Evidence-Based Grants: Tiers, Registries & How to Win
Updated 21 official sources
Quick Answer: Some funding opportunities require an intervention to meet a defined evidence standard, while others ask for an evidence-informed rationale or a local evaluation plan.
Full answer and how we checked it
Start with the opportunity’s exact standard, then check whether the cited research or registry matches the proposed population, setting, outcomes, and delivery model. A registry listing alone does not prove that a model satisfies every competition’s requirement.
Who this is for: Program directors, nonprofit grant writers, school district leaders, justice agencies, and evaluators who need to prove that a proposed program meets a funder's evidence standard.
Data note: This guide was checked against current What Works Clearinghouse, CrimeSolutions, SAMHSA Evidence-Based Practices Resource Center, OJJDP, eCFR (34 CFR 77.1), and ED materials on October 12, 2026. The live funding opportunity controls its evidence standard and accepted sources.
In this guide
- June 2026 Source Map For Evidence-Based Grants
- 6 Evidence Claim Checks Before Writing
- What "Evidence-Based" Means in Federal Grants: 4 Tiers
- The ESSA Evidence Tiers: 4 Levels Most Applicants See
- Where to Find 6 Approved Evidence-Based Program Sources
- Which 4 Federal Agency Groups Require Evidence
- How to Build Evidence for Your Own Program in 5 Steps
- Pay for Success and Social Impact Bonds
- Common Mistakes with Evidence Requirements
- What OMB's Evidence-Based Policy Guidelines Doesn't Tell You About Evidence-Based Grant Programs
- Related Guides
- Sources
If you've applied for a federal education, criminal justice, or behavioral health grant in the past five years, you've probably encountered a requirement that your proposed program be "evidence-based." Maybe the NOFO asked you to implement a program rated as effective by the What Works Clearinghouse. Maybe it required you to cite research meeting "moderate evidence" standards. Maybe it used the phrase "demonstrates a rationale" and you had no idea what that meant.
If you're a program director building an evidence base, a researcher designing evaluations, or a grant writer making evidence claims in proposals, this guide covers what 'evidence-based' means across federal agencies.
You're not alone. Evidence requirements have become one of the fastest-growing and most confusing features of federal grantmaking. A 2016 Government Accountability Office report (opens in a new tab) found that tiered evidence models were spreading across agencies, but implementation varied widely and lessons from early programs hadn't been systematically shared.
This guide explains what "evidence-based" means in the federal grant context, walks through the tiered evidence frameworks that agencies use, shows you where to find approved programs, and covers how to build evidence for your own interventions over time.
For open opportunities that explicitly ask for evidence tiers, use the Grantsights grant database alongside the agency NOFO. The grant page tells you where the opportunity sits, but the NOFO tells you whether a Tier 1 study, a Tier 3 study, or a Tier 4 logic model is enough.
June 2026 Source Map For Evidence-Based Grants
Use the source named by the NOFO first. If the NOFO does not name a registry, use the agency's evidence framework and downgrade any claim that does not match the study population, setting, dosage, or outcome.
According to the What Works Clearinghouse ESSA page, WWC tools help administrators find studies that meet strong and moderate ESSA evidence definitions. According to the CrimeSolutions FAQ (opens in a new tab), programs rated from March 2025 on get a rating for each outcome rather than one overall program rating. According to SAMHSA, the Evidence-Based Practices Resource Center gives communities, clinicians, program planners, and policymakers practice resources rather than the retired NREPP rating model. According to OJJDP, the Model Programs Guide used CrimeSolutions' review process and ratings for juvenile justice and youth prevention programs, but its pages are now marked as an archive that's no longer updated. According to ED, the EIR program uses early-phase, mid-phase, and expansion lanes, so the evidence claim must match the award lane before the budget is drafted.
