (ABP) in establishing monetary eligibility issues multiple denials on the same claim, how should these be processed for sampling?

A. Per UIPL No. 28-01 (April 10, 2001), “If the State agency issues more than one determination that denies eligibility for monetary reasons on the same claim during the week for which transactions are being assembled, only the most recent determination should be included in the UI transactions file for that week.” By representing such cases as a single record, the claim has the same chance of being selected for the sample as claims for which a single denial determination is issued. It is possible that some claimants who were initially denied will become eligible under one of the ABP formulas during the two-week lag in sampling monetary denials.

10. UIPL No. 10-01 refers to the rec1.dat file consisting of the BAM paid and denied sample cases and the sfsum.dat file consisting of the sample and population characteristics. Are these rec1.dat and sfsum.dat files created for DCA only? Or, are the rec1.dat file and the sfsum.dat file referred to in UIPL 10-01 to be used to select both the BAM paid claims and DCA samples?

A. The rec1.dat and sfsum.dat files include both paid and denied claims. The Sample Indicator field (pos. 168) identifies the four different sample types in the rec1.dat file, and the Sample Type field (pos. 7) does the same in sfsum.dat. The Case Conversion module is executed from the BAM menu to load the paid claims, and the Case Conversion module on the DCA menu is executed to load the denied claims.

11. Will there continue to be separate data collection instruments (DCI) for BAM paid claims and DCA or will a single DCI be used for both paid and denied claims?

A. Currently, the paid and denied claims data are stored in separate tables in the UI database with different data collection instruments. Eventually (probably concurrent with the conversion to Web-based data entry software), a single database with a common set of data elements will be created for both paid and denied claims.

Coding and Investigation Issues

12. Claimants for benefits under income-support programs are sometimes required to file a claim for UC. How should these “pro-forma” claims be coded?

A. Basic case control and claimant information (gender, ethnicity/race, age, etc.) should be coded, as well as all of the data elements relative to the claim (base period wages, high quarter wages, weeks worked, if any). States may use the State option field in the Reason for Monetary Denial (Before / After Investigation) to identify these cases, for example by using codes 51 to 59. The Action Code Flag should be coded as appropriate. If States pick up a significant number of these claims in their sample, they may increase their sample size to obtain additional denials that are not for “pro-forma” claims.

13. Can the agency request the claimant’s e-mail address on the claimant questionnaire? Can the claimant be contacted via an e-mail address?

A. Yes. Under the alternative data collection methods, a State agency may use e-mail, fax, or other electronic means (for example Internet) to collect information from all of the relevant parties.

14. If the Benefits Rights Interview is given by telephone or through the Internet, how should this be coded?

A. Use code ‘4’: video / electronic / other multimedia.

15. How should the Dollar Amount of Error be coded in the Error Issue table?

A. Per ET Handbook 395, p. VIII-A-78:

! If a specific number of weeks were denied (Weeks Denied Before Investigation is greater than 0 and not equal to 99, indefinite denial), the amount of error equals the number of weeks erroneously denied times the claimant’s weekly benefit amount that should have applied after the DCA investigation (WBA After Investigation).

! For determinations that resulted in an indefinite period of disqualification (Weeks Denied Before Investigation equals 99, indefinite denial), the dollar amount of error can be coded zero (0). Optionally, the State can code the dollar amount of error estimated for the indefinite period affected by the erroneous denial (for example, the claimant’s maximum benefit amount that should have applied after the DCA investigation or the claimant’s weekly benefit amount that should have applied after the DCA investigation times average duration for the State).

Although States have options, States must be consistent in the way in which they code this data element, both among investigators and over time.

16. In a combined wage claim (CWC), if out-of-State wages are insufficient to establish eligibility when combined with in-State wages, do we code these wages?

A. Yes, even if wages are still insufficient to establish eligibility, they will be coded in the Base Period Wages After Investigation data element.

17. When should Method of Filing Week Claimed be coded ‘0’?

A. This code is used only when Claim Type is ‘0’ (No week claimed).

18. If there are multiple weeks claimed and different methods of filing were used, what code do you use?

A. Method of Filing Week Claimed should be coded to reflect the most recent (latest) week affected by the denial. As a general principle, when multiple weeks are involved, coding should reflect the most recent (latest) week.

