International Journal of Biometeorology

, Volume 56, Issue 1, pp 129–136

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

Authors

    • ALECTIA A/S
  • Henrik Spiild
    • Department of Infomatics and Mathematical ModelingTechnical University of Denmark
  • Jørn Toftum
    • International Centre for Indoor Environment and Energy, Department of Civil EngineeringTechnical University of Denmark
Original Paper

DOI: 10.1007/s00484-011-0404-y

Cite this article as:
Jensen, K.L., Spiild, H. & Toftum, J. Int J Biometeorol (2012) 56: 129. doi:10.1007/s00484-011-0404-y

Abstract

The aim of the current study was to apply multivariate mixed-effects modeling to analyze experimental data on the relation between air quality and the performance of office work. The method estimates in one step the effect of the exposure on a multi-dimensional response variable, and yields important information on the correlation between the different dimensions of the response variable, which in this study was composed of both subjective perceptions and a two-dimensional performance task outcome. Such correlation is typically not included in the output from univariate analysis methods. Data originated from three different series of experiments investigating the effects of air quality on performance. The example analyses resulted in a significant and positive correlation between two performance tasks, indicating that the two tasks to some extent measured the same dimension of mental performance. The analysis seems superior to conventional univariate statistics and the information provided may be important for the design of performance experiments in general and for the conclusions that can be based on such studies.

Keywords

Experimental designIndoor air qualityMultivariate mixed-effects modelingPerformanceStatistical analysis

Supplementary material

484_2011_404_MOESM1_ESM.doc (28 kb)
ESM 1(DOC 28 kb)

Copyright information

© ISB 2011