| Applicant question | Primary source to check | Why it matters |
|---|---|---|
| Does an education intervention meet ESSA evidence levels? | What Works Clearinghouse ESSA page and WWC handbook | WWC ties education studies to ESSA evidence tiers |
| Does ED reward stronger evidence with larger awards? | Education Innovation and Research program page and NOFO | EIR separates early-phase, mid-phase, and expansion logic |
| Does a justice program have reviewed evidence? | CrimeSolutions (OJJDP's Model Programs Guide is archived) | DOJ and OJJDP use reviewed program and practice evidence |
| Does a behavioral-health model have implementation resources? | SAMHSA Evidence-Based Practices Resource Center | SAMHSA provides practice resources rather than the old NREPP rating model |
| Is the evidence acceptable for this competition? | The live NOFO on Grants.gov or the agency portal | The NOFO controls over registry shorthand |
Get grant deadline alerts by email
About once a week, we email a short list of open grant calls with real deadlines from different funders. It isn't limited to the programs in this guide.
No spam. Unsubscribe anytime.
6 Evidence Claim Checks Before Writing
| Check | Pass signal | Reject-risk signal |
|---|---|---|
| Tier named | The narrative says Tier 1, 2, 3, or 4 and cites the study | The narrative says "evidence-based" without a tier |
| Population match | The study population overlaps with the proposed service population | Study used a different age group, risk level, or setting |
| Outcome match | The study measured the same outcome the NOFO rewards | Evidence is for attendance but the grant scores reading growth |
| Registry fit | The cited registry is named or accepted by the NOFO | Applicant cites a registry the funder did not request |
| Fidelity plan | Staff training, dosage, monitoring, and adaptation rules are named | The proposal adapts the model without explaining how fidelity is protected |
| Evaluation plan | Tier 4 or local models include a clear evaluation path | Applicant claims strong evidence from pre-post data |
What "Evidence-Based" Means in Federal Grants: 4 Tiers
In everyday language, "evidence-based" just means "supported by research." In federal grants, it means something much more specific: your proposed program must be supported by a defined level of research rigor that the funding agency has spelled out in its NOFO.
The federal government's push toward evidence-based grantmaking started in the early 2000s and accelerated dramatically after 2009, when the Obama administration launched several tiered-evidence grant programs. The idea, outlined by Results for America (opens in a new tab), works like venture capital: agencies invest larger amounts in programs with stronger evidence and smaller amounts in programs that are earlier in their evidence-building journey.
According to the What Works Clearinghouse ESSA page (opens in a new tab), ESSA evidence levels require more than a positive study name. Tier 1 and Tier 2 evidence both require statistically significant positive effects, at least 350 students, and at least 2 educational sites. According to CrimeSolutions (opens in a new tab), DOJ ratings are based on reviewed evaluations and meta-analyses, with programs rated by evidence strength and direction of effect. According to ED, the Education Innovation and Research program uses 3 grant types that move from early-phase to mid-phase to expansion evidence, with expansion awards reaching up to $15 million.
A common mistake is overstating the tier. For example, an applicant for a youth mentoring grant that has only a local pre-post survey can't call its model "strong evidence." A stronger framing would call it Tier 4, cite the research behind the parent model, add a fidelity plan, and budget for an outside evaluator. That framing doesn't hide the weak evidence. It shows the reviewer the applicant understands the standard.
Where official sources stop: WWC, CrimeSolutions, and ED explain evidence levels separately, but they do not give applicants a single reject-risk rule. Our working rule is simple: if the study population, delivery setting, dosage, and outcome do not match the NOFO, downgrade the claim one tier before writing the narrative. That tells the applicant whether to apply, switch models, or budget for evaluation first.
Under the 34 CFR 77.1 definitions (opens in a new tab) used in ED discretionary competitions, strong evidence needs a study sample that overlaps with both the populations and settings you'll serve, while moderate evidence needs overlap with the populations or settings. Bad fit cases are common: skip a registry-backed program if the study population, implementation setting, or dosage does not match your NOFO or your community.
This approach has three goals:
- Direct more money toward programs that work. If randomized controlled trials show that a reading intervention improves outcomes, fund that intervention rather than unproven alternatives.
- Encourage innovation with guardrails. Smaller grants for emerging approaches let agencies test new ideas while requiring evaluation to determine whether they actually work.
