Electronic Nicotine Vaping Outcomes Scale (ENVO)

Abstract

The Electronic Nicotine Vaping Outcomes Scale (ENVO), introduced by Harrell et al. in 2023, is a robust, 32-item psychological instrument meticulously designed to quantify vaping expectancies, or the outcome beliefs associated with using Electronic Nicotine Delivery Systems (ENDS). The scale addresses a critical gap in existing literature, which often failed to capture the nuances of vaping beliefs specific to contemporary and racially diverse young adult populations.

The development procedure was stringent, commencing with an extensive pool of 150 items derived from prior cigarette smoking expectancy measures and refined through focus group data collected from young adults. This pool was initially reduced to 51 items following refinement by a Delphi panel. Subsequent psychometric evaluation involved administering the 51-item version to a diverse sample of college students across various U.S. educational settings, including community colleges, four-year universities, and a Historically Black College or University (HBCU).

Final statistical analyses, including Factor Analysis and Item Response Theory (IRT), confirmed a final structure of 32 items loading onto five distinct factors. The validation study provided strong evidence supporting the scale’s reliability, validity, and its critical adherence to measurement invariance across key demographic groups.

Keywords

Addiction, College Students, E-Cigarettes, Electronic Nicotine Vaping, Expectations, Health Outcomes, Incremental Validity, Negative Affect Reduction, Nicotine, Positive Reinforcement, Substance Use and Addiction Measures, Vaping, Vaping Expectancies, Weight Control

Authors

Paul T. Harrell, Thomas H. Brandon, Stephen E. Stark, Vani N. Simmons, Tracey E. Barnett, Gwendolyn P. Quinn, Seokjoon Chun

[quads id=5]

Purpose

The fundamental purpose of the ENVO scale is to offer a comprehensive and psychometrically sound assessment of the anticipated outcomes, or expectancies, that young adults hold regarding the use of electronic nicotine vaping products. This assessment is crucial for understanding motivational pathways toward the adoption and continued use of ENDS.

The instrument was specifically developed to be sensitive to a diverse range of outcomes, including positive reinforcement (e.g., pleasure, sensory feedback), negative reinforcement (e.g., mood regulation), and perceived negative consequences (e.g., health risks, addiction potential), thereby offering a detailed profile of an individual’s psychological relationship with vaping.

Construct

The primary psychological construct measured by the ENVO scale is Electronic Nicotine Vaping Outcomes (or Vaping Expectancies). This construct encompasses the set of beliefs an individual holds about the physical, social, and psychological effects that they expect to occur as a result of using electronic cigarettes.

The scale operationalizes this broad construct into five empirically derived subscales, reflecting the multidimensional nature of substance use expectancies in the context of modern vaping devices, ensuring relevance to contemporary usage patterns.

Validity

The validation studies provided robust statistical support for the ENVO scale across multiple domains of validity.

Convergent and Divergent Validity: The scale demonstrated expected correlational patterns. Vaping susceptibility and lifetime vaping history showed significant positive correlations with subscales reflecting rewarding outcomes, such as Positive Reinforcement (PR) and Negative Affect Reduction (NAR), and marginal correlations with Addiction and Weight Control (WC). Crucially, both susceptibility and history were negatively correlated with the Negative Consequences (NC) subscale (e.g., PR: r = .49, p < .001; NC: r = -.18, p < .001 for lifetime history), supporting divergent validity.

Incremental Validity: Hierarchical linear regression analyses confirmed the measure’s value by demonstrating its incremental validity. The factors of the ENVO scale were statistically significant predictors of lifetime vaping behavior, even after controlling for established covariates like demographic variables, exposure to advertising, and the influence of peer and family vaping habits, confirming its utility beyond simple demographic data.

Item Response Theory (IRT) Characteristics: Further advanced validation was provided through IRT analyses. These results confirmed that individual items effectively measure their intended latent constructs, evidenced by discrimination parameters (‘a’ parameters) ranging from 1.26 to 3.18. The breadth of the items across the continuum of expectancies was also supported, with difficulty parameters (‘b’ parameters) spanning from 0.72 to 2.47.

Reliability

The ENVO scale exhibits excellent internal consistency across all five derived subscales, demonstrating high reliability within the sample tested.

Internal consistency was assessed using Cronbach’s alpha coefficients, with all factors yielding values well above the acceptable threshold of .70, indicating strong inter-item correlation within each factor:

• Positive Reinforcement: α = .92

• Negative Consequences: α = .94

• Negative Affect Reduction: α = .95

• Weight Control: α = .92

• Addiction: α = .87

Factor Analysis

The scale structure was established through a rigorous two-step process involving Exploratory and Confirmatory Factor Analysis (FA).

