startthat.work home

CM1-03 Mill's method of difference and most similar systems design (MSSD; joint method of agreement and difference)

Looks for the factor present where an outcome occurs and absent in an otherwise similar case where it does not, producing a short list of the remaining differences between matched cases.

CompareEvidence: DocumentedPrimary technique, runs in step 5.3Strongest statement: show a link between things

What it does

Looks for the factor present where the outcome occurs and absent in an otherwise similar case where it does not. Artefact: the short list of remaining differences between matched cases.

When to use it

Use when a user has a case with an outcome and a close match without it and asks what made the difference.

When it does not fit

The cases still differ on many characteristics. Przeworski and Teune concede that no two countries differ in only two characteristics, so rival hypotheses always remain; cross-national statistics also differ in definition and coverage.

Result format

The result sets the case where the outcome occurs beside an otherwise similar case where it does not, and gives a short list of the remaining differences between the matched cases. It names the factor present where the outcome occurs and absent where it does not.

Requirements on the material

PropertyRequirement typeRequirement
Content typeAlways appliesrequires=text; requires=tabular
Unit of analysisAlways appliesrequires=topic; requires=record
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 materialAt least one case with the outcome, one similar case without it, and detailed background data on both.
Additional sources needednone
Not for (rules)none
Not for (text)Wrong choice when the cases still differ on many characteristics, because rival hypotheses then remain. No two cases differ in only one or two characteristics, and cross-national statistics differ in definition and coverage.

Choosing and ordering

Kind of techniquePrimary technique, runs in step 5.3
Phase taganalyse
Other phase tagsnone
Kind of findingcause; difference
Question answeredwhy; what
Strongest statement it supportsshow a link between things
Needs as input
Makes as output
Must exist firstNone
Loop targetreturn to gather

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-informingalternative explanations checknoneno
decision-informingspot checkoptionalno
high-stakesassumption and limitation statementnoneno
high-stakesalternative explanations checknoneno
high-stakesspot checkoptionalno
high-stakessecond technique with different assumptionsnoneconditional
high-stakesexpert reviewneededno

Failure modes and limits

Model suitabilityhuman judgement required
Typical model failureTreats two cases as matched without checking background differences, and names the single visible difference as the cause.
Error sourcesinput data; source selection; coder or rater judgement
Certainty limitsCertainty drops when the matched cases still differ on several factors, when definitions or coverage differ between the cases (inconsistency), or when the pair was chosen after the outcome was known (risk of bias).

Evidence and tag basis

Documented. As for the method of agreement. The joint method is described as weaker than the direct method of difference. (Britannica, Mill's methods; Wikipedia, Mill's Methods; Przeworski and Teune design summary)

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