- Build a body of knowledge over time. Every grant that includes an evaluation component adds to the research base, helping future funders make better decisions.
The result is that many federal grants now include specific evidence requirements. How those requirements work depends on the framework the agency uses.
The ESSA Evidence Tiers: 4 Levels Most Applicants See
The Every Student Succeeds Act of 2015 established the evidence framework that most people encounter first, and it's become the template for other agencies. The statutory definition sits in ESEA section 8101(21) (opens in a new tab), and ED's discretionary competitions use the more detailed definitions in 34 CFR 77.1 (opens in a new tab). Together they set four tiers of evidence:
Tier 1: Strong Evidence
Requires at least one well-designed and well-implemented experimental study, meaning a randomized controlled trial (RCT). The study must show a statistically significant positive effect on a relevant outcome. For ED discretionary grants, 34 CFR 77.1 adds that a single study must meet What Works Clearinghouse standards "without reservations," include at least 350 students or other individuals across more than one site, and show no overriding statistically significant negative effects.
What this means for applicants: You're implementing a program that's been tested in an RCT and proven effective. You can point to a specific study in a peer-reviewed journal or a WWC intervention report. Programs at this tier receive the most competitive advantage and often the largest grant awards.
Tier 2: Moderate Evidence
Requires at least one well-designed and well-implemented quasi-experimental study. A quasi-experimental study compares outcomes between a treatment group and a comparison group, but without random assignment. Common designs include regression discontinuity, difference-in-differences, and propensity score matching.
What this means for applicants: Your program has been studied rigorously but not necessarily through a full RCT. Under 34 CFR 77.1, an experimental or quasi-experimental study that meets WWC standards "with or without reservations" can count, again with at least 350 participants across more than one site. The comparison group has to be similar enough to the treatment group to draw valid conclusions.
Tier 3: Promising Evidence
Requires at least one well-designed and well-implemented correlational study with statistical controls for selection bias. This is a lower bar than quasi-experimental because the study design doesn't need to approximate random assignment as closely, but it must still control for confounding variables.
What this means for applicants: You have research showing a positive correlation between your program and desired outcomes, with enough statistical rigor to rule out obvious alternative explanations. Tier 3 is the minimum for school improvement activities funded under ESEA section 1003, which can't rely on Tier 4.
Tier 4: Demonstrates a Rationale
The statute requires a rationale based on high-quality research findings or positive evaluation, plus ongoing efforts to examine the effects. Under 34 CFR 77.1, a key component in the project's logic model must be supported by citations of high-quality research or evaluation findings. This is the entry-level evidence tier. You don't need studies of your specific program. You need a logical argument, backed by research on similar approaches, for why your program should work.
What this means for applicants: You're proposing something new or adapted, and you can articulate why it should produce results based on what we know from related research. You also need a plan to examine the effects, which usually means an evaluation plan in your application. This tier is designed for new approaches that haven't been formally studied yet.
How the Tiers Map to Funding
The tiered model creates a direct relationship between evidence strength and funding levels. The Department of Education's Education Innovation and Research (EIR) (opens in a new tab) program illustrates this clearly:
| Tier | EIR Grant Type | Most recent maximum award | Evidence Required |
|---|---|---|---|
| Tier 4 | Early-phase | Up to $6 million (FY2024; no FY2025 early-phase competition) | Demonstrates a rationale |
| Tier 2 | Mid-phase | Up to $10 million (FY2025) | Moderate evidence |
| Tier 1 | Expansion | Up to $15 million (FY2025) | Strong evidence |
According to the FY2025 EIR mid-phase notice (opens in a new tab), ED estimated $253 million for new mid-phase and expansion awards combined, required a 10% match, and didn't run an early-phase competition that year. Grantsights compared the FY2024 and FY2025 EIR notices: ED estimated $251 million for all three grant types in FY2024 (opens in a new tab), when early-phase awards could reach $6 million, and $253 million for only mid-phase and expansion in FY2025. That funding trend means a Tier 4 applicant can't count on an early-phase round every year. ED hasn't posted an FY2026 EIR notice, so check the EIR program page (opens in a new tab) before planning around these caps. The structure still rewards stronger evidence with bigger checks. Early-phase grants fund development and feasibility testing. Mid-phase grants expand and rigorously evaluate. Expansion grants scale proven programs nationally.