19. For nonseparation denials, if claimant is reporting bi-weekly, what date should be reflected as claim date?

A. Code the most recent (latest) week affected. The UI transactions file should contain separate records if one week was allowed, and the other denied.

20. If several nonseparation weeks are denied, what date is used for claim date?

A. Use the date of the last week denied prior to the week ending date of the batch in which the sample is selected.

21. What should be done with UCX separation denials (dishonorable discharge)?

A. These should be investigated like any other separation denial.

22. Should high-quarter (HQ) wages be coded if the State doesn’t use them in computing monetary eligibility?

A. No. If the State does not use HQ wages to determine monetary eligibility, these data elements can be coded ‘-2’, not applicable. The same applies to the number of weeks worked in the base period.

23. Is there a mandatory rounding procedure, or may each State follow its usual practice?

A. States should follow State practice but that practice must be consistent among investigators and over time.

24. Should a monetary redetermination be coded based only on State agency action exclusive of DCA or should it be coded based on DCA outcome?

A. Consider only monetary redeterminations conducted by the State agency independent of the DCA investigation. Do not consider monetary redeterminations conducted by the State agency as a result of the DCA investigation. The monetary redetermination data element should be coded ‘Yes’ to indicate that the State agency redetermined claimant's monetary eligibility prior to or during the course of the DCA investigation.

25. If a case is reopened due to appeal action, is the DCA time lapse affected?

A. Reopen code ‘4’, used when an appeals decision requires changes on a closed case, will not affect DCA time lapse.

26. Is denial of dependent’s allowances required to be in writing with appeal rights?

A. Yes, all denials of eligibility for unemployment compensation must be provided to the claimant in writing with the right to appeal the determination.

27. If, despite the two-week lag in sampling monetary denials, a case is selected which DCA finds is now eligible due to receipt of out-of-state, UCFE, and/or UCX wage credits, how should this be coded?

A. These cases should be coded ‘0’ in the Action Code Flag data element (Monetary eligibility established upon receipt of CWC, UCFE, and/or UCX wage credits; claimant eligibility initially denied). These cases will not be included in the estimation of the DCA accuracy rate.

28. When a claim is withdrawn after it has been denied and selected for the sample, how should that denial be investigated and reported?

A. Some State laws allow a claimant to withdraw a claim following a monetary denial and file the claim in another State, where the claimant has wage credits, in hopes of qualifying under that State’s eligibility requirements. These cases should be coded ‘8’ in the Action Code Flag data element. No investigation is required for these cases, but the current version of the DCA Update Cases software requires all data element fields to be coded in order to close out the case. See question #55 for a discussion of data element coding requirements.

The edit criteria in the Update Cases software will be modified in a future software release to allow States to close out these cases without having to code as many data elements.

29. When a denial has been selected for a week in which partial payment is allowed - as in the case of a deduction for receipt of Social Security benefits - how do you code the number of weeks denied when that number is variable depending on dollars paid?

A. This would be coded 99 (indefinite).

30. When an appeal is withdrawn, how is the appeal information coded?

A. A determination in which the appeal is withdrawn should be coded ‘0’ (denial not appealed) which is the correct status after the withdrawal.

31. When an appeal is rejected by the appeals board for lack of jurisdiction (as in late filing), how is that coded?

A. An appeal formally rejected should be coded as appeal decision issued. The rejection is the decision in this case, and the Results of Appeal of Initial Determination should be coded ‘2’ (affirmed, ineligible).

32. When a denial decision is determined to be wrong, but the claimant should still have been denied for another reason, how is this coded?

A. These cases will be coded ‘30’ in the Action data element in the Error Issue table. The purpose of DCA is to see if agency determinations (not just results) are correct. In this instance, the denial decision is erroneous, and should be coded as an improper denial. The fact that the claimant may have been denied for another reason is not material to the fact that an improper denial was issued for the case selected for review.

This code should be used, for example, if the DCA investigation establishes that the set of facts that the original determination was based on is complete and accurate, but the agency issued the denial under the wrong section of State law (they misinterpreted the issue or the provision of law governing it).

33. If more than one nonseparation determination denying benefits has been issued (for example, pension and work search) for a given week claimed, which determination is investigated?

