Skip to content
No Vape

An ordered vaping-survey outcome is not one yes-or-no event

Read a proportional-odds result through its category order and cumulative splits before interpreting a quoted odds ratio.

No Vape · Updated

Disposable vape resting on concrete
Illustrative image · Markus Winkler / Unsplash

Find the actual ordered response

The published study of perceived social unacceptability of vaping among regular users in a Middle Eastern sample describes ordinal regression and checks of the proportional-odds assumption. Its outcome concerns reported perceptions. That methods example is distinct from measuring abstinence or proving that any perception is medically correct.

Copy the outcome categories in their reported order and read how they were coded. A numerical label can represent rank without equal spacing between adjacent categories. Do not turn a model of ordered answers into an arithmetic score-change claim simply because the categories were numbered in a dataset.

Frontiers in Public Health: Correlates of perceived social unacceptability of vaping among regular e-cigarette users: a cross-sectional study of a sample of Middle Eastern countries

Draw the cumulative splits in a fictional response

Imagine an invented three-category response: low, middle and high. One cumulative split separates low from middle-or-high; another separates low-or-middle from high. Those comparisons use the same ordered categories differently from asking only whether a person chose middle. The example is a reading sketch, not a validated vaping questionnaire.

The UVA explanation describes cumulative category comparisons and a shared slope in the proportional-odds model. Check the direction used in the actual software and paper before saying higher or lower odds. Reversing the category ordering can reverse your verbal description even though the underlying observations did not change.

University of Virginia Library: Fitting and Interpreting a Proportional Odds Model

Retain the proportional-odds assumption

A common slope across cumulative splits is a model assumption, not proof that every adjacent category has the same probability. Find how the report describes assessing it and whether it used a different model when needed. Do not decide that one reported test settles every aspect of fit or that the assumption applies because the outcome has an obvious order.

This task does not select a statistical test or tell you to merge categories. If the article does not explain its model, write the precise question you cannot answer. Reconstructing a different binary outcome from a headline would silently change the analysis instead of clarifying the original one.

Summarize the relationship at the scale reported

A careful sentence identifies the ordered outcome, direction of its categories, model and the stated estimate with uncertainty. It does not translate an odds ratio into the number of category steps each participant supposedly moved. A comparison across people is also different from observing a particular person change over time.

The method cannot tell you which clinical support to choose, validate a personal craving scale or classify your health. Where a consequential decision depends on it, seek qualified interpretation of the full report. For ordinary preparation, describe your actual situation in words you understand rather than importing an ordinal coefficient as a personal target.

Sources and scope

Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.

  1. Frontiers in Public Health — Correlates of perceived social unacceptability of vaping among regular e-cigarette users: a cross-sectional study of a sample of Middle Eastern countries
  2. University of Virginia Library — Fitting and Interpreting a Proportional Odds Model
  3. No Vape — No Vape — app features and store links

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