Where to Find 6 Approved Evidence-Based Program Sources
If your NOFO requires an evidence-based program, you need to know where to look. Several registries catalog and rate programs based on their evidence:
What Works Clearinghouse (WWC)
Run by the Institute of Education Sciences (opens in a new tab) within the Department of Education, the WWC reviews education research and rates interventions based on the quality and results of existing studies. It's the gold standard for education grants.
The WWC maintains intervention reports that summarize the evidence for specific programs and practice guides that synthesize research on broader topics. Each intervention receives a rating based on the quality of studies reviewed and the direction of effects found.
Use for: Department of Education grants, ESSA-funded programs, any education-focused federal grant.
CrimeSolutions.gov
Operated by the National Institute of Justice (opens in a new tab) within the Department of Justice, CrimeSolutions rates criminal justice, juvenile justice, and victim services programs. Programs receive ratings of "Effective," "Promising," or "Ineffective" based on the quality of evaluation evidence (CrimeSolutions changed the old "No Effects" label to "Ineffective" in January 2025), and interventions without enough evidence go on its Inconclusive lists.
According to How CrimeSolutions Works (opens in a new tab), each screened program and practice is reviewed by 2 certified reviewers using objective scoring instruments. That matters because a program listed there is not just a program description. It has gone through an evidence review.
Use for: DOJ grants (BJA, OJJDP, OVW), state criminal justice planning grants, juvenile justice programs.
Blueprints for Healthy Youth Development
Maintained by the University of Colorado Boulder's Institute of Behavioral Science (opens in a new tab), Blueprints rates youth development programs as "Promising," "Model," or "Model Plus." Model programs need stronger evaluation evidence plus positive effects that last at least 12 months after the program ends; Model Plus programs have also had a high-quality independent replication.
Use for: Youth-focused grants from OJJDP, SAMHSA, and some Department of Education programs. Several NOFOs specifically reference Blueprints.
SAMHSA Evidence-Based Practices Resource Center
After SAMHSA retired the National Registry of Evidence-Based Programs and Practices (NREPP) in 2018, it replaced it with the Evidence-Based Practices Resource Center (opens in a new tab). This resource doesn't rate programs the same way NREPP did. Instead, it provides resources, toolkits, and implementation guides for evidence-based approaches in substance abuse and mental health.
Use for: SAMHSA grants, behavioral health programs, substance use prevention and treatment.
OJJDP Model Programs Guide
The Office of Juvenile Justice and Delinquency Prevention's Model Programs Guide (opens in a new tab) rated juvenile justice and youth prevention programs using CrimeSolutions' review process, scoring instrument, and ratings. Its pages are now marked as an archive that's no longer being updated, so confirm a program's current rating on CrimeSolutions.
Use for: OJJDP grants, state juvenile justice formula grants, when the NOFO still names it.
HHS Home Visiting and Prevention Clearinghouses
ACF's Home Visiting Evidence of Effectiveness (HomVEE) (opens in a new tab) review identifies home visiting models that meet HHS criteria for evidence of effectiveness, and the Title IV-E Prevention Services Clearinghouse (opens in a new tab) rates mental health, substance use, parenting, and kinship programs as well-supported, supported, or promising for Family First prevention spending.
Use for: MIECHV home visiting grants and Title IV-E prevention services plans.
Important Note About Registry Overlap
Different registries can rate the same program differently because they use different review criteria. A program rated "Effective" on CrimeSolutions might not appear in the What Works Clearinghouse at all if it hasn't been reviewed there. Always check which registry your specific funder recognizes. Some NOFOs name acceptable registries explicitly. Others defer to the ESSA tiers without specifying a particular registry.
