Causal inference & experimentsCausal mechanismsSingle caseQualitative methodsHypothesis testingHistorical evidence
Process Tracing
Process Tracing
Also known as: Process tracking、Process tracing、Causal mechanism analysis
Tests evidence for causal pathways within one case by hypothesis. Considers not just event order but how well evidence distinguishes competing explanations. [1]
- Difficulty
- Intermediate
- Cost
- Medium
- Duration
- Illustrative planning durations, not common standards: 2–6 months for one case including collection/analysis
Content updated: 2026-09-13Source checked: 2026-09-13 (includes AI cross-checking)No documented human reviewIncludes fictional examples
How it works at a glance
Specify main/competing hypotheses and expected evidence beforehand. Check origins, timing and independence of evidence. [1]
Hoop tests check necessary conditions; passing does not prove hypotheses. Smoking-gun evidence may strongly support a hypothesis, but absence need not refute it. [1]
- Formulate causal and competing hypotheses
- Translate hypotheses into expected signals
- Systematically collect evidence
- Conduct evidence tests
- Reconstruct mechanisms
- Examine alternatives and report conclusions
When to use it and when it is unsuitable
When to use it
- Explain mechanisms of why programs work or fail
- Trace causal pathways deeply in 1 to several cases
- Explain mechanisms/context after quantitative effects are found
- Compare evidence consistency of competing causal hypotheses A and B
When it is unsuitable
- Large-sample statistical effects/generalization suit quantitative or large comparative studies
- Confidential or inaccessible past records prevent intermediate-process evidence
- Only quantitative effect magnitude, rather than mechanisms, is required
Participants, roles, and deliverables
Participants and roles
1–2 evaluators experienced in qualitative/case research, with document/interview field staff and access to key program/policy informants
Main deliverables
- Diagram: program → mechanism → outcome
- Test matrix: hypotheses × evidence × tests × judgments
- Grounds for retaining/rejecting competitors
- Theoretical and practical implications
Implementation steps
- Formulate causal and competing hypothesesSpecify intervention-outcome pathways and important alternatives. Focus on explanatory differences, not counts.
- Translate hypotheses into expected signalsPre-list documentary, interview or observation fingerprints expected under each hypothesis. Organize implications of presence/absence through 4 test types.
- Systematically collect evidenceCollect program documents, minutes, staff interviews, administrative records and media. Attend to unexpected evidence suggesting new hypotheses too.
- Conduct evidence testsAssess support/refutation through hoop, smoking gun, doubly-decisive and straw-in-the-wind tests. Judge diagnostic power to distinguish hypotheses.
- Reconstruct mechanismsIntegrate tests and reconstruct what happened and how outcomes arose over time. Exclude refuted hypotheses and present the most evidence-consistent explanation.
- Examine alternatives and report conclusionsHonestly report remaining explanations/uncertainty. State limits of single-case generalization and theoretical/practical implications.
Strengths and limitations
Strengths
- Reveals causal black boxes and answers why programs worked
- Rigorous inference with few cases for context-dependent paths beyond quantitative methods
- Systematic transparent qualitative evidence assessment
- Substantially deepens understanding when complementing quantitative evaluation
Limitations
- Limited generalization from one/few cases elsewhere
- Subjective interpretations may yield different evaluator conclusions
- Collection/analysis take time and effort, unsuited to whole large-program assessment
- Sparse records or limited informant access hinder discrimination
Common pitfalls
- Chronological storytelling is not process tracing. Preformulated hypotheses and discriminating evidence tests are central.
- Predetermined conclusions and no competing hypotheses create confirmation bias. Always compare alternatives against evidence.
- Avoid seeking only decisive evidence or treating many weak items as confirmation. Passing hoops does not confirm; assess discrimination, credibility and dependence.
- Do not generalize a case-specific mechanism everywhere. State scope and conditions.
Conditions of use and evidence
- Finding and not finding evidence may have different hypothesis implications. [1]
Trace whether training materials informed meeting decisions
A fictional illustration examining adoption of new indicators.
Scroll horizontally to view the table
| Evidence | Test treatment | Fictional judgment |
|---|---|---|
| Staff reply confirming receipt 2 days before meeting | Temporal-order hoop | Consistent with necessary order; decision use unconfirmed |
| Minutes explaining adoption reasons not yet obtained | Mechanism evidence unconfirmed | Defer judgment on decision use |
| Funder-mandated indicators unconfirmed | Examine competing H2 | Do not reject H2 |
Common errors in judgment
Common mistake
Assume prior timing proves causality
How to revise
Treat order as necessary and examine mechanisms/competitors
ReasonOther factors can have the same timing.
Common mistake
Reject hypotheses because evidence is absent
How to revise
Check observability and recording completeness
ReasonSome events leave no records.
Organize your thinking by writing
Process tracing evidence-test worksheet
A practical template is available to fill in and print on A4 paper.
Open worksheetRelated practical guides
Related methods
References and checked sections
- Process tracingBetterEvaluation
- Section consulted
- Process tracing involves four types of causal tests
- How it relates to the text
- Finding and not finding evidence may have different hypothesis implications.
Editorial and checking information
How to cite this page
EvalCompass Editorial Team. “Process Tracing”. EvalCompass, updated 2026-09-13. https://evalcompass.jun-nakatani.com/en/methods/process-tracing/
Add the date you accessed the page. Text created by this site is available under CC BY 4.0. Quotations, external materials, and bundled fonts have their own terms of use. When reusing content, credit the source, link to the license, and indicate any changes.
Change history
- — Checked definition/condition sources and added fictional examples, judgment rationales and completed fields. Revised unsupported history/uptake claims, existing procedures, fixed counts/thresholds and institutional generalizations. Unified fictional evidence as receipt acknowledgments, distinguishing sending, receipt, reading and decision use. Rechecked examples against completion checks and conditional procedures.