EAT Step 3 Walk-Through — Identify Burdens, Benefits, and Disparities

Content

IMPORTANT — Scope of this resource: This walks staff through the methodology of the DHS Equity Analysis Toolkit. It does not perform equity analyses. Actual equity analyses are conducted by the appropriate authorities — your Equity Champion, the One DSD Equity Team, or the DHS Equity Director for your work area. What this step is for Step 3 (Engaging Communities) and Step 4 (Benefits and Burdens) of the DHS EAT are deeply intertwined — community engagement is how you learn about burdens and benefits, and identifying burdens and benefits is why community engagement matters. This walk-through focuses on the analytical work of Step 4: surfacing disparate impacts, using the Equity Lens Needs Assessment as a guide, and applying an intersectionality lens to understand how multiple identities interact to shape who bears burdens and who receives benefits. Plain-language overview Step 4 of the EAT asks: Who benefits, and who bears burdens — and are those distributions equitable? The DHS EAT Guide (https://mn.gov/dhs/) frames this step as the moment where data and community engagement come together to produce an honest assessment of impact. "Use data and community engagement to inform your process." The analysis at this step must be specific, not general. "The proposal may have unintended consequences" is not a finding. "The income documentation requirement disproportionately denies access to Somali-origin applicants, who are 3.2 times more likely to be denied for this reason than white applicants" is a finding. This step also requires honesty about unintended consequences — impacts that the proposal did not aim to produce but may produce regardless. Walk-through (the methodology) Core questions for Steps 3 and 4 For community engagement (Step 3), the EAT Guide asks: • Who are the most affected community members? How have you involved them in development of the proposal? • What has your engagement process told you about burdens or benefits? • What factors produce and perpetuate racial inequity in this context? What barriers prevent full participation of some groups? For benefits and burdens analysis (Step 4), the EAT Guide asks: • How will the proposal increase or decrease racial equity? • Who would benefit and who would be burdened? • What are possible unintended consequences? • How can you modify the proposal to correct negative impacts and enhance positive impacts? • Are the impacts aligned with community outcomes defined in Step 1? • How will you address impacts? Are there complementary strategies? How to surface disparate impacts Disparate impact means that a policy, practice, or program produces significantly different outcomes for different groups — even when the policy appears neutral on its face. The Race Forward Racial Equity Impact Assessment (https://www.raceforward.org/sites/default/files/RacialJusticeImpactAssessment_v5.pdf) describes this as "considering adverse impacts": identifying which racial and ethnic groups could be negatively affected and how those impacts could be prevented or minimized. Methods for surfacing disparate impacts: 1. Cross-tabulate data by race. Take the outcome measures identified in Step 2 and examine how outcomes differ across racial and ethnic groups. Look for gaps that are statistically meaningful and operationally significant. 2. Apply the "same rule, different reality" test. A rule that treats everyone identically can still produce disparate outcomes if different groups start from different positions. A documentation requirement that assumes everyone has access to formal employment records treats everyone "the same" — but produces different results for communities with high rates of informal employment. 3. Look for access barriers, not just outcome gaps. Disparate impact can occur before the service is delivered — in who applies, who completes the application, who receives eligibility determination, and who accesses the service at full intensity. Map the entire pathway, not just the endpoint. 4. Use community testimony to identify informal impacts. Data systems often miss impacts that communities experience but that are not formally measured. Staff overtime required to navigate a new process. Informal workarounds that emerge when a policy is unworkable. Trust erosion that reduces future engagement. Community engagement surfaces these. Using the Equity Lens Needs Assessment The Equity Lens Needs Assessment, completed before Step 1, established the depth of analysis required for this decision. At Step 4, return to it: are the impacts you are finding consistent with what the Needs Assessment suggested? If the analysis is revealing more significant disparities than anticipated, the scope may need to expand. If the analysis is revealing fewer disparities than anticipated, document that finding explicitly — do not simply move forward without noting the discrepancy. The Needs Assessment also established which communities are in scope. At Step 4, verify that those communities have been adequately represented in