Which 4 Federal Agency Groups Require Evidence
Not every federal grant requires evidence-based programming. Here's where you'll encounter evidence requirements most often:
Department of Education
Education is the agency most committed to evidence requirements. Since ESSA's passage in 2015, evidence tiers have been embedded in:
- Title I school improvement (ESEA section 1003): The old School Improvement Grants program ended with ESSA. Now, school improvement funds for schools the state identifies for improvement support must go to evidence-based interventions at Tier 1, 2, or 3.
- Title I, Part A: Improvement plans for the lowest-performing schools must include evidence-based interventions. See our Title I grants guide for how those funds flow.
- Education Innovation and Research: The flagship tiered-evidence program, with three grant tiers matching evidence levels.
- Full-Service Community Schools: Has asked for evidence-based practices, but ED discontinued a set of FSCS grants (opens in a new tab) in December 2025, so check the current notice before planning around it.
- Teacher quality and STEM grants: Many require applicants to demonstrate evidence for proposed professional development or instructional approaches.
If you're applying for any Department of Education grant, expect to encounter evidence requirements.
Department of Justice
DOJ has been moving toward evidence requirements across several programs:
- Byrne JAG (Justice Assistance Grants): A formula grant, so there's no competitive scoring, but BJA encourages states and localities to fund evidence-based programs
- Second Chance Act grants: NOFOs emphasize evidence-based reentry practices (see our Second Chance Act guide)
- COPS grants: Increasingly weight evidence in competitive scoring
- OJJDP grants: Many programs reference the Model Programs Guide and CrimeSolutions
DOJ programs typically don't use the ESSA tier structure (our DOJ grants guide covers the main offices). Instead, they reference their own registry (CrimeSolutions) or describe evidence requirements in the NOFO's selection criteria.
SAMHSA
SAMHSA's approach to evidence has evolved significantly. After retiring NREPP in 2018, the agency shifted to:
- Requiring evidence-based practices in program design: Many SAMHSA grant programs require applicants to describe the evidence supporting their proposed approach
- Referencing multiple registries: NOFOs may point to CrimeSolutions, Blueprints, or general literature reviews rather than a single SAMHSA registry
- Emphasizing implementation fidelity: SAMHSA wants to see that you'll implement the evidence-based model as designed, with training and technical assistance from the program developer
HHS (Broader)
Beyond SAMHSA, other HHS agencies use evidence frameworks:
- Office of Population Affairs (OASH): The Teen Pregnancy Prevention program funds replication of effective programs as its Tier 1 track. The FY2026 Tier 1 replication NOFO (AH-TP1-26-001) (opens in a new tab) closed July 23, 2026.
- HRSA and ACF: The Maternal, Infant, and Early Childhood Home Visiting (MIECHV) program, run by HRSA in partnership with ACF, requires most funds to go to home visiting models that meet HHS evidence-of-effectiveness criteria, as identified by the HomVEE review.
- CDC: Some grant programs reference evidence-based practices, though CDC typically focuses on evidence-based public health strategies rather than requiring specific program registries.
Agencies That Generate (Rather Than Require) Evidence
NIH, NSF, and similar research agencies fund the studies that build evidence. They don't typically require applicants to implement existing evidence-based programs. If you're writing an NIH R01 or NSF research proposal, you're proposing to create evidence, not deploy it. The evidence requirements described in this guide apply primarily to service delivery and implementation grants, not research funding.
How to Build Evidence for Your Own Program in 5 Steps
If your program isn't in any registry yet, you're not locked out of federal funding. You can start at Tier 4 and build upward over multiple grant cycles. Here's the progression:
Step 1: Develop a Strong Logic Model (Tier 4)
Your logic model is your evidence starting point. It needs to:
- Identify the specific problem and population you're addressing
- Describe your program activities and how they connect to short-term outputs and long-term outcomes
- Cite existing research that supports each link in your causal chain
- Explain why your approach should work based on theory and related evidence
A Tier 4 logic model isn't just a diagram. It's a research-backed argument. If you're running a mentoring program for justice-involved youth, you should cite research on mentoring effectiveness, the specific risk factors your program addresses, and the theoretical framework (cognitive behavioral, social learning, etc.) that explains your mechanism of change.
