The study proposes a definition of organizational humanization as the affirmation of autonomy, individuality and dignity within an organization. This framing distinguishes humanization from the absence of dehumanization. An organization might avoid overtly disrespectful treatment while remaining impersonal or inattentive, but a lack of denial does not establish that employees’ agency, distinctiveness or worth has been actively recognized. Autonomy in the OHS means meaningful influence over how, when and by what criteria organizational activities are performed, rather than complete freedom from the constraints of a role. Individuality concerns whether a person’s uniqueness is valued, embraced and allowed to be expressed. Dignity concerns self- and other-recognized worth through respectful interaction, recognition of contributions and equal treatment.

The OHS was not created as an unrelated theoretical domain. Its items were adapted from established instruments measuring work autonomy, work-group inclusion and workplace dignity, with three items assigned to each dimension. Autonomy items addressed influence over methods, scheduling and evaluation. Individuality items concerned the valuing and embracing of uniqueness and its permitted expression, drawing primarily on the Work Group Inclusion Measure. Dignity items covered respectful interaction, respect for contribution and equal respect, corresponding to components of the Workplace Dignity Scale. Wording was broadened from “work” to “activities” to permit potential use in non-work organizations, and factor-specific stems were introduced. The researchers did not obtain formal ratings from an independent expert panel, leaving further work needed on the adapted items’ relevance, clarity and coverage.

Participants were recruited through the Prolific crowdsourcing platform, with the survey administered through Qualtrics. Eligibility required full-time employment in the United States and at least one year in the current organization, a criterion intended to ensure sustained exposure to organizational practices and interactions. The questionnaire contained 57 items across six measures, five demographic questions and six attention checks; researchers randomized the order of measures and items. Of 501 completed responses, 54 were excluded for straightlining across three or more scales and six more for failing an attention check, leaving 441 participants in the primary analysis. This sample included 255 men, 183 women and three participants who selected “prefer not to say,” with representation across age groups, job levels, industries and organization sizes.

Confirmatory factor analysis supported the proposed higher-order structure. A one-factor model fit poorly, producing a comparative fit index of 0.82. The correlated three-factor and second-order models had identical fit because they are mathematically equivalent when a second-order factor contains exactly three first-order factors; both produced a comparative fit index of 0.99, a Tucker-Lewis index of 0.98, a root mean square error of approximation of 0.05 and a standardized root mean square residual of 0.03. A bifactor model also fit well. Standardized loadings for the nine individual items ranged from 0.79 to 0.93, while second-order loadings were 0.83 for autonomy, 0.93 for individuality and 0.80 for dignity. The bifactor analysis yielded explained common variance of 0.73 and omega hierarchical of 0.85, supporting interpretation of a general humanization score alongside the three subscales.

Internal consistency was strong across the instrument. For the full nine-item OHS, Cronbach’s alpha and McDonald’s omega were both 0.93; for the autonomy, individuality and dignity subscales, both estimates ranged from 0.86 to 0.93. Convergent validity also appeared. The OHS autonomy subscale correlated most strongly with the Work Autonomy Scales, at r = 0.79, while OHS dignity correlated most strongly with the Workplace Dignity Scale, at r = 0.85. The Work Group Inclusion Measure correlated equally with OHS individuality and OHS dignity, at r = 0.76 for each, indicating substantial convergence with the intended dimension without complete separation from related content. Average variance extracted was 0.67 for autonomy, 0.81 for individuality and 0.73 for dignity. Heterotrait-monotrait ratios of 0.66 to 0.78 remained below the 0.85 criterion, providing evidence that the subscales were empirically distinguishable while sharing a broader structure.

Structural equation models examined associations with workplace social capital, which includes networks, norms and trust that facilitate coordination and cooperation, and work engagement, measured through vigor, dedication and absorption. In the primary second-order model, the humanization factor was associated with social capital at a standardized beta of 0.87 and with engagement at beta = 0.69, both with p < 0.001. The model accounted for 75% of the variance in social capital and 47% of the variance in engagement. When the general humanization factor was modeled alongside autonomy, individuality and dignity, their shared component remained strongly associated with both outcomes. Dignity retained an additional association with social capital at beta = 0.37, while unique paths for autonomy and individuality were not significant in that specification. These cross-sectional self-report results do not establish causal direction or rule out reciprocal relationships among the constructs.

Measurement appeared comparable across several broad demographic and organizational divisions. Configural, metric and scalar invariance models supported item-level comparability across age groups, sex, management versus non-management positions, and organizations with 1,000 or fewer versus more than 1,000 employees. Industry-level invariance was not tested because group sizes were too small to combine defensibly, and equality of second-order loadings was not assessed. A sensitivity analysis that retained the 54 responses excluded for straightlining, producing a sample of 494, preserved the principal factor, reliability, validity and association results. The scale was administered only once, however, so its temporal stability and test-retest reliability remain unknown.

The most important caution concerns shared response method. The same participants completed the humanization, social-capital and engagement measures, using five-point agreement scales for the OHS and social capital and a seven-point frequency scale for engagement. A supplementary model allowing an added factor to load freely on all 20 indicators produced the best global fit and sharply reduced the humanization–social-capital path from beta = 0.866 to 0.041, with a 95% confidence interval of −0.259 to 0.341. A more constrained model, allowing the added factor to operate only across the 17 agreement-format indicators, preserved the primary estimates, including beta = 0.859 for social capital and beta = 0.687 for engagement. Neither specification proves that the added factor is purely methodological, and overlapping content between humanization and social capital may contribute substantively to their association. The magnitude of the relationship therefore remains unresolved.

For researchers, the OHS offers a compact starting point for studying whether employees perceive their autonomy, individuality and dignity as affirmed, including in settings involving restructuring, growth or artificial-intelligence adoption. The study itself did not measure AI implementation or organizational resilience, and its findings cannot show that a particular policy raises humanization, social capital or engagement. The scale records individual perceptions rather than independently observed practices, so its scores should not be interpreted as a direct organizational audit. Future studies should replicate the factor structure across countries, industries and employment arrangements, test the scale in non-work organizations, compare it directly with established dehumanization measures, obtain expert ratings of its items and use longitudinal, multimethod designs to examine change and temporal stability. The central scientific advance is a testable framework for examining whether organizations affirm people as autonomous, distinct and dignified, rather than merely avoiding obvious mistreatment.

Subject of Research: Initial validation of a nine-item measure of perceived organizational humanization

Article Title: Organizational humanization: definition and measurement

Article References: Stroop, K. D. (2026). Organizational humanization: definition and measurement. Frontiers in Psychology, 17, Article 1906924. https://doi.org/10.3389/fpsyg.2026.1906924

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1906924

Keywords: organizational psychology, workplace humanization, measurement, autonomy, individuality, workplace dignity, psychometrics, work engagement, social capital, employee perceptions, Organizational, humanization