the community engagement (Step 3) and in the data (Step 2). If a community identified in the Needs Assessment has not been reached, that is a gap that must be named. Applying an intersectionality lens Intersectionality refers to the way multiple identities — race, disability, gender, immigration status, age, sexual orientation — overlap and interact to shape individual experiences. A Black woman with a physical disability experiences health and human services in ways that are not fully captured by data about Black people, or data about women, or data about people with disabilities as separate categories. Her experience is shaped by all three identities simultaneously. The DHS EAT Guide's commitment to examining impacts on BIPOC, Tribal Nations, Veterans, LGBTQIA2S+, and persons with disabilities implicitly calls for an intersectional analysis. In practice, this means: • Ask whether the communities most burdened are disproportionately those who hold multiple marginalized identities. • Resist the impulse to rank which identity is "most" affected. Intersectionality is not a hierarchy; it is a recognition of complexity. • When data does not allow intersectional analysis (because it does not capture multiple identity dimensions simultaneously), name that as a data gap and use community engagement to supplement. • Apply the GARE Racial Equity Toolkit (https://belonging.berkeley.edu/sites/default/files/ARCHIVE/gare-racial_equity_toolkit.pdf) principle: "Ensure racial equity impacts are at the core of decision-making" — while holding that racial equity does not exist in isolation from other forms of inequity. Modification opportunities Step 4 is not just diagnosis — it is the moment where the analysis identifies how the proposal can be changed to reduce burdens and enhance benefits. The EAT Guide asks: "How can you modify the proposal to correct negative impacts and enhance positive impacts?" And separately: "How will you address impacts? Are there complementary strategies?" This distinction matters: • Modification changes the proposal itself — adjusting the policy or practice to reduce its disparate impact. • Complementary strategies add new actions alongside the proposal to address burdens that the proposal itself cannot eliminate. Both are needed. A policy that cannot be fully modified (due to legal or fiscal constraints) may still be accompanied by complementary strategies — expanded outreach to underserved communities, translation and interpretation support, waiver processes for documentation requirements, dedicated staff support for communities with highest barriers. Worked example fragment (anonymized, methodology illustration only) Continuing the anonymized waiver program scenario: The Equity Champion reviews community engagement findings from Step 3 alongside Step 2 data. Findings: Burdens identified: • Income documentation requirement disproportionately affects Somali-origin, Karen-origin, and recent immigrant households (confirmed by both data and community testimony). • Application abandonment is highest among clients with primary languages other than English and Spanish — indicating a language access barrier not captured in formal denial data. • Community trust barriers: Karen community advocacy partners report that some eligible individuals are avoiding the application entirely due to concerns about data sharing with immigration enforcement (an unintended consequence not anticipated in the proposal's design). Benefits identified: • Proposed change to accept alternative income documentation would directly benefit the populations most burdened. • Simplification of the documentation checklist (a complementary strategy) would reduce application abandonment across all communities. Modification proposed: Accept sworn statements of household income as an alternative to formal documentation, with community-based organization support available for completion. Add Somali and Karen language application support as a complementary strategy. Unintended consequence flagged: Immigration-related trust concerns require a communication strategy clarifying data use limitations — a complementary strategy, not a modification to the policy itself. This resource walks through methodology only. The actual analysis would be conducted by the Equity Champion working with the DHS Equity Director. IS / IS NOT Step 3/4 IS — Step 3/4 IS NOT An honest, specific assessment of who benefits and who bears burdens — A general statement that "all communities are affected" An intersectional analysis that recognizes overlapping identities — A single-axis analysis (race only, or disability only, but not both) A moment to identify modification opportunities and complementary strategies — A moment to confirm that the original proposal was right Built on both data and community engagement — Built on data alone or community testimony alone A place to name unintended consequences, including trust erosion — A place to minimize inconvenient findings Common questions What if the analysis shows that the proposal disproportionately benefits one community over