Exploratory Factor Analysis (EFA): An initial 51-item pool was subjected to EFA using a split-half sample. Items were systematically deleted due to cross-loadings (secondary loadings > .2) or conceptual redundancy, resulting in the elimination of 19 items. Multiple statistical criteria—including the Kaiser-Guttman criterion, Catell’s scree test, parallel analysis, and Velicer’s MAP test—unanimously supported a robust 5-factor solution. The final 32 items demonstrated very high primary factor loadings, ranging from .60 to 1.00.

Confirmatory Factor Analysis (CFA) and Measurement Invariance: The hypothesized 5-factor model was subsequently tested using CFA and demonstrated excellent fit indices (CFI = 0.95, TLI = 0.94, RMSEA = 0.05). Furthermore, the measure exhibited strong measurement invariance (configural, scalar, and metric) across critical subgroups, including tobacco use status, lifetime and current vaping status, self-reported gender, and race (e.g., white vs. other, black vs. other). This invariance confirms that the scale measures the same construct consistently across diverse populations.

Instrument

Test Type: Original Inventory/Questionnaire

Format: Items are rated on a 5-point Likert-type scale, with response options ranging from 1 (strongly disagree) to 5 (strongly agree).

Language Available: English

Population Group: Human; Male; Female

Age Group: Adulthood (18 years and older); Young Adulthood (18–29 years)

Population Details: The validation sample consisted of diverse young adult college students recruited from three distinct U.S. institutions: a Community College, a Four-Year University, and a Historically Black College or University (HBCU).

Test Methodology: The rigorous development involved Test Validity (Convergent, Discriminant, Incremental), Test Reliability (Internal Consistency using Cronbach’s alpha), Factor Analysis (Exploratory and Confirmatory), Item Response Theory, and Measurement Invariance analyses.

Keywords

Addiction, College Students, Electronic Cigarettes, Nicotine, Electronic Nicotine Vaping, Expectations, Health Outcomes, Negative Affect Reduction, Positive Reinforcement, Vaping Expectancies, Weight Control, Psychometrics, ENDS

[quads id=5]

Authors

Harrell, Paul T.

• Author ORCID Identifier: No data is Available

• Affiliation: Department of Pediatrics, Division of Community Health & Research Eastern Virginia Medical School (EVMS)

• Email addresses: [email protected]

• Correspondence Address: Eastern Virginia Medical School, Division of Community Health & Research, P.O. Box 1980, Norfolk, Virginia, United States, 23501-1980

Brandon, Thomas H.

• Author ORCID Identifier: No data is Available

• Affiliation: Department of Health Outcomes and Behavior, Moffitt Cancer Center

Stark, Stephen E.

• Author ORCID Identifier: No data is Available

• Affiliation: Department of Psychology, University of South Florida

Simmons, Vani N.

• Author ORCID Identifier: No data is Available

• Affiliation: Department of Health Outcomes and Behavior, Moffitt Cancer Center

Barnett, Tracey E.

• Author ORCID Identifier: http://orcid.org/0000-0002-2932-1540

• Affiliation: Department of Biostatistics and Epidemiology, School of Public Health University of North Texas, Health Science Center

Quinn, Gwendolyn P.

• Author ORCID Identifier: No data is Available

• Affiliation: Department of Obstetrics & Gynecology, Grossman School of Medicine, New York University

Chun, Seokjoon

• Author ORCID Identifier: No data is Available

• Affiliation: Department of Psychology, University of South Florida

Permissions & Fee and Test Year

Permissions: The scale may be used freely for academic research and teaching purposes.

Fee: No fee is required for non-commercial use.

Commercial: Not authorized for commercial use.

Test Year: 2023

Reference’s

Harrell, P. T., Brandon, T. H., Stark, S. E., Simmons, V. N., Barnett, T. E., Quinn, G. P., & Chun, S. (2023). Measuring vaping-related expectancies in young adults: Psychometric evaluation of the electronic nicotine vaping outcomes (ENVO) scale. Drug and Alcohol Dependence, 246, 1–9. DOI: https://doi.org/10.1016/j.drugalcdep.2023.109861

[quads id=5]

Items of the Electronic Nicotine Vaping Outcomes Scale (ENVO)

This measure is composed of 32 individual items, which are grouped into five distinct factors or subscales. The complete list of the 32 test items is available for review in the source publication, specifically in Table 2 on page 5.

1. Positive Reinforcement

2. Negative Consequences

3. Negative Affect Reduction

4. Weight Control

5. Addiction

Positive Reinforcement

Negative Consequences

Weight and Appetite Control

Addiction

Negative Affect Reduction / Negative Reinforcement

Note. Response options range from 1 (strongly disagree) to 5 (strongly agree).