UNEMPLOYMENT INSURANCE PROGRAM LETTER NO. 11-14...UI program administrators must conduct statistical or other quantifiable data analyses of demographic records and data, to determine

UNEMPLOYMENT INSURANCE PROGRAM LETTER NO. 11-14

Purpose

The purpose of this directive is to remind all State administrators of Unemployment Insurance (UI) programs of their responsibility:

(1) to collect and analyze claimant demographic data for possible indications of systemic discrimination, and

(2) to investigate any such indications of potential discrimination that the analyses disclose. This Unemployment Insurance Program Letter (UIPL) rescinds and replaces UIPL No. 46-89.

DOL’s regulations require recipients to provide information about the reasons for a data request, and the ways in which the data may be used, before asking about a claimant’s race/ethnicity, sex, age, or disability status.

All requests for information must inform the claimant that the provision of such information is voluntary, and that such information may not be used as determining factor of eligibility for participation in the unemployment insurance program.

This does not preclude state UI agencies from doing fact-finding if there is an indication the individual may not have been “able and available” for work and making an eligibility determination based on the fact finding.

Each recipient must record the

race/ethnicity,

sex,

age,

and where known, disability status,

of every applicant, registrant, participant, terminee, applicant for employment, and employee.

Beginning on January 3, 2019, each recipient must also record the limited English proficiency and preferred language of each applicant, registrant, participant, and terminee.

Such information must be stored in a manner that ensures confidentiality,

and must be used only for the purposes of recordkeeping and reporting;

determining eligibility, where appropriate,

for WIOA Title I-financially assisted programs or activities;

determining the extent to which the recipient is operating its WIOA Title I-financially assisted program or activity in a nondiscriminatory manner;

or other use authorized by law.

The guidelines state that “Self-reporting or self-identification using two separate questions is the preferred method for collecting data on race and ethnicity. In situations where self-reporting is not practicable or feasible, the combined format may be used.”

In DOL’s view, the UI program presents few, if any, circumstances under which self-reporting of race and ethnicity is not practicable or feasible; therefore, the two-question format should be used. The guidelines require that under this format, ethnicity must be collected first.

According to the guidance, the minimum designations for race and ethnicity in the two-question format are:

Ethnicity:

Hispanic or Latino

Not Hispanic or Latino

Race:

American Indian or Alaska Native

Asian

Black or African American

Native Hawaiian or Other Pacific Islander

White

Individuals shall be offered the option of selecting one or more racial designations.

Recommended forms for the instruction accompanying the multiple response question are "Mark one or more" and "Select one or more."

The guidance permits collection of data re: race and/or ethnicity in greater detail;

“however, any collection that uses more detail [must] be organized in such a way that the additional categories can be aggregated into these minimum categories for data

on race and ethnicity.”

DOL’s regulations require that all demographic data, including disability status, whether hard-copy or electronic, “must be stored in a manner that ensuresconfidentiality, and must be used only for the purposes of recordkeeping and reporting; determining eligibility, where appropriate . . .; or other use authorized by law.”

The system for maintenance of demographic data, if stored separately, must be cross-referenced to individual records, in the event that it is necessary for DOL or other authorized agencies to identify the race, gender, etc., of a particular claimant in the course of a complaint investigation or compliance review.

Medical and disability-related information must be collected and maintained on separate forms and in separate medical files.

To successfully meet these requirements, hard-copy files should be stored in locked cabinets, safe files, or secured rooms; electronic records should be password-protected.

Similar to demographic data, medical information must be cross referenced to individual records for the same reasons.

Access to demographic data, medical, or disability-related information, must be strictly limited and must be available only on a “need to know” basis.

The above confidentiality requirements related to demographic data are in addition to those imposed by the regulations related to confidentiality and disclosure of state UC information that are published at 20 CFR Part 603.

UI Monitoring Requierments

UI program administrators must conduct statistical or other quantifiable data analyses of demographic records and data, to determine whether their UI programs and activities are being conducted in a nondiscriminatory way.

These analyses must, at a minimum, include analyses by race/ethnicity, sex, age, and disability status, to identify any statistically significant differences in the success rates of claimants who are members of these demographic categories.

Overall Single-Claimant Claims Processed:

Total number of new initial claims Total number of additional initial claims Total number of initial claims (new and additional)

Single-Claimant Monetary Determinations:

By race/ethnic identification, sex, age, and disability status for the following:

• Total number of monetary determinations made

• Total number of monetary determinations resulting in ineligibility

Single-Claimant Non-Monetary Determinations:

• Total number of non-monetary determinations made

• Total number of monetary determinations denying benefits

Separation issues:

A) Total number of separation from work determinations made because of: • Voluntary quits • Discharge for misconduct • Other

B) Total number of separation from work determinations denying benefits because of:• Voluntary quits • Discharge for misconduct • Other

Non-separation issues:

A) Total number of non-separation determinations made by the following issues:

• Able, available, and actively seeking work • Disqualifying or deductible income • Refusal of suitable work • Reporting requirements • Other

Non-separation issues

B) Total number of non-separation determinations denying benefits by the following issues:

• Able, available, and actively seeking work • Disqualifying or deductible income • Refusal of suitable work • Reporting requirements • Other

Single-Claimant Appeals

Lower authorityTotal number of lower authority appeals decisions made by the following:• Separation issues

o Voluntary quits o Discharge for misconduct o Other

Single-Claimant AppealsLower authority

• Non-separation issues

o Able, available, and actively seeking work o Disqualifying or deductible income o Refusal of suitable work o Reporting requirements o Other

