One questionnaire, several subscales: read the proposed structure
Follow the item-to-factor and factor-to-score map behind a vaping questionnaire rather than assuming every item belongs in one total.
No Vape · Updated

Identify the score whose structure is being claimed
A questionnaire may produce one total, several subscale scores, or both. Write down the scoring arrangement before reading a validation headline. The same printed questions can be grouped in different ways, so evidence about one arrangement should remain attached to that arrangement. A shortened form also needs its version identified.
COSMIN’s structural-validity section relates the measurement claim to dimensionality and the scoring model actually used. It distinguishes models supporting different arrangements rather than treating every factor analysis as interchangeable evidence. This is a guide to locating that information, not a formal factor-analysis appraisal or a method for inventing a clinical score.
COSMIN Initiative: COSMIN Manual Version2.0 — structural validity
Sketch an item-to-score diagram
Use a blank sheet with three columns: items, proposed dimensions, and reported scores. Copy labels from the report rather than giving groups your own psychological names. Note whether each dimension leads to its own subscale and whether the authors separately justify a total. When the mapping is unavailable, leave the connection open.
In a fictional example, a generic planning questionnaire groups items into organizing time and arranging materials. A study supports an account with those two dimensions, while an advertisement displays one overall number. The reader asks what evidence supports that overall number. The invented example supplies neither a real instrument nor an assertion that combining scores is necessarily wrong.
Read what was explored and what was tested
A methods section may name exploratory or confirmatory factor analysis. Preserve whether the work investigated a possible structure or examined a specified model, and locate its actual results. Merely seeing an analysis name does not tell you which structure was supported or whether the researchers changed the model during the work.
A diagram also needs its population and version. A result for one language, item set or group cannot silently validate a different interface questionnaire. You need not fit a model, choose a fit threshold or decide which alternative is best. Instead, ask which scoring arrangement the cited evidence addresses and which public claim extends beyond it.
Do not turn the factor labels into personal categories
A subscale name describes an intended measured concept within an instrument. It does not assign you a fixed personality, diagnose a condition or show what treatment you need. The interpretation of a numerical score is another question requiring the actual instrument and appropriate professional guidance. A tidy diagram alone establishes none of those conclusions.
Finish with a concise score map: this version used these items, this proposed structure was studied, these scores were reported, and this uncertainty remains. Item consistency, content coverage and effectiveness each need their own evidence. The map is useful when it prevents a broad validated label from hiding the difference between a supported subscale and an unexamined total.
Sources and scope
Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.
Written with AI assistance and checked against the cited sources. No clinical review is claimed. This guide covers practical planning, not individual medical treatment. How we research and write