Step 2: Conduct Formative and Process Evaluation
During your first implementation, collect data systematically:
- Track who participates, how much service they receive (dosage), and whether you're implementing as planned (fidelity)
- Measure outcomes using validated instruments
- Document implementation challenges and adaptations
This data won't meet Tier 3 standards on its own, but it builds your case for future proposals and helps you refine your program before investing in a rigorous study.
Step 3: Correlational Study with Controls (Tier 3)
Commission a study that compares your participants' outcomes to a comparison group, controlling for observable differences statistically. This might use:
- Propensity score matching to identify similar non-participants
- Pre-post designs with comparison groups
- Regression analysis controlling for demographics and baseline characteristics
You'll likely need a research partner for this, either a university researcher or an evaluation firm. Get a quote early, because cost depends heavily on sample size and data collection requirements.
Step 4: Quasi-Experimental Study (Tier 2)
Advance to a more rigorous design:
- Regression discontinuity: If your program uses a cutoff score for admission, this design compares outcomes just above and below the cutoff
- Difference-in-differences: Compares changes over time between treatment and comparison groups
- Matched comparison with strong controls: More sophisticated matching methods than Tier 3
These studies cost more than correlational work and require careful planning before implementation begins, not after.
Step 5: Randomized Controlled Trial (Tier 1)
An RCT randomly assigns eligible participants to treatment and control conditions. This is the strongest evidence design but also the most expensive, ethically complex, and logistically challenging. You'll need:
- Enough participants to detect meaningful effects (statistical power analysis)
- An ethical framework for randomly denying services (often handled through waitlists or staggered rollout)
- An independent evaluator with RCT experience
- IRB approval
- A budget sized to the trial's scale and data collection, which is usually the largest evaluation cost in the progression
The full progression from Tier 4 to Tier 1 usually takes several years and multiple grant cycles. That's by design. The tiered model is built for iterative learning, not overnight results.
Pay for Success and Social Impact Bonds
Pay for Success (PFS), sometimes called Social Impact Bonds, represents a different approach to evidence-based funding. Instead of government paying upfront for services and hoping for results, private investors fund evidence-based programs, and government repays investors only if predetermined outcomes are achieved. An independent evaluation verifies whether targets were met.
According to an archived Obama White House Social Innovation and Civic Participation page (opens in a new tab), from its first major award in 2012 the administration dedicated nearly $100 million to PFS through eight federal agencies, drawing at least $65 million in philanthropic, state, and local funds.
Notable PFS projects include:
- Denver Supportive Housing Social Impact Bond: Concluded at the end of 2020, with outcome payments tied to housing stability and reduced jail days. The city reported repaying investors about $9.6 million, close to the payment projected in the Harvard Kennedy School Government Performance Lab summary (opens in a new tab)
- DOL Pay for Success pilots: Case in point: in 2013, DOL awarded nearly $24 million (opens in a new tab) in Workforce Innovation Fund grants, $12,000,000 to the New York State Department of Labor and $11,670,000 to Massachusetts, for employment and reentry services for formerly incarcerated people. DOL's final process report (opens in a new tab) found neither project achieved the impacts that would have triggered DOL payment
- State and local PFS projects: Covering recidivism reduction, early childhood education, chronic homelessness, and workforce development
PFS matters for evidence-based grantmaking because it creates a financial incentive to use proven programs. Investors won't fund interventions without strong evidence because their returns depend on outcomes. This drives demand for Tier 1 and Tier 2 programs.
However, PFS remains a niche mechanism. Legal, evaluation, and intermediary costs are high, deal structures are complex, and the number of completed projects remains small. It's worth understanding as part of the evidence-based funding ecosystem, but it won't be relevant for most grant applicants.
Common Mistakes with Evidence Requirements
Watch for these errors:
Claiming Evidence You Don't Have
Applicants sometimes cite a single pre-post study of their own program and call it "strong evidence." It's not. The evidence tiers have specific definitions. A pre-post study without a comparison group doesn't meet even Tier 3 standards. Be honest about where your evidence actually sits and frame your proposal accordingly.
