Comparison Readout Annotation¶
Interpretation protocol — instantiates Dimensioned Comparison Framing
Attaches to a finished comparison an explicit statement of what relation it actually supports — rank, dominance, trade-off, equivalence, or non-comparability — together with the uncertainty, scope limits, and uncompared dimensions, so the result cannot be over-read.
A comparison can be computed flawlessly and still be read wrongly. The Comparison Readout Annotation is the closing bookend that governs interpretation: it takes a finished comparison and states, in plain language, exactly what relation the numbers legitimately support — and refuses to let a reader promote it into a stronger claim than the data earns. Its one defining move is to treat the type of relation as a decision rather than an assumption. A composite total looks like a ranking, but the annotation asks whether it is really a rank at all, or a trade-off, a statistical tie, or an outright non-comparability — and then attaches the caveats that must travel with the verdict: the uncertainty around it, the scope it holds within, and the dimensions that were never compared. It adds no score and rebuilds no grid; it is the discipline that turns a computed number into an honest statement.
Example¶
A family foundation is deciding which of two after-school education nonprofits to fund. An analyst has already built the comparison — cost per student, retention, test-score gains, and reach — and Nonprofit A edges out B on the weighted composite. The program officer could stop there and write "A is the stronger program." Instead she runs the readout annotation over the finished comparison.
First she classifies the relation. A's lead comes entirely from a lower cost-per-student; on retention and score gains, B is actually ahead. That is not a dominance — it is a trade-off, and the honest readout is "A is cheaper per student, B retains and teaches better," not "A wins." Second she fixes the boundary: both nonprofits operate only in urban middle schools, so the verdict says nothing about the rural high-school context the board is also weighing. Third she registers the uncompared dimension — long-term life outcomes, the thing the foundation ultimately cares about, could not be measured for either organization — and marks it unknown, not equal, so no one reads the tie as evidence of parity. Finally she records the whole readout, caveats included, so next year's grant cycle can check whether the same reading still holds.
The annotation changed no cell in the grid. It changed the sentence the board acted on — from a false winner into a genuine trade-off with its limits attached.
How it works¶
- Classify the relation. Inspect the computed result and name which relation it actually supports — similarity, difference, rank, dominance, equivalence, deviation-from-baseline, trade-off, or non-comparability — using a stated rule rather than defaulting to "whatever is on top wins."
- Attach the boundary and uncertainty. Record the scope the verdict holds within (population, setting, time window) and the uncertainty around it, so those limits travel with the conclusion into wherever it is quoted next.
- Register the uncompared dimensions. Carry forward the dimensions that could not be compared and mark them unknown, guarding against the silent inference that an absent comparison means the items are equal on that axis.
- Record the readout. Write the labeled relation together with its caveats and the frame that produced it into a note a later reviewer can re-open and challenge.
Tuning parameters¶
- Downgrade strictness — how readily a headline rank is demoted to a trade-off or a tie. Strict rules refuse to declare a winner unless the margin clears the uncertainty; loose rules report a rank more often but risk crowning noise.
- Caveat verbosity — from a one-line boundary note to a full limits appendix. More caveats protect against over-reading but, past a point, drown the finding; match it to how far the conclusion will travel from the people who built it.
- Uncertainty representation — whether the verdict carries a point claim, a "too close to call" band, or an explicit confidence statement. Bands are honest but harder for decision-makers to act on than a clean call.
- Consequence framing — whether the annotation stops at "what the relation is" or extends to "what to do given it." Adding the action step makes the readout usable but risks smuggling a recommendation past the evidence.
When it helps, and when it misleads¶
Its strength is that it closes the gap between what a comparison computed and what people conclude from it — the place where a modest edge quietly becomes "the best," a trade-off collapses into a ranking, and a dimension nobody could measure is read as a draw. It is the only sibling whose whole subject is that last interpretive step, and it makes a conclusion contestable by writing down the exact relation being claimed.
Its honest failure mode is the opposite excess: over-hedging, where every result is buried under so many caveats that it dissolves into "it depends" and offers a decision-maker no purchase. A comparison that can never yield an actionable reading has failed as surely as one that over-claims. The classic misuse is selective annotation — reporting the favored option crisply while hedging the rival into oblivion, so the caveats become an instrument of the very bias they are meant to check. The guarding discipline is symmetry (the same downgrade and caveat rules applied to every comparand) and reserving the strongest label, non-comparability, for cases that genuinely resist a common frame rather than using it to duck a hard call — the real distinction being value incommensurability[1], where dimensions cannot be placed on one scale at all.
How it implements the components¶
The annotation realizes the report-out core of the archetype — the components that turn a computed result into an honest, bounded statement:
relation_readout_rule— its central act: it applies the rule that decides whether the result is a rank, dominance, trade-off, equivalence, or non-comparability, and states that label explicitly so no stronger claim is read into it.context_and_boundary_record— it attaches the scope limits and uncertainty the verdict holds within, so the boundaries travel with the conclusion rather than being stripped off when it is quoted.missing_dimension_register— it carries forward the dimensions that could not be compared and marks them unknown, so their absence is never mistaken for parity.comparison_audit_trail— it records the labeled readout together with its caveats and the frame that produced it, so a later reviewer can check whether the same conclusion still follows.
It does not build the scored grid, choose dimensions, or normalize cells — that is the Dimensioned Comparison Matrix's work — and it does not adjudicate which items legitimately belong in the set, which is the Comparator Set Audit's job.
Related¶
- Instantiates: Dimensioned Comparison Framing — the annotation is the archetype's interpretation-and-report-out layer, the last step before the result becomes a claim.
- Consumes: Dimensioned Comparison Matrix supplies the computed relation it labels; the Dimension Weight Sensitivity Panel supplies the robustness caveats it carries forward.
- Sibling mechanisms: Dimensioned Comparison Matrix · Comparison Basis Checklist · Pairwise Comparison Protocol · Comparator Set Audit · Dimension Weight Sensitivity Panel · Matched Case Comparison Sheet · Counterbalanced Comparison Display
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: The mechanism inspects a completed comparison and issues a bounded interpretation of the relation it actually supports, together with uncertainty, scope, and uncompared dimensions, so its operative form is readout assessment.
Nearest alternative: Representation, Specification & Plan — The annotation preserves the finding, but the defining act is evaluating the finished comparison under a stated interpretation rule rather than merely externalizing information.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Philosophy
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Value theory and logic supplied distinctions among ranking, trade-off, equivalence, and incommensurability.
Related originating lineages:
- Statistics & Experimental Design — Uncertainty reporting contributes interval, scope, tie, and non-comparability qualifications on quantitative results.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Independent reviewer agreement; medium confidence.
Notes¶
The annotation shares two components — context_and_boundary_record and missing_dimension_register — with the Comparison Basis Checklist, but they sit at opposite ends of the pipeline and do opposite work. The checklist declares scope and exclusions at intake, as a gate that can stop a comparison before it starts; the annotation re-attaches them to the finished result as interpretation. Keeping both is not redundant: intake scope is a promise, report-out scope is a receipt, and a comparison can honor the first while quietly drifting past it by the time a number is read.
References¶
[1] Raz, J. The Morality of Freedom. Clarendon Press (1986). Defines value incommensurability as the absence of a common ordering in which alternatives are better, worse, or equal. registry ↩