Skip to main content

Table 4 Variation matrix indicating the dispersion of each movement behavior relative to other movement behaviors within the work and leisure domains

From: A comparison of standard and compositional data analysis in studies addressing group differences in sedentary behavior and physical activity

Movement behavior

Sedentary

Stand

PA

 

Work

Sedentary

0

1.32

0.98

Stand

1.32

0

0.30

PA

0.98

0.30

0

 

Total variance = 0.86

 

Leisure

Sedentary

0

0.31

0.33

Stand

0.31

0

0.16

PA

0.33

0.16

0

 

Total variance = 0.27

  1. PA physical activity
  2. Numbers in cells show the variance in the data set of the log-ratio between behaviors stated by the row (numerator) and the column (denominator). Values close to 0 indicate that the two behaviors involved are consistently proportional. For example, during leisure, the variance in ln(PA/stand) is 0.16, suggesting that these two behaviors are highly proportional, or co-varying. “Total variance” indicates the total relative variability of the time-use composition during work and leisure. The formula to calculate total variance is presented in Pawlowsky-Glahn & Buccianti [28]