A. When assembling the batch transaction file for sample selection, separate records should be created for each issue. DCA will investigate the issue selected. If an agency issues separate determinations that deny eligibility based on more than one nonmonetary nonseparation issue, States should distinguish these separate records in the UI transactions file by the transaction date, if the determinations were issued on different dates. If the determinations were issued on the same date, States can use positions 72-80 in the UI transactions file record for the nonmonetary nonseparation issue code. This code should be mapped to positions 163-164 (reason for nonmonetary nonseparation denial) in the rec1.dat file.

34. The concept of “denial period” seems to be fluid depending on the type of denial being investigated. Is there a workable definition of “denial period” for all denials?

A. Denial period is defined by denial type and State law.

! For Monetary Denials, the period should start with the effective date of the claim and be indefinite (code ‘99’).

! For Separation Denials, the period should not extend back beyond the benefit year beginning date (or the date of additional claim if a benefit year separation issue) of the claim under investigation, and be either indefinite or fixed, depending on State law.

! For Nonmonetary Nonseparation Denials, the period should start with the applicable week claimed, and be either indefinite or fixed, depending on State law.

35. Should the number of weeks denied before and after investigation include unclaimed weeks prior to the denial?

A. No. The number of weeks denied should be a function of the denial period as defined above (32). It should however, include unclaimed weeks subsequent to the denial that would be expected to be claimed but for the denial.

36. For monetary denials due to a lack of sufficient qualifying employment, would the number of weeks denied (before and after, if no change) be coded ‘99’ (indefinite disqualification)?

A Yes. This would be appropriate coding to reflect the effect of the denial for this category.

37. Can code ‘0’ (no week claimed) for Claim Type be used for monetary denials? (When a claimant is monetarily ineligible and does not file a weekly certification.)

A. The ‘0’ code for Claim Type cannot be used for monetary denials. Claim Type for a monetary denial needs to reflect the appropriate code (‘1’ - new initial, ‘3’ - transitional). Edits in the Update Cases and Reopen Cases programs will not allow a Claim Type code of ‘0’ for monetary denials.

38. If a separation denial is selected on which the disqualification is cleared by subsequent work, as reflected in the filing of a new claim, how would the number of weeks denied (before and after) be coded?

A. The number of weeks denied would not be affected by the existence of a new claim. It would be affected solely by the claim date of the denial selected and State law. In this case, the new claim is simply the vehicle for clearing the denial and its existence is not a factor in the coding of the denial determination.

39. In a case involving payment of partial benefits, if the agency pays an incorrect amount of benefits, which is later amended as a result of the DCA investigation, how should the number of weeks denied after investigation be coded? How should the dollar amount of the error be coded?

A. Number of weeks denied after investigation would be coded ‘0’ if the denial review determined that a partial payment was due, but the amount was erroneous. The dollar amount of the error is the difference between the amount the claimant should have received and the actual amount paid (as would be done for BAM paid claims). If the error is an overpayment, the amount of error cannot exceed the original amount paid.

40. Is it possible in States that reduce MBA due to disqualification that unadjusted MBA (before nonmon. reductions) minus benefits already paid will never agree with the “actual” remaining balance reported?

A. Yes. States should code MBA based on the monetary determination, not the adjusted MBA based on monetary redetermination made as a result of nonmonetary issues (for example, separation or administrative penalty).

41. Because the DCA methodology requires the investigation of only the issue on which the denial determination was based, some of the data elements will be not applicable. (For example, a deductible income issue does not require a work search investigation.) Will the DCA software allow cases to be closed without these data elements coded?

A. Update Cases and Reopen Cases will accept codes of ‘-2’ (not applicable) for data elements that may not be required for all cases. Several of the data elements can be coded ‘-2’ in the DCA software’s Stamp program.

42. Is it necessary to code the industry (NAICS) code of the claimant’s last and primary base period employers?

A. Yes, if it is relevant to the issue on which the denial determination was based. However, not applicable (‘-2’) can be coded if industry code is not needed to determine the accuracy of the denial determination.

43. For the Industry Code of Last Employer, do we code the last employer or the last bona fide (adjudicable) employer?

A. Code the last employer, as in BAM paid claims.

44. Is it necessary to enter the occupational code for usual job and employment being sought?

A. Yes, if it is relevant to the issue on which the denial determination was based (for example, if it is needed to determine whether the claimant made reasonable efforts to find employment). However, not applicable (‘-2’) can be coded if the occupation is not needed to determine the accuracy of the denial determination.

