Theory of Change vs Logic Model

A theory of change is a causal explanation — why an intervention should work, what conditions must hold first, and what the argument assumes. A logic model is an operational depiction of one project’s resources, activities, and results. A theory of change asks why. A logic model asks what.

Key takeaways

• A theory of change explains causation; a logic model depicts operations.
• A theory of change is built backward from the long-term goal.
• A logic model is usually a subset of a larger theory of change.
• The ceiling of accountability marks where a project stops claiming credit.
• Relabeling a logic model without causal reasoning fools no reviewer.

What is a theory of change?

A theory of change is a documented causal argument. The Center for Theory of Change defines it as “a comprehensive description and illustration of how and why a desired change is expected to happen in a particular context,” focused on “filling in what has been described as the ‘missing middle’ between what a program does (its activities or interventions) and how these lead to desired goals being achieved” (Center for Theory of Change).

The method has a traceable lineage. The Aspen Institute Roundtable on Community Change published New Approaches to Evaluating Comprehensive Community Initiatives in 1995, and the evaluation theorist Carol Weiss popularized the term as a way of forcing initiatives to state the intermediate steps they were assuming. ActKnowledge, a New York social enterprise, developed much of the subsequent practical method and the outcomes-mapping software built around it (Theory of Change origins).

A theory of change is not confined to what your grant will pay for. A pathway can include preconditions produced by partners, by other funders, or by conditions already present in the community. Naming those preconditions is the point: a reviewer wants to know whether the applicant understands the whole causal chain, not only the segment the applicant is billing for. Work on this sits alongside everything else in the evidence, evaluation, and data hub.

How does a theory of change differ from a logic model?

A theory of change differs from a logic model in altitude, direction, and burden of explanation. A theory of change carries the reasoning and the assumptions. A logic model carries the operations. Candid puts the relationship plainly: a theory of change works “at the 30,000-foot level,” a logic model “zooms in more closely — typically to the program level,” with “the theory serving as the umbrella under which various logic models sit” (Candid, Laurel Molloy, 2020).

The table below compares the two instruments across the five dimensions that decide which one a funder is asking for.

Dimension	Theory of change	Logic model
Question answered	Why should this work?	What will we do and produce?
Scope	Organizational or initiative-wide; may include unfunded pathways	One project, bounded by the award
Construction direction	Backward from the long-term goal through preconditions	Presented forward, inputs through impact
Required contents	Preconditions, causal rationales, assumptions, indicators, narrative	Inputs, activities, outputs, outcomes, impact
Typical form	Pathway diagram plus written narrative	Linear table or one-page graphic

The relationship is nested, not opposed. A logic model is usually a subset of a larger theory of change — the operational slice a specific grant funds. The Pennsylvania Coalition Against Rape compresses the distinction into one sentence: “a theory of change is what puts the logic in a logic model” (PCAR, 2018).

When do funders require a theory of change?

Funders rarely use the two terms consistently, so the requirement is decided by the notice of funding opportunity rather than by the literature. Federal education regulations do not define a theory of change at all; they define a “logic model (also referred to as theory of action)” and score its quality as a selection criterion (34 CFR 77.1; 34 CFR 75.210).

Three patterns are worth recognizing. Human-services, education, and criminal-justice competitions usually ask for a logic model as a diagram, plus narrative reasoning that functions as a theory of change whether or not it is called one. Large private foundations and international funders more often ask for a theory of change by name, and expect a pathway diagram with assumptions and a narrative. Health funders sit between the two: the CDC framework treats a program description as often including a logic model, with evaluation priorities resting on “a shared understanding of the theory of change identified in the logic model” (CDC, Developing an Effective Evaluation Plan).

The Kellogg guide itself concedes that “the term logic model is frequently used interchangeably with the term program theory in the evaluation field” (W.K. Kellogg Foundation, 2004). Follow the funder’s usage. When a solicitation asks for both, supply a one-page logic model as the graphic and a narrative theory of change that states the assumptions and cites evidence for each causal link.

