Skip to main content
. 2014 Mar 28;11:44. doi: 10.1186/1479-5868-11-44

Table 3.

Results of correlated component regression analyses for healthy dietary intake outcome measures

    BOYS (n = 275)
 
    GIRLS (n = 353)
 
Variable CV predictor count a β Model goodness of fit indices b Variable CV predictor count a β Model goodness of fit indices b
Fruit intake (8 predictors)
 
 
Fruit intake (8 predictors)
 
 
 SEIFA*
100
0.138
R2 = 0.052
 Employment*
100
−0.122
R2 = 0.014
 Mother’s education*
100
0.128
R2(CV) = 0.011
 Child age*
89
−0.063
R2(CV) = 0.012
 Employment*
96
−0.087
SD (CV) = 0.004
 Mother’s occupation*
89
−0.063
SD (CV) = 0.009
 Mother’s occupation*
92
0.067
 
 Marital status*
82
0.033
 
 Marital status*
82
0.044
 
 Mother’s education*
77
−0.010
 
 Mother’s age*
79
−0.040
 
 Mother’s age*
71
−0.013
 
 Child age*
64
−0.035
 
 SEIFA*
71
−0.002
 
 Income*
47
0.024
 
 Income*
71
−0.006
 
Vegetable intake (1 predictor)
 
 
Vegetable intake (1 predictor)
 
 
 Mother’s education*
90
0.120
R2 = 0.015
 Employment*
83
−0.099
R2 = 0.010
 SEIFA
67
 
R2(CV) = 0.007
 Mother’s education
74
 
R2(CV) = 0.005
 Employment
42
 
SD (CV) = 0.009
 Mother’s occupation
10
 
SD (CV) = 0.003
 Mother’s age
39
 
 
 SEIFA
9
 
 
 Child age
22
 
 
 Income
8
 
 
 Income
17
 
 
 Mother’s age
7
 
 
 Mother’s occupation
2
 
 
 Child age
5
 
 
 Marital status
1
 
 
 Marital status
4
 
 
Healthy behavioursC (4 predictors)
 
Healthy behavioursC (1 predictors)
 
 
 Employment*
100
−0.116
R2 = 0.036
 Marital status*
54
0.077
R2 = 0.006
 Marital status*
59
0.076
R2(CV) = 0.008
 SEIFA
31
 
R2(CV) = 0.022
 Mother’s occupation*
57
0.082
SD (CV) = 0.004
 Child age
16
 
SD (CV) = 0.013
 Mother’s education*
48
0.062
 
 Mother’s age
12
 
 
 Income
12
 
 
 Mother’s occupation
10
 
 
 SEIFA
3
 
 
 Income
3
 
 
 Mother’s age
1
 
 
 Employment
3
 
 
 Child age^ --      Mother’s education 1    

*Predictor retained in final model.

^Predictor not retained in any model.

ß = standardised regression coefficient.

aCross-validation predictor count - Represents number of regressions in which predictor appeared. Predictor count of 100 indicates that predictor was present in all 100 regressions. Indicates importance of predictor together with standardised regression coefficient (β).

bModel goodness of fit indices: R2(CV) = cross-validated R2; SD (CV) = Standard deviation for cross-validated R2.

cHealthy behaviours: Breakfast intake, carrying water bottle, help parents with groceries, help to prepare dinner, eat dinner with the family.