Claim-Evidence-Reasoning Card¶
Template — instantiates Evidentiary Trace Warranting
Captures a compact claim, evidence trace, and reasoning bridge for review.
A Claim-Evidence-Reasoning Card is a single, small fill-in artifact that holds exactly one argument: one claim, the one trace offered in support of it, and the one sentence of reasoning that bridges the two. Its defining trait is that it is a single-relation, prose-first template built for legibility — its job is to make one inference readable end to end so a reviewer can see whether the leap from observation to conclusion actually holds. It does not rank strength, tabulate provenance, or combine multiple pieces of evidence; the moment you need to hold and weigh many relations at once, you have outgrown the card. The card forces the reasoning step into the open, which is precisely the part people leave implicit when they say "the data shows."
Example¶
A seventh-grader in a life-science class is asked to argue whether the fertilizer helped her bean plants. Left to instinct she would write "the fertilizer worked." The card makes her fill three boxes instead. Claim: the plants given fertilizer grew taller than the untreated plants. Evidence: over three weeks, the six fertilized plants averaged 14 cm while the six control plants averaged 9 cm, measured with the same ruler on the same days. Reasoning: because the two groups had the same light, water, and soil and differed only in fertilizer, the height difference is best explained by the fertilizer rather than by growing conditions.
Writing the reasoning box is where the learning happens. The student discovers that her claim needs the control group to license the bridge — that the evidence alone ("they were taller") does not carry her to "the fertilizer did it" without the sentence that rules out the ordinary alternative. The card is deliberately confined to this one argument; a second question ("did they also flower earlier?") gets its own card. That confinement is the point: one card, one legible inference.
How it works¶
The card is three named slots, filled in order and kept short:
- Claim — the single assertion being made, stated so it could in principle be wrong. Vague or compound claims are split until each card carries one.
- Evidence — the specific trace offered: what was observed, with the minimum detail (counts, conditions, when and how) needed to recognize it, not a summary conclusion.
- Reasoning — one or two sentences naming the principle or rule that connects this evidence to this claim. This is the load-bearing slot and the one most often skipped.
The discipline is one relation per card and no smuggling: the claim may not appear inside the evidence box, and the reasoning may not simply restate the claim. Cards are cheap, so a complex position becomes a small stack of them rather than one overloaded card.
Tuning parameters¶
- Granularity — how much a "single claim" is allowed to bundle. Finer cards are clearer but multiply; coarser cards are fewer but hide sub-inferences.
- Reasoning depth — a one-line warrant versus a short paragraph naming the backing principle. Deeper reasoning surfaces hidden assumptions at the cost of brevity.
- Evidence specificity — how much raw detail the evidence slot must carry before it counts as a trace rather than a paraphrase.
- Prompts on the template — sentence stems ("This is best explained by… rather than…") that scaffold novices but can make expert cards formulaic.
- Review target — whether the card is written for a peer, a grader, or one's future self; the audience sets how much can be left tacit.
When it helps, and when it misleads¶
Its strength is that it makes a single inference inspectable by anyone, which is why the Claim-Evidence-Reasoning framework is a staple of science education — it externalizes the reasoning move that separates "I observed X" from "therefore Y."[n1] For a first-encounter reviewer, one card answers the only question that matters: does the reasoning actually get you from this evidence to this claim?
Its failure mode is scale and self-containment. Because a card holds one relation, it cannot show how several traces converge, conflict, or jointly support a claim — force a whole case onto one card and the reasoning box becomes a paragraph doing five jobs badly. The classic misuse is treating a filled card as validation: the slots being non-empty says the argument is stated, not that it is strong or survivable. The guarding discipline is to keep cards atomic and hand off any weighing, provenance tracking, or cross-trace synthesis to mechanisms built for it — the card's honesty comes from staying small.
How it implements the components¶
observable_trace_record— the Evidence slot captures the specific observation with enough detail to be recognized as a trace, not a conclusion.hypothesis_or_claim_anchor— the Claim slot pins the one assertion the trace is being asked to support.inferential_link_model— the Reasoning slot names the bridge principle that licenses moving from this trace to this claim.
It does not record where the trace came from, grade its strength, or combine it with others (provenance_chain, scoped_evidential_weight, evidence_synthesis_interface) — that many-relation, cross-tabulating work belongs to the Evidence Relation Matrix, its nearest twin. The card holds one legible argument; the matrix holds and combines many.
Related¶
- Instantiates: Evidentiary Trace Warranting — the card is the archetype's smallest usable unit: one trace-claim-reasoning relation made explicit.
- Sibling mechanisms: Admissibility or Relevance Gate · Defeater Register · Evidence Relation Matrix · Evidence Strength Ladder · Evidence Update Review · Relevance and Alternative Explanation Check · Trace-to-Claim Diagram
Editorial Notes¶
Form Classification¶
Form family: Representation, Specification & Plan
Rationale: The practitioner fills a persistent three-slot template that externalizes one claim, its specific evidence trace, and the reasoning bridge without smuggling conclusions between fields, so its concrete form is a structured representation.
Nearest alternative: Interface, Display & Cue — The card guides the author's attention while being filled, but its continuing operative value is the completed reviewable argument artifact rather than an interactive affordance.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Education & Pedagogy
Origin pattern: Single lineage
Present-day reach: Multi-domain
Rationale: Science education established the Claim-Evidence-Reasoning scaffold as a compact template that forces students to state the bridge from observation to conclusion.
Related originating lineages:
- Philosophy — Informal logic supplies the distinction between conclusion, premises, and warrant.
- Rhetoric — Argumentation practice contributes clear, reviewable presentation of a single inference.
Review resolution: The source explicitly names the K-12 Claim-Evidence-Reasoning scaffold, fixing education as the primary lineage. Informal logic and rhetoric are antecedent resources, but the compact one-inference card cohered as a single educational method.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] The Claim-Evidence-Reasoning framework, developed for K-12 science instruction by Katherine McNeill and Joseph Krajcik, structures a scientific explanation into a claim, supporting evidence, and reasoning that links them — with the reasoning step explicitly required so students cannot substitute an observation for an argument. ↩