Using the Wrong Registry
If the NOFO says "programs rated effective by What Works Clearinghouse" and your program is rated effective on CrimeSolutions but hasn't been reviewed by WWC, you don't meet the requirement. Read the NOFO's evidence section carefully and use the specific registries it names.
Citing Outdated Evidence
Research doesn't age well in fast-changing fields. If your strongest study is from 2005 and the population, context, or intervention has changed significantly, reviewers will notice. Supplement older foundational studies with more recent research that supports your approach.
Ignoring Implementation Fidelity
Having evidence for the general model isn't enough if you're planning to implement it differently. If the evidence supports a 12-week program and you're proposing a 6-week version, you can't claim the same evidence base. Document any adaptations and explain why they shouldn't undermine effectiveness. Better yet, cite research on adapted versions.
Treating the Evaluation Plan as an Afterthought
Many applicants write detailed program narratives and then tack on a one-page evaluation section. In evidence-based programs, the evaluation is as important as the intervention. Build your evaluation budget, timeline, and methodology into the proposal from the start. Include a qualified external evaluator on your team, and check whether the NOFO sets an evaluation budget expectation or requires participation in a national evaluation.
Confusing Research Citations with Program Evidence
Citing general research ("studies show that mentoring improves outcomes for youth") isn't the same as citing evidence for your specific program model. Reviewers want to see evidence for the particular intervention you're proposing, not just the general approach. If your program is new, acknowledge that and explain how the general research supports your specific design.
What OMB's Evidence-Based Policy Guidelines Doesn't Tell You About Evidence-Based Grant Programs
The four tiers (strong, moderate, promising, and demonstrates a rationale) come from ESSA and ED's 34 CFR 77.1 regulations, not from the Foundations for Evidence-Based Policymaking Act of 2018, which requires agencies to build evidence plans and evaluation capacity but doesn't define grant evidence tiers. Many applicants claim 'evidence-based' status for their program without specifying which tier. Reviewers increasingly require you to identify the specific tier and cite the studies that support your claim.
Agencies interpret evidence requirements differently. ED's tiered evidence framework for its Education Innovation and Research (EIR) grants is the most prescriptive: mid-phase and expansion applicants need studies reviewed under What Works Clearinghouse standards, either already in the WWC or assessed by ED or its reviewers. SAMHSA maintains its own Evidence-Based Practices Resource Center. Using evidence from the wrong clearinghouse for a given agency can weaken your application.
Building your own evidence base through a logic model with measurable outcomes is the starting point. But a 'strong evidence' claim usually rests on a multi-site RCT that took years to run and publish. If your program is newer, target the 'promising' or 'demonstrates a rationale' tiers and build your evaluation plan into the proposal budget.
Related Guides
Sources
- What Works Clearinghouse ESSA Tiers (opens in a new tab)
- GAO Report on Tiered Evidence Grants (GAO-16-818) (opens in a new tab)
- Education Innovation and Research Program (opens in a new tab)
- CrimeSolutions.gov Evidence-Based Programs (opens in a new tab)
- CrimeSolutions rating process FAQ (opens in a new tab)
- SAMHSA Evidence-Based Practices Resource Center (opens in a new tab)
- OJJDP evidence-based programs (opens in a new tab)
- OJJDP Model Programs Guide (opens in a new tab)
- Results for America: Evidence-Based Grantmaking (opens in a new tab)
Next steps
Where to go next
These next pages help you move from research into a specific grant choice.
Build the logic model
Use when the evidence claim is Tier 4 or needs a theory of action before the NOFO narrative.
Compare ED grants
Check how ESSA tiers and EIR phase logic apply to education opportunities.
Compare SAMHSA grants
Use when the evidence claim is behavioral health, substance use, or recovery support.
Find grants that fit you
Free eligibility check: answer 9 questions and see ranked matches with deadlines and fit reasons.
Frequently Asked Questions
What are the four ESSA evidence tiers?
