Worked-example setupScope and assumptions
- Alder Software and its compensation plan are fictional.
- The packet establishes only the existence of a quarter-end revenue bonus and a descriptive increase in contracts signed near quarter end.
- No causal comparison design, contract testing, communication evidence, control-operation evidence, or legal analysis is supplied.
- Period
- Fictional fiscal second quarter
- Units
- Binary claim-support indicators: 1 means the stated evidence condition is met; 0 means it is not
- Rounding
- Exact binary indicators
Packet and initial claim
Alder Software pays a bonus when reported quarterly revenue exceeds a target. The supplied schedule shows a larger share of signed contracts near the end of the second quarter. A reviewer writes: “The bonus caused managers to manipulate revenue, so the plan must be abolished.”
That sentence compresses four analytical jobs: description, mechanism, causation, and recommendation.
Audit each transition
- Description supported: the supplied schedule can support a bounded claim about contract-signing timing in the displayed records.
- Mechanism plausible: the bonus changes the expected payoff from reaching the revenue target. That is an incentive hypothesis.
- Causation unsupported: the packet lacks a comparison design and evidence about customer demand, seasonal selling patterns, contract terms, recognition, overrides, or rival explanations.
- Misconduct and intent unsupported: a signing pattern and incentive do not establish manipulation, improper recognition, or state of mind.
- Recommendation unsupported: the writer names neither a decision criterion nor feasible redesigns, side effects, authority, or expected consequences.
The verified claim audit therefore marks description and a model implication as supported, while the causal and recommendation transitions remain open.
Rewrite the conclusion
Alder's bonus creates an incentive to reach the quarterly revenue target, and the supplied records show more contracts signed near quarter end. Those facts warrant targeted testing; they do not establish improper recognition, causation, or intent. Obtain contract-level recognition evidence, approval and override logs, communications, customer and seasonal controls, comparison periods, and compensation-governance records before reaching those conclusions or recommending a remedy.
This bounded conclusion is not timid. It directs procedures at the mechanism while preserving the evidence standard for stronger claims.
Verified calculation · economics foundations analysis
The curriculum loader recomputed this example before it entered the site build. Expand any structured input to inspect the stated facts.
- claim audits
- 1 field
Inspect data
{
"alder_bonus_claim": {
"causal_design_supplied": 0,
"decision_authority_identified": 0,
"feasible_alternatives_compared": 0,
"model_assumptions_stated": 1,
"observation_supplied": 1,
"value_criterion_stated": 0
}
}Recomputed result
| Measure | Value |
|---|---|
| causal claim supported | 0 |
| descriptive claim supported | 1 |
| model implication supported | 1 |
| normative criterion visible | 0 |
| recommendation supported | 0 |
| unsupported causal jump | 1 |
| unsupported normative jump | 1 |