startthat.work home

HE-05 Margin of error and credibility interval check

Checks that a reported margin of error matches the sampling method and produces a judgement on whether the figure is valid.

Handle evidenceEvidence: DocumentedPreparation technique, runs in step 5.2Strongest statement: describe

What it does

Checks that a reported margin of error matches the sampling method. Artefact: a judgement on whether the figure is valid. Read the sampling type first, because a margin of sampling error applies to probability samples and opt-in polls report model-based credibility intervals that depend on assumptions.

When to use it

Use when a poll or survey reports a margin of error and you need to know whether that figure means what it says.

When it does not fit

Fits any reported survey precision. The rule is mechanical, but judging the assumptions behind a credibility interval needs a person.

Result format

The result is a judgement on whether the reported margin of error is valid for the sampling method. It names the sampling type first and says whether the figure is a margin of sampling error or a model-based credibility interval that depends on assumptions.

Requirements on the material

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

Not restricted on: Structure, 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 materialnone
Additional sources needednone
Not for (rules)none
Not for (text)Judging the assumptions behind a model-based credibility interval needs a person, even though the rule for probability samples is mechanical.

Choosing and ordering

Kind of techniquePreparation technique, runs in step 5.2
Phase tagverify and prepare
Other phase tagsnone
Kind of findingrisk
Question answeredhow much
Strongest statement it supportsdescribe
Needs as input
Makes as output
Must exist firstNone
Loop targetproceed with limit stated

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
exploratoryspot checkoptionalno
decision-informingspot checkoptionalno
decision-informinghuman review sampleneeded; left out below high stakesno
high-stakesspot checkoptionalno
high-stakeshuman review sampleneededno
high-stakesindependent reviewneededno
high-stakessecond technique with different assumptionsnoneconditional

Failure modes and limits

Model suitabilityhuman judgement required
Typical model failureAccepts a margin of sampling error for an opt-in poll, where only a model-based credibility interval can apply.
Error sourcesinput data; statistical assumptions
Certainty limitsCertainty drops when the sampling type is not disclosed or the model behind a credibility interval is not stated (reporting bias).

Evidence and tag basis

Documented. The source is a statement of position and not an experiment. (AAPOR statement on credibility intervals)

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