45. Is it necessary to code recall status (before) and (after) for denial cases?

A. Yes, if it is relevant to the issue on which the denial determination was based. However, not applicable (‘-2’) can be coded if recall status is not needed to determine the accuracy of the denial determination.

46. Will work search contacts be investigated in all cases?

A. Work search contacts will be investigated only when a denial based on a work search issue is selected. Work search related data elements (investigated, acceptable, not acceptable, not verifiable) can be coded ‘-2’ (not applicable) if they are not needed to determine the accuracy of the denial determination.

47. What methodology was most successful for the pilot States in obtaining a completed claimant questionnaire: in-person, mail, or phone?

A. The five pilot States used different data collection methods. One State relied mostly on telephone for all three types of denials. Another State used all three methods, but slightly favored mail for monetary denials and telephone for separation and nonseparation denials. The third State used mostly in-person interviews for separation and nonseparation denials and in-person and telephone equally for monetary denials. The fourth State favored mail for all three types of denials, and the fifth pilot State used in-person investigation for the majority of cases for all three types of denials.

All pilot States tended to prefer doing the claimant interview in-person when possible, on the premise that an in-person interview yields the highest quality information. However, cost considerations, as well as the difficulty of gaining the cooperation of denied claimants for in-person interviews, often led States to an alternate data collection method.

48. A claimant was denied benefits on a separation decision, which the claimant appealed. At the time the case was coded for DCA the appeals referee had not heard the appeal. Should DCA wait until the appeal is heard since the determination was appealed in a timely manner - it is not final?

A. The DCA investigation is conducted independently from the appeals process. The results of the DCA investigation should be coded (denial was either proper or improper). The fact that the case is under appeal should be coded in data elements Initial Determination Appealed (code=1, claimant appealed) and Results of Appeal (code=5, appeals decision pending).

49. DCA mailed a call-in notice to a claimant who was monetarily ineligible; the claimant failed to report for the interview to discuss the monetary denial. The following week, which was the change of quarter, the claimant filed a new claim that was eligible, and he drew the first week of benefits on the new eligible claim.

(1) Does the claimant’s failure to report affect the DCA investigation in terms of whether the monetary denial that was sampled was proper?

A. No, the issue for the DCA investigation is whether the claimant met the State’s monetary eligibility requirements for the claim that was sampled. Although the claimant subsequently established eligibility on a new claim, under the DCA methodology only the issues that affect the original denial decision are investigated.

(2) Can the claimant be disqualified in the new benefit year because he failed to report for an interview on the denied claim?

A. Yes, if State law provides for such disqualification. However, this should be addressed through an adjudication independent of the DCA investigation.

50. A claimant filed a claim in a prior benefit year. He had been discharged for a dishonest act; the wage credits were removed; and he was determined monetarily ineligible. Claimant returned to work for a short period of time, lost that job, and filed a new claim, which was also monetarily ineligible because he had no other employment in the base period. This latter denial is sampled by DCA. Does DCA investigate only the base period wages, or does DCA also investigate the separation issue in the prior benefit year, which resulted in the removal of wages?

A. The question to investigate is whether the claimant had sufficient wages for the sampled denial. The prior discharge separation issue should not be investigated for the monetary denial because under the DCA methodology, only the accuracy of the monetary determination, not the determination of the prior separation issue, is investigated. Unless the DCA investigation turns up additional wage credits that had not been taken into account in making the monetary determination, then the denial is proper.

The separation determination on the discharge should have been included in the DCA separation sampling frame for the week in which the determination was issued, to be eligible for selection.

51. Should the three data elements Union Service, Union Assistance Requested, and Claimant Union-Assisted be coded for all three types of denials? Can values for these elements be stamped?

A. In the current release of the DCA software, the three union assistance data elements are only accessible through the nonseparation screen. However, these elements can apply to any type of denial. In a subsequent software release, these three elements will be moved from the nonsep. screen to the record information screen, where they can be coded for all three types of denials.

The current version of the DCA software does not include the three union assistance data elements in the Stamp program. This change will be requested for a future software release. However, this software revision will not be available until after the August implementation of DCA. States should continue to code these data elements for nonseparation denials from the claimant interview responses, or these fields can be coded in the rec1.dat file, which is downloaded from the State’s mainframe or LAN to the Sun server.

52. If DCA determines that the claimant was erroneously denied benefits, is the State required to initiate official action to redetermine claimant eligibility?

