People who reported stronger management support also tended to report greater compliance with laboratory safety policies, according to a questionnaire survey of 510 respondents in university biomedical laboratories. The pattern included both a direct association between management support and compliance and indirect links through perceived trust and self-efficacy.
The survey captured responses at one point in time, so it cannot show that management support caused trust, self-efficacy or compliance. The relationships are correlational rather than proof of a cause-and-effect chain.
Trust and confidence were part of the pattern
Management support was associated with both perceived trust and self-efficacy. On the study's standardized scale, the reported coefficients were 0.809 for trust and 0.798 for self-efficacy, with reported test statistics of 18.478 and 18.552, respectively.
Perceived trust and self-efficacy were each associated with reported compliance. The coefficient was 0.411 for perceived trust and 0.248 for self-efficacy, with reported test statistics of 3.956 and 3.265. The estimated indirect association through trust was 0.332, compared with 0.198 through self-efficacy. The reported 95% bootstrap confidence intervals ranged from 0.190 to 0.564 for the trust pathway and from 0.082 to 0.335 for the self-efficacy pathway.
The analysis also reported a direct management-support and compliance association of 0.247, with a p-value of 0.003. Taken together, the results describe both a direct association and indirect associations through the two measures. The reported total indirect coefficient was 0.530.
The larger point estimate for the trust pathway should not be read as evidence that trust matters more than self-efficacy. The 95% confidence interval for the difference between the two indirect associations ran from -0.117 to 0.465 and included zero, so the study did not find a statistically significant difference between them.
A concentrated sample
The researchers used a questionnaire survey with convenience sampling and set allocation targets for undergraduate students, graduate students and in-service researchers because no complete sampling frame was available. The final analysis included 510 respondents: 478 undergraduates, 14 graduate students and 18 in-service researchers. Undergraduates made up 93.72% of the sample, while graduate students accounted for 2.75% and in-service researchers for 3.53%.
The analysis used partial least squares structural equation modeling in SmartPLS 4.0 with reflective indicators. Researchers used 5,000 bootstrap subsamples to calculate 95% bias-corrected and accelerated confidence intervals. Less than 1% of responses were missing and were handled by listwise deletion.
Why the findings need caution
Convenience sampling and the makeup of the respondents narrow the scope of the finding. Undergraduates accounted for 93.72% of the sample, while graduate students and in-service researchers accounted for 2.75% and 3.53%, respectively. The evidence therefore comes from a concentrated university-laboratory sample rather than a broad population.
The questionnaire design also raised a risk of common method bias, in which the same measurement approach can push several answers in the same direction. Harman's single-factor test explained 64.17% of the variance, while full-collinearity variance inflation factor values ranged from 3.57 to 7.00. The authors warned that the reported coefficients, especially mediation estimates, may be inflated.
The reported model diagnostics included a standardized root mean square residual of 0.038 and a normed fit index of 0.915. The model explained 63.7% of the reported variation in self-efficacy, 65.5% in perceived trust and 69.8% in compliance behavior. Those figures describe the model's fit to these survey responses, but they do not overcome the cross-sectional limit on causal inference.
Taken together, the results support an association model in which management support is related to compliance both directly and through trust and self-efficacy. They do not show that management support caused any of these outcomes, or that trust was more influential than self-efficacy.
The study was funded by the Youth Doctoral Innovation Program of Zhejiang Shuren University under grant 2019QC09. The authors said original contributions are included in the article or supplementary material, with further inquiries directed to the corresponding author.
Paper data and sources
Original title: Direct and indirect associations of management support with compliance with laboratory safety policies: the mediating roles of perceived trust and self-efficacy
Authors: Hong L, Yang K, Qu J
Journal/Repository: Frontiers in public health
Status: Peer-reviewed
First online: 2026-10-08
DOI: Not available
Original paper · Full text