startthat.work home

HE-06 Opt-in panel respondent quality check (bogus respondent screen)

Looks for implausible answers that show careless or fraudulent respondents inflating rare responses in an opt-in survey and produces a judgement of inflation risk.

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

What it does

Looks for implausible answers that show careless or fraudulent respondents inflating rare responses. Artefact: a judgement of inflation risk. Read results for under-30 and Hispanic subgroups in opt-in samples first, the subgroups Pew found affected.

When to use it

Use when an opt-in online survey reports a rare behaviour or a result for a young or minority subgroup that looks too high.

When it does not fit

Fits opt-in online surveys. Pew found few implausible answers in probability panels. Its finding comes from one study of six samples and may not apply to business panels.

Result format

The result is a judgement of the risk that careless or fraudulent respondents have inflated rare responses, with the implausible answers that support it. It reports the results for the under-30 and Hispanic subgroups first when the sample is opt-in.

Requirements on the material

PropertyRequirement typeRequirement
StructureApplies in some casesrequires=structured
Content typeAlways appliesrequires=tabular
Unit of analysisAlways appliesrequires=record; requires=dataset
How the sample was chosenApplies in some casesrequires=self-selected or opt-in
Level of detail availableAlways appliesrequires=record-level
Appraises source credibilityApplies in some casesno

Not restricted on: Variable types, Time structure, 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 neededcomparator
Not for (rules)none
Not for (text)The survey uses a probability panel, where few implausible answers were found. The supporting finding comes from one study of six samples and may not apply to business panels.

Choosing and ordering

Kind of techniquePreparation technique, runs in step 5.2
Phase tagverify and prepare
Other phase tagsnone
Kind of findingrisk; difference
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: deterministic.

StakesCheckPerson involvedCounts as a different method
exploratoryspot checkoptionalno
decision-informingspot checkoptionalno
decision-informingrepeat runsnoneno
decision-informinghuman review sampleneeded; left out below high stakesno
high-stakesspot checkoptionalno
high-stakesrepeat runsnoneno
high-stakeshuman review sampleneededno
high-stakessecond technique with different assumptionsnoneconditional
high-stakesindependent reviewneededno

Failure modes and limits

Model suitabilitymodel plus code
Typical model failureFlags real but unusual respondents as bogus, and does not compare rates with a probability sample.
Error sourcesinput data; statistical assumptions
Certainty limitsCertainty drops because the evidence base is one study and results for subgroups are imprecise (imprecision and indirectness).

Evidence and tag basis

Documented. 12 per cent of opt-in respondents under 30 claimed a licence to operate a nuclear submarine, against a true share that rounds to 0 per cent. (Pew Research Center 2024)

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