Implementation of multivariate linear mixed-effects models in the analysis of indoor climate performance experiments

Area of Science:

• Environmental Health
• Occupational Psychology
• Statistical Modeling

Background:

• Air quality is a critical factor influencing cognitive functions and workplace productivity.
• Traditional statistical methods often analyze performance dimensions separately, potentially missing interdependencies.
• Understanding the relationship between environmental factors and human performance is vital for optimizing work environments.

Purpose of the Study:

• To apply multivariate mixed-effects modeling to analyze the relationship between air quality and office work performance.
• To estimate the effects of air quality on multi-dimensional response variables, including subjective perceptions and objective performance tasks.
• To uncover correlations between different dimensions of performance that are often overlooked in univariate analyses.

Main Methods:

• Utilized multivariate mixed-effects modeling to analyze data from three experimental series.
• Integrated subjective perceptions and a two-dimensional performance task outcome into a single analytical framework.
• Estimated the direct effects of air quality exposure and the correlations among response variables.

Main Results:

• Identified a significant positive correlation between two distinct performance tasks.
• Demonstrated that the multivariate approach captures inter-dimensional performance correlations.
• The analysis provided a more comprehensive understanding of performance metrics compared to univariate methods.

Conclusions:

• Multivariate mixed-effects modeling offers a superior analytical approach for performance studies compared to conventional univariate statistics.
• The identified correlations between performance tasks suggest they measure overlapping aspects of mental performance.
• Findings have implications for designing more robust performance experiments and interpreting their results accurately.