How do you build a theory of change?

Building a theory of change runs backward, in six stages. The ActKnowledge method begins with the long-term goal and works down to the earliest changes required. There are six stages:

1. Define the long-term outcome precisely — the population, the amount of change, and the timeframe. Getting the long-term outcome right is treated as the most consequential step in the entire process, and a first working session often produces nothing else.
2. Map preconditions backward. Ask what conditions must exist for the long-term outcome to occur, then repeat for each answer. The result is a pathway of change: vertical chains of outcomes, earliest at the bottom.
3. Test each link for necessity and sufficiency. The method “hinges upon defining all of the necessary and sufficient conditions required to bring about a given long term outcome” (Center for Theory of Change). A precondition that is necessary but not sufficient signals a missing sibling precondition.
4. State the assumptions explicitly. Three types matter: assertions about the connections between outcomes, the claim that all important preconditions have been identified, and justifications for the links between activities and the outcomes they should produce. Contextual and environmental factors form a fourth type.
5. Place the interventions. Interventions attach to the connectors between preconditions, not to the outcome boxes. One intervention may serve several pathways, and some pathways may carry no intervention at all because someone else supplies them.
6. Attach indicators and write the narrative. Every outcome in the pathway gets at least one indicator, and a written narrative explains the logic that the diagram compresses.

The direction of travel is what makes the method useful. Working backward surfaces the steps an applicant was silently assuming, which is exactly the material that separates a causal argument from an activity list.

What is the ceiling of accountability?

The ceiling of accountability is the line above which a project stops claiming credit. Most theories of change reach toward a goal larger than any single organization can deliver — ending poverty, eliminating a disease. The ceiling, usually drawn as a dashed line, separates the outcomes the project will measure and be judged on from the broader change it merely contributes to.

The ActKnowledge method draws the line as a dashed “accountability ceiling” above the long-term outcome, with an ultimate goal sitting above it that “your group will not hold itself accountable for” but may want in the framework to communicate the larger vision (Theory of Change Basics).

Practitioners define the line functionally. In the theory-of-change adaptation of the Medical Research Council framework for complex interventions, the ceiling of accountability is the “level at which you stop using indicators to measure whether the outcomes have been achieved and therefore stop accepting responsibility for achieving those outcomes,” and it “is often drawn between the impact and the long-term outcome” (De Silva et al., Trials 15:267, 2014).

Drawing the ceiling is a credibility move, not a retreat. A proposal that claims responsibility for community-wide condition change on a three-year award invites a reviewer to test the claim and find it unsupported. A proposal that says explicitly where its accountability stops, and links its measured outcomes to the larger goal as a contribution, reads as competent. The same discipline governs how far you can push claims about outputs, outcomes, and impact.

What goes wrong when a logic model is relabeled?

The dominant failure in relabeling is cosmetic conversion: taking an existing logic model, retitling the page “Theory of Change,” and submitting it without adding a single causal claim. A theory of change that contains no rationales, no assumptions, and no preconditions the project does not fund is a logic model in a different font, and experienced reviewers read the absence immediately.

Four specific gaps produce that outcome. The first is missing rationales — arrows drawn between boxes with nothing explaining why one produces the next. The second is a pathway with no assumptions, which implies either that nothing could interfere or that no one asked. The third is a theory built forward from current activities, which reliably justifies existing work rather than reaching the goal. The fourth is unattached evidence: research cited in a bibliography rather than at the causal link it supports, which matters because rationale-tier evidence requirements are defined at the level of the individual project component under the applicable levels of evidence.

There is also a failure in the other direction. A theory of change submitted where the funder asked for an operational depiction, with no dose, no outputs, and no measurable targets, reads as evasion. Match the instrument to the ask, and let the project design section carry whichever one the narrative needs.

Frequently asked questions

Can a proposal include both a theory of change and a logic model?

