Skip to main content
. 2017 Feb 8;17:175. doi: 10.1186/s12889-017-4070-9

Table 4.

Multivariate linear regression (for continuous outcomes) or logistic regression (for categorical outcomes) of association of socioeconomic status measures and physical activity (PA) and sitting time among boys

Total PA score
(MET-min/week)
Vigorous PA score
(MET-min/week)
Moderate PA score
(MET-min/week)
Walking score
(MET-min/week)
Active commuting Sport participation Sitting time(min/day)
β P β P β P β P OR [95% CI] P OR [95% CI] P β P
unstd (std) eta2 unstd (std) eta2 unstd (std) eta2 unstd (std) eta2
Socioeconomic characteristics
 Social, professional class of the family head 0.0148
0.003
  Executives, intermediate jobs, farmers, shopkeepers, craftsmen, managers 0
  Employees and workers (unskilled or skilled) 333.6 (0.08) 0.0061
  Retired, inactive, unemployed 421.5 (0.04) 0.1403
 Family income level 0.0135
0.005
0.0092
  Low or average 0 1
  High 299.4 (0.07) 1.4 [1.1; 1.7]
 Residence area 0.0448
0.002
<0.0001 0.0033
  Urban 0 1 1
  Rural 123.3 (0.06) 0.5 [0.4; 0.7] 1.4 [1.1; 1.8]
School-related characteristics
 School type
  General and technological
  Professional
 Schooling placement 0.0231
0.004
  Typical or advanced 0
  Late 106.1 (0.06)
 School boarding status 0.0017
0.01
<0.0001
  Non-boarder 0 1
  Half-boarder −252 (−0.011) 0.0013 0.3 [0.2; 0.5]
  Full-Boarder −52.5 0.6544 1.9 [1.2; 3.0]
Family characteristics
 Family composition 0.0021
0.009
  Two-parents 0
  Single-parent 192.3 (0.09)
 Perceived parental PA level <0.0001
0.01
<0.0001
0.16
0.0006
0.007
0.0162
  Low or average 0 0 0 1
  High 625.7 (0.11) 482.3 (0.12) 204.3 (0.1) 1.3 [1.1; 1.7]

β linear regression coefficient, OR odds ratio (logistic regression) and [95% CI confidence interval], P p-value, statistically significantly (p <0.05), unstd unstandardised, std standardised, unstandardised β divided by the ratio of the standard error of the dependent variable to the standard deviation of the regression, eta semi partial ŋ2 correlation ratio of the squared semi partial correlation