startthat.work home

CH-09 Structured self-critique

Asks about sources of uncertainty, evidence, gaps, critical assumptions and alternative hypotheses, and produces a written critique of one own analysis.

ChallengeEvidence: DocumentedChallenge technique, runs in step 6Strongest statement: describe

What it does

Asks about sources of uncertainty, evidence, gaps, critical assumptions and alternative hypotheses. Artefact: a written critique of one's own analysis.

When to use it

Use when you have drafted an analysis and want to find its gaps and weak assumptions before it is reviewed by others.

When it does not fit

It is applied dishonestly; the handbook says it is only as good as the honesty with which it is applied.

Result format

The result is a written critique of the analyst's own analysis that covers the sources of uncertainty, the evidence, the gaps, the critical assumptions and the alternative hypotheses. Each of these headings has its own section.

Requirements on the material

PropertyRequirement typeRequirement
Unit of analysisAlways appliesrequires=claim
Appraises source credibilityApplies in some casesno

Not restricted on: Structure, Content type, Variable types, Time structure, How the sample was chosen, Level of detail available, Number of sources, Whether sources can be joined, Amount of material, Known data problems, Producer interest, Source dependence, Origin, Document genre, Language, Personal information, Security classification, Licence.

Minimum materialA draft analysis.
Additional sources needednone
Not for (rules)none
Not for (text)It is applied dishonestly, since the handbook says it is only as good as the honesty with which it is applied.

Choosing and ordering

Kind of techniqueChallenge technique, runs in step 6
Phase tagvalidate
Other phase tagsconclude
Kind of findinggap; risk
Question answeredwhat; why
Strongest statement it supportsdescribe
Needs as input
Makes as output
Must exist firstNone
Loop targetreturn to analyse

Library checks by stakes

Each stakes level includes the checks of the level below it. Checks that need a person apply at high stakes only, so the process leaves them out at lower stakes (decision 38). Repeatability: judgement dependent.

StakesCheckPerson involvedCounts as a different method
exploratoryassumption and limitation statementnoneno
decision-informingassumption and limitation statementnoneno
decision-informingindependent reviewneeded; left out below high stakesno
high-stakesassumption and limitation statementnoneno
high-stakesindependent reviewneededno
high-stakesexpert reviewneededno
high-stakessecond technique with different assumptionsnoneconditional

Failure modes and limits

Model suitabilitymodel alone
Typical model failureReports that the analysis is sound and lists minor gaps, because the model reviews its own output with the same biases.
Error sourcesinput data; coder or rater judgement; language model
Certainty limitsA self-critique is not independent of the analysis it reviews; the result carries risk of bias and the technique overlaps with other checks.

Evidence and tag basis

Documented. Overlaps with a key assumptions check plus alternative hypotheses (inference). (UK Red Teaming Handbook)

Tag basis: input profile inference; selection inference; sequencing inference; robustness inference. All tag values are inference unless a source is named.