A. This depends on State law. This issue is discussed in the Benefit Accuracy Measurement State Operations Handbook, ET Handbook No. 395, 3rd edition (chapter II, p. 1).

53. How do claim type and claim date relate to the sample type and the transaction date? What information are we trying to capture in the elements claim type and claim date?

A. For BAM paid claims, the records are payments issued within the batch (sampling week); for denied claims, the records are determinations that deny eligibility for unemployment compensation issued within the batch. The transaction date data element in the UI transactions file, which is the input file for the COBOL program, is one of several edit criteria used in determining which records are included in that batch’s sampling frame.

Claim type and date are separate data elements, which should reflect the claim for which an issue was identified, was adjudicated, and for which a denial was issued – for example, new initial claim (lack of wages); additional claim (VQ), continued week claimed (A&A).

The following table summarizes the claim types that are valid for the purposes of defining the BAM sampling frames (� = valid claim type X = invalid claim type).

Claim Code / Type Paid Monetary Separation Nonseparation Claim Denial Denial Denial 01 - New Initial X 02 - Additional X X 03 - Transitional X X 04 - Reopen X X X 13 - Continued* X 00 - No Week Claimed X X

• Codes ‘12’ (first payment) and ‘14’ (final payment) are also valid.

54. Are dependents issues addressed only for monetary denials?

A. The dependents allowance data elements are coded only for monetary denial sample types. Per Handbook 401 (p. I-4-5), one of the examples of actions that DO NOT constitute a nonmonetary determination and should NOT be reported is, “A determination on the existence of and/or the number of dependents. This is part of the monetary determination function and under no circumstances should be reported as a nonmonetary determination.”

These data elements should be used to record before and after DCA investigation information. For claimants erroneously denied, code the number of dependents and the amount of dependents allowance that should have applied if the monetary had been correctly determined.

55. What data elements are required to be coded and which data elements can be coded “not applicable”?

A. States must code all data elements that are relevant to determining the accuracy of the determination to deny benefits that they are investigating. Certain case identification and claimant information data elements are required for all cases: batch, sequence number, sample type, SSN, claim type, claim date, state, local office, investigator ID, program type, and combined wage claim (cwc). Claimant demographic information (date of birth, gender, ethnic, education, citizenship) must be coded with the appropriate values or missing (‘-1’). Benefit year data elements (original amount paid, number of weeks denied (B/A), wba (B/A), mba (B/A), and remaining balance (B/A)) can all be coded ‘0’ (zero), if appropriate. Reason for monetary, separation, and nonseparation denial (B/A) must reflect the appropriate issue code. Case outcome information (action code, appeal, investigator and supervisor completion fields) must also be coded. Other data elements may be coded ‘-2’ (not applicable), if the element is not relevant to the issue on which the denial was based. For example, the data elements related to work search can be coded ‘-2’ (not applicable) if claimant work search is not relevant to the issue on which the denial was issued.

56. Why is the ethnic data element in the UI transactions file record still one position, but the paid claims ethnic code (b13) and the DCA ethnic data element are both two positions and capture both race and ethnicity?

A. These data elements perform different functions. The sole purpose of the ethnic data element in the UI Transactions file is to produce aggregate sample and population counts that are reported in the sfsum.dat file and stored in the b_comparison and b_dca_comparision tables in the UI database on the State Sun server. Sample and population data are aggregated for three categories: white (code ‘1’), nonwhite (codes ‘2’, ‘3’, ‘4’, and ‘5’) and missing (code ‘8’).

When coding records for the UI transaction file, States should consider only the claimant’s race, not ethnicity (Hispanic / non-Hispanic). The valid codes in the UI transactions file record are:

1 = White 2 = Black or African American 3 = Asian 4 = American Indian or Alaska Native 5 = Native Hawaiian or Other Pacific Islander 8 = Information Not Available or Missing

Ethnicity (Hispanic / non-Hispanic) is not coded in the UI transactions file. It must be coded, either manually by the case investigator through the Update Cases software on the Sun server or automatically loaded from the rec1.dat file (position 169), in the first position of data element b13 in the b_master table or the ethnic data element in b_dca_master.

57. If a monetary denial is due to the failure to requalify for a new benefit year, would DCA need to investigate the base period wages used to establish the original benefit year?

A. Only the wages needed to requalify for the new benefit year need to be investigated.