Non-separation issues:B) Total number of appeals decisions in favor of claimants

C) Total number of appeals decisions not in favor of claimants

Higher authority:

A) Total number of higher authority appeals decisions made

B) Total number of appeals decisions in favor of claimants

C) Total number of appeals decisions not in favor of claimants

1. Map service delivery process & Obtain program data

2. Obtain civilian labor force or population data for your service area

3. Review any anecdotal evidence you received during the period

4. Analyze the data using the 80% Rule or the Two Standard Deviation Test

5. Investigate significant differences

6. Justify or take mitigating actions

7. Follow-up

U.S. Census Bureau

American FactFinder

American Community Survey

Compare the eligible population in your service area to your applicants (compare percentages)

Unemployment Compensation

Total Monetary Determinations to Monetary Determinations denied

Total Non-monetary Determinations to Non-monetary Determinations denied

Total Separation Issues to Separation Issues denied

Total Non-separation Issues to Non-separation Issues denied

Total Lower Appeals to Lower Appeals against claimants

Total Higher Appeals to Higher Appeals against Claimants

• Find the success rates for each demographic of the group

• Find the best rate of the group

• Compare each of the other rates to the best rate

• A result below 80% indicates a significant difference that needs to be investigated

Compare success rates, failure rates, Before-Pools and After-Pools

Analyze using the Z Test for Statistical Significance

A result should be greater than 2 standard deviations indicates a significant difference that needs to be investigated

Four-fifths or 80% Rule Two Standard Deviation Test

Data analysis: statistical evidence

Participation Rate =For each demographic in the group

𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑝𝑝𝑁𝑁𝑁𝑁𝑝𝑝𝑜𝑜𝑝𝑝𝑝𝑝 𝑝𝑝𝑁𝑁𝑠𝑠𝑠𝑠𝑁𝑁𝑝𝑝𝑝𝑝𝑜𝑜𝑁𝑁𝑠𝑠𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑝𝑝𝑁𝑁𝑁𝑁𝑝𝑝𝑜𝑜𝑝𝑝𝑝𝑝 𝑎𝑎𝑎𝑎𝑎𝑎𝑁𝑁𝑁𝑁𝑝𝑝𝑎𝑎𝑎𝑎𝑝𝑝𝑎𝑎

Impact Ratio =𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃𝑟𝑟 𝑃𝑃𝑜𝑜 𝑃𝑃 𝑑𝑑𝑟𝑟𝑑𝑑𝑃𝑃𝑑𝑑𝑃𝑃𝑃𝑃𝑃𝑃𝑑𝑃𝑃𝑃𝑃

𝐵𝐵𝑟𝑟𝐵𝐵𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃𝑟𝑟 𝑃𝑃𝑜𝑜 𝑃𝑃𝑑𝑟𝑟 𝑑𝑑𝑃𝑃𝑃𝑃𝑔𝑔𝑃𝑃

Below 80% = significant difference

𝑆𝑆𝑆𝑆𝑔𝑔−𝑆𝑆𝑆𝑆𝑜𝑜

(𝑂𝑂𝑂𝑂𝑆𝑆 ×𝑂𝑂𝑂𝑂𝑆𝑆)×( 1𝑇𝑇𝑔𝑔+ 1𝑇𝑇𝑜𝑜

)

Where:

SRg = the selection rate for the group under consideration

SRo = the selection rate of othersOAR = the overall selection rateODR = the overall denial rateTg = the number of applicants for the group under

considerationTo = the number of other applicants

Above 2 standard deviations = significant difference

Two Standard Deviation TestFour-fifths or 80% Rule

A quick way to calculate the 80% Rule: 1) select the best rate in the group and 2) determine 80% of that rate. Other rates in the group that are below that denial rate show a significant difference (adverse impact)!

43.24% divided by 80% =54.1%

(.4324 divided by .80 = .541)

Any rate greater than 54.1% shows a significant difference (adverse impact)

The Allow rates could be used instead. In that case the best rate would be the highest rate – 56.75%. You would then multiply the best rate by 80%. The result is 45.4%. Therefore, any rate below 45.4% would show significant difference.

Need proof that you conducted an investigation into the adverse impact areas by the following actions:

Indicates a significant difference

A red flag; not necessarily proof of discrimination

Course of action:

1. Investigate possible reasons for the significant differences

2. If justification (business necessity) is found, document it; no further action necessary for that demographic in that step

3. If justification is not found, take mitigating actions

4. Follow-up to see if the mitigating actions made a difference

Collecting Limited English Proficiency Data Effective January 3, 2019

Beginning on January 3, 2019, each recipient must also record the limited English proficiency and preferred language of each applicant, registrant, participant, and terminee.

Such information must be stored in a manner that ensures confidentiality, and must be used only for the purposes of recordkeeping and reporting;

22

59

4 8 6

60

3 1 5

25

1 1 6 1 4 1 3

17

196

126 2 7

19

0

50

100

150

200

250

LEP Assessment Data PY 16-17

1. Include a summary of your organization setup

2. Map out your service delivery process

3. Describe your civilian labor force or population data for your service area

4. Describe the investigation of the significant differences with the adverse impact identified

5. Describe the justification or mitigating actions taken

6. Describe the follow-up method that will be used

7. Provide back up copies of your data

Danielle Smith, State WIOA EO Officer [email protected] Phone: 573-751-2428

Kristin Funk, EO Disability Coordinator [email protected] Phone: 573-751-1098

Isaac Hagan, EO Investigator [email protected] Phone: 573-751-8626