The study used structural equation modeling to assess an integrated model linking transformational and servant leadership with workplace harassment, teacher self-efficacy and life satisfaction. The sample included 49.6% men and 50.4% women, with 61.66% of participants working at public institutions. The authors directly estimated residual paths from both leadership styles to life satisfaction, allowing them to test whether self-efficacy accounted for part of the relationships between leadership and well-being.

Ordinal indicators were analyzed with robust weighted least squares mean and variance adjusted estimation, with maximum likelihood estimation used as a sensitivity analysis. The researchers evaluated indirect effects using the product-of-coefficients method and bootstrap bias-corrected and accelerated confidence intervals based on 5,000 resamples. This design tested statistical relationships among measured variables rather than effects produced by a controlled leadership program or another intervention.

Both leadership styles were negatively associated with workplace harassment, which the study defined as systematic, recurrent negative acts that constrain a target’s capacity for self-defense. The estimated coefficient was -0.123 for transformational leadership and -0.186 for servant leadership. Servant leadership had the strongest reported association with teacher self-efficacy, with a coefficient of 0.505, while self-efficacy was positively associated with life satisfaction, with a coefficient of 0.429. The model accounted for 39.3% of the variance in life satisfaction.

Servant leadership showed what the authors described as complementary partial mediation through self-efficacy. Its standardized indirect effect through self-efficacy was 0.217, with a bias-corrected and accelerated bootstrap confidence interval from 0.146 to 0.354. A direct residual effect on life satisfaction was also significant, with a coefficient of 0.251 and a bootstrap confidence interval from 0.198 to 0.597. Transformational leadership had a smaller significant indirect effect through self-efficacy, with a standardized coefficient of 0.033 and a confidence interval from 0.003 to 0.101, but its direct path to life satisfaction was not significant.

Paths running through workplace harassment were not significant in the overall model. The researchers next tested whether measurements and structural relationships could be compared across gender groups. Measurement invariance was supported across configural, metric and scalar models, with no more than a 0.003 change in the comparative fit index at any step. This supported proceeding with a formal comparison of the structural paths between women and men.

Full structural equivalence between the gender groups was rejected, with a chi-square difference of 17.86 for eight degrees of freedom and a p-value of 0.022. Wald tests localized the differences to transformational-leadership paths rather than servant leadership or the self-efficacy-to-life-satisfaction relationship. Transformational leadership was associated with less harassment among women, with a coefficient of -0.238, while the corresponding association among men was not significant, at -0.022. Conversely, transformational leadership was associated with higher self-efficacy among men, with a coefficient of 0.157, but not among women, whose coefficient was -0.024.

An additional model adjusted for age, university type and region, and the relevant path coefficients changed by no more than 0.012 in absolute value, which the authors described as evidence of robustness to those characteristics. The findings therefore support different statistical patterns for the two leadership styles in this sample: servant leadership was associated with faculty well-being outcomes in a similar way for women and men, whereas transformational leadership showed gender-specific associations. Because the study was cross-sectional and did not test an intervention, it cannot determine whether changing leadership practices would alter harassment, self-efficacy or life satisfaction, and the sample of Peruvian university teachers also limits how broadly the results may apply.

Subject of Research: Leadership styles, harassment, self-efficacy and life satisfaction among Peruvian university teachers

Article Title: Multivariate modeling of organizational dynamics and faculty well-being in higher education

Article References: Alave Mamani, C. G., Tocto-Cano, E., Gonzales Medina, R. I., López-Gonzales, J. L., & Turpo-Chaparro, J. E. (2026). Multivariate modeling of organizational dynamics and faculty well-being in higher education. BMC Psychology, 14(1), Article 1387. https://doi.org/10.1186/s40359-026-05450-0

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05450-0

Keywords: servant leadership, transformational leadership, faculty well-being, workplace harassment, teacher self-efficacy, life satisfaction, gender differences, higher education, structural equation modeling, Peru, Multivariate, modeling