Yes, and strong proposals often do. The usual pattern is a narrative theory of change that states assumptions and cites evidence for each causal link, plus a one-page logic model graphic showing the funded project’s inputs, activities, outputs, and outcomes. Check page limits before adding both.

Is a theory of change ever required for a federal grant?

Federal notices more often require a logic model, sometimes called a theory of action. Some competitions ask for a “conceptual framework” or an explanation of “why the proposed components will produce the intended outcomes,” which is a theory of change under another name. Read the selection criteria rather than the program description.

Does a theory of change have to be a diagram?

No. A theory of change can be delivered as narrative alone, and the written narrative is a required element of the standard method even when a diagram exists. Diagrams help reviewers who skim; narrative carries the reasoning that diagrams cannot hold.

How long should a theory of change be?

Long enough to state the long-term outcome, the pathway of preconditions, the assumptions, and the evidence behind each link. In practice that runs one to three narrative pages plus a diagram. A theory of change that fits on a slide has usually dropped the assumptions, which is the part reviewers most want.

Who should be in the room when building a theory of change?

People who will implement the work, people who receive the service, and at least one person who will have to evaluate it. Backward mapping surfaces disagreements about what actually causes what, and those disagreements are cheaper to resolve during planning than during year two of an award.

Related topics

• Evidence, Evaluation, and Data — the hub for proving a project will work, and then proving it did
• How to Build a Logic Model
• Writing an Evaluation Plan
• Performance Measurement and Indicators
• Writing a Statement of Need

Sources

1. Center for Theory of Change. What is Theory of Change? https://www.theoryofchange.org/what-is-theory-of-change/ (accessed 2026-08-11)
2. Center for Theory of Change. How Does Theory of Change Work? https://www.theoryofchange.org/what-is-theory-of-change/how-does-theory-of-change-work/ (accessed 2026-08-11)
3. Center for Theory of Change. Theory of Change Origins. https://www.theoryofchange.org/what-is-theory-of-change/toc-background/toc-origins/ (accessed 2026-08-11)
4. Taplin, D. H., and Clark, H. Theory of Change Basics: A Primer on Theory of Change. ActKnowledge / Center for Theory of Change. https://www.theoryofchange.org/wp-content/uploads/toco_library/pdf/ToCBasics.pdf (accessed 2026-08-11)
5. W.K. Kellogg Foundation. Logic Model Development Guide. 2004. https://www.naccho.org/uploads/downloadable-resources/Programs/Public-Health-Infrastructure/KelloggLogicModelGuide_161122_162808.pdf (accessed 2026-08-11)
6. U.S. Department of Education. 34 CFR 77.1 — Definitions that apply to all Department programs. Legal Information Institute, Cornell Law School. https://www.law.cornell.edu/cfr/text/34/77.1 (accessed 2026-08-11)
7. U.S. Department of Education. 34 CFR 75.210 — General selection criteria. Legal Information Institute, Cornell Law School. https://www.law.cornell.edu/cfr/text/34/75.210 (accessed 2026-08-11)
8. Molloy, L. What is a theory of change? Or a logic model? Candid, February 14, 2020. https://candid.org/blogs/what-is-a-theory-of-change-or-a-logic-model/ (accessed 2026-08-11)
9. Pennsylvania Coalition Against Rape. Theory of Change and Logic Models. 2018. https://pcar.org/sites/default/files/resource-pdfs/tab_2018_logic_models_508.pdf (accessed 2026-08-11)
10. De Silva, M. J., Breuer, E., Lee, L., Asher, L., Chowdhary, N., Lund, C., and Patel, V. Theory of Change: a theory-driven approach to enhance the Medical Research Council’s framework for complex interventions. Trials 15:267, 2014. https://pmc.ncbi.nlm.nih.gov/articles/PMC4227087/ (accessed 2026-08-11)
11. Centers for Disease Control and Prevention. Developing an Effective Evaluation Plan. https://www.cdc.gov/tobacco/stateandcommunity/tobacco-control/pdfs/developing_eval_plan.pdf (accessed 2026-08-11)