another? Does that mean the proposal is wrong? Not necessarily. The finding means the team must now decide what to do about the disparity. Modification, complementary strategies, or — in some cases — reconsidering the proposal entirely are all on the table. The analysis does not make the decision; it informs it. What if community members disagree with our data findings? Take the disagreement seriously. It is more likely that your data is incomplete or your interpretation is partial than that the community is simply wrong about their own experience. Use the disagreement as a signal to probe further — what does the community know that the data does not show? How do we handle the intersectionality of disability and race when our data systems do not capture both simultaneously? This is a common gap in human services data. Name it as a data gap. Use community engagement — particularly with disability advocacy organizations that specifically represent communities of color and immigrant communities — to fill the gap qualitatively. Recommend that future data collection capture both dimensions. Connection to DSD goals DSD's commitment to eliminating disparities based on race, ethnicity, age, abilities, veteran status, sexual orientation, and gender identity and expression — as articulated in the EAT Guide — is operationalized in Step 4. This is where the commitment meets specific decisions and produces specific findings. Without Step 4, DSD's equity commitments remain aspirational rather than analytical. When to escalate / where to refer • If Step 4 findings reveal significant, unexpected disparities: share with your Equity Director before finalizing the analysis. • If community engagement (Step 3) has not been completed when Step 4 analysis is needed: do not proceed with Step 4 as if Step 3 were done. Incomplete engagement produces incomplete findings. • If intersectionality analysis reveals impacts on Tribal Nations members: refer to the ADSA Tribal Collective for any engagement or interpretation of findings related to tribal communities. • If trust barriers related to immigration enforcement are identified: this is a sensitive finding requiring careful handling. Consult your Equity Director and DSD leadership before communicating findings publicly. Reflective questions 1. Think of a service your program area delivers. Who is most likely to benefit from it? Who is least likely to access it, and why? Have you ever formally analyzed that gap? 2. How would you describe the difference between a disparate impact and a disparate intent? Why does the distinction matter legally and ethically? 3. What modification would be most difficult for your program area to make — but would have the most significant equity impact? What would it take to make that change? 4. Whose voice is most often missing from benefit-and-burden analyses in your experience? What would it take to include them? 5. How does disability interact with race in the populations your program area serves? Where do you have data on that interaction, and where are you missing it? Sources • DHS Equity Analysis Toolkit Guide — Minnesota DHS (https://mn.gov/dhs/) (primary source, Steps 3–4) • Racial Equity Impact Assessment Guide — Race Forward (https://www.raceforward.org/sites/default/files/RacialJusticeImpactAssessment_v5.pdf) (sections 6–8: adverse impacts, equitable impacts, examining alternatives) • King County Equity Impact Review — Phase 2 and 3 (https://blog.homelessinfo.org/wp-content/uploads/King-County-Equity-Impact-Review-checklist_Mar2017.pdf) • Racial Equity Toolkit — GARE, via Othering and Belonging Institute (https://belonging.berkeley.edu/sites/default/files/ARCHIVE/gare-racial_equity_toolkit.pdf) • One DSD Program Charter v1.1 (one-dsd-program-charter) Always defer to the person in front of you.

Ask: Who are the most affected community members? How have you involved them in development of the proposal?

Ask: What has your engagement process told you about burdens or benefits?

Ask: What factors produce and perpetuate racial inequity in this context? What barriers prevent full participation of some groups?

Ask: How will the proposal increase or decrease racial equity?

Ask: Who would benefit and who would be burdened?

Ask: What are possible unintended consequences?

Ask: How can you modify the proposal to correct negative impacts and enhance positive impacts?

For all staff, use EAT Step 3 Walk-Through — Identify Burdens, Benefits, and Disparities to define the decision, compare evidence and impacts, document limits, and name the owner; use the guidance in the immediate task, check impact and access, and ask for help when authority is unclear. Start by ask: Who are the most affected community members? How have you involved them in development of the proposal?

Do not replace official legal, policy, clinical, supervisory, program, or Tribal authority.

Do not infer an individual's identity, preferences, needs, or experience from group-level information.

State uncertainty, use current authoritative sources, and escalate when the decision exceeds the user's role.