ESEA section 8101(21) defines four levels. Tier 1 (Strong) needs at least one well-designed and well-implemented experimental study showing a statistically significant effect. Tier 2 (Moderate) needs at least one well-designed quasi-experimental study. Tier 3 (Promising) needs at least one well-designed correlational study with statistical controls for selection bias. Tier 4 (Demonstrates a Rationale) needs a rationale based on high-quality research plus ongoing efforts to examine the effects. Title I school improvement funds under ESEA section 1003 accept only Tiers 1 to 3, and ED's larger discretionary awards, such as EIR expansion grants, typically require Tier 1 or Tier 2.
Where can I find approved evidence-based programs for my grant application?
The main federal registries are the What Works Clearinghouse (ies.ed.gov/ncee/wwc) for education interventions, CrimeSolutions.gov for criminal justice programs, and the Blueprints for Healthy Youth Development registry (blueprintsprograms.org) at the University of Colorado Boulder. SAMHSA's NREPP was retired in 2018 and replaced by the Evidence-Based Practices Resource Center. OJJDP's Model Programs Guide, which used CrimeSolutions ratings, is now an archive page that's no longer updated. For home visiting and child welfare models, HHS uses the HomVEE review and the Title IV-E Prevention Services Clearinghouse. Each registry uses its own rating system, so check which registry your specific funder recognizes. Some NOFOs specify which registries they accept.
Do all federal grants require evidence-based programs?
No. Evidence requirements vary significantly by agency and program. The Department of Education is the most aggressive about requiring evidence, with programs like Education Innovation and Research (EIR) and several ESSA-funded grant programs built entirely around evidence tiers. The Department of Justice requires evidence for some programs but not all. SAMHSA has shifted toward requiring evidence-based practices in substance abuse and mental health grants. Agencies like NSF and NIH fund research that generates evidence rather than requiring applicants to use existing evidence-based programs. When a program does require evidence, the NOFO will specify the tier and acceptable sources.
How do I build evidence for my own program?
Start at Tier 4 by developing a strong logic model that connects your activities to outcomes through a clear theory of change, supported by existing research on similar approaches. Conduct a formative evaluation during your first implementation to document outputs and early outcomes. Move to Tier 3 by commissioning a correlational study with statistical controls, comparing your participants to a matched comparison group. Advance to Tier 2 through a quasi-experimental design using methods like regression discontinuity or propensity score matching. Reaching Tier 1 requires a randomized controlled trial, which typically needs external research partners and dedicated evaluation funding. This progression usually takes several years and multiple grant cycles.
What is Pay for Success and how does it relate to evidence-based grants?
Pay for Success (also called Social Impact Bonds) is an outcomes-based financing model where private investors fund evidence-based social programs upfront, and government pays back investors only if the program achieves predetermined outcomes verified by independent evaluation. An Obama White House archive page says that from 2012 the administration dedicated nearly $100 million to PFS through eight federal agencies, drawing at least $65 million in philanthropic, state, and local funds. PFS projects need strong evidence because investors need confidence that the program will deliver measurable results. The model has been used for homelessness reduction, recidivism prevention, early childhood education, and workforce development. It remains a niche mechanism because deals are complex and costly to structure, and not every project pays out: neither of DOL's two 2013 PFS pilots reached the impacts that would have triggered DOL payment.
Last updated October 12, 2026. Corrected October 12, 2026: what changed. Published by Grantsights from public sources. See our editorial policy, how we build and check guides, corrections and how we make money.
Related Guides
Complete Guide
Federal Grant Proposal Guide: Sections & Scoring
Federal grant proposals average 50-100 pages. This guide covers every section and scoring criteria for writing a proposal that reviewers score highly.
CDC Ethiopia Faith Based HIV TB Grant Eligibility Guide
CDC withdrew its 2026 Ethiopia faith-based HIV and TB forecast (JG-26-0136). What it offered, who fit, and 5 tests to use if CDC reissues it.
Best Federal Grants for Faith-Based Nonprofits 2026
9 federal grants for faith-based nonprofits in 2026, ranked by realistic access: EFSP, CDBG, NSGP, AmeriCorps and more, with each program's faith rule.
Grants for Faith-Based Organizations (2026)
Faith-based groups can compete for federal grants on equal terms. See which agencies fund FBOs, the 2024 notice rule, and what you can't spend money on.