# Psychometric evaluation of the e-cigarette knowledge, attitude, and practice questionnaire among young students: a structural equation modeling approach

## TL;DR

This study developed and validated a questionnaire to assess young students' knowledge, attitudes, and practices regarding e-cigarettes using structural equation modeling.

## Contribution

The study introduces a validated questionnaire for evaluating e-cigarette knowledge, attitude, and practice among young students.

## Key findings

- The final questionnaire had 15 items with acceptable reliability and validity.
- The three-factor structure was confirmed with good model fit indices.
- Subscale Cronbach’s α values ranged from 0.775 to 0.847, indicating strong internal consistency.

## Abstract

The growing popularity of e-cigarettes among young students has emerged as a pressing public health concern. Misconceptions about e-cigarettes persist widely, yet standardized tools to systematically evaluate youth knowledge, attitudes, and behaviors are limited. This study aimed to develop and validate a questionnaire evaluating e-cigarette knowledge, attitude, and practice among young students using a structural equation modeling framework.

An initial e-cigarette knowledge, attitude, and practice questionnaire was developed and refined using classical test theory, including item analysis and exploratory factor analysis. A sample of 1,400 students undergraduate students were recruited from Kunming, China and Jakarta, Indonesia, yielding 1,333 valid responses (Kunming, n = 631; Jakarta, n = 702). Reliability was assessed using Cronbach’s α, and model fit was examined through confirmatory factor analysis within the SEM framework.

The final questionnaire retained 15 items: knowledge (4), attitude (8), and practice (3). Pearson correlation coefficients between each item and its respective dimension exceeded 0.4. Except for items P2 and P3, t-test results between high- and low- scoring groups were statistically significant (p < 0.05). The final model (M3) demonstrated excellent fit indices (χ2/df = 1.350, RMSEA = 0.050 [90% CI: 0.022–0.072], SRMR = 0.0643, CFI = 0.968, TLI = 0.960), indicating optimal model parsimony and goodness of fit. The overall Cronbach’s α was 0.643, with subscale coefficients of 0.819 (knowledge), 0.847 (attitude), and 0.775 (practice). Exploratory factor analysis confirmed the expected three-factor structure, with factor loadings between 0.521 and 0.819 and a cumulative variance explanation of 61.943%.

The final instrument demonstrated acceptable reliability and validity within the sampled populations from China and Indonesia. This tool provides a practical and evidence-based approach to assess young students’ perceptions and attitudes towards e-cigarettes in public health research.