Demographics
Table 1 displays demographics data for the total sample, including head circumference. The average age of children with ASD was 4.20 and sex ratio was 5.2:1 which similarly to sex bias in epidemiological surveys, demonstrating that age and sex ratio of the sample was representative in preschool-aged children with ASD. There were no statistical differences in age and sex ratio between the ASD with IQ ≥ 70 group and IQ<70 group (all P > 0.05). In terms of diagnosis, age of diagnosis was significantly delayed in ASD with IQ ≥ 70 (t = 2.11, P = 0.037, Cohen's d = 0.36). Meanwhile, age when parents firstly noticed developmental abnormalities was also later in ASD with IQ ≥ 70 (t = 2.07, P = 0.041, Cohen's d = 0.33). The results showed that higher IQ could lead to a delay in diagnosis of ASD. Head circumference as endophenotype of autism, was not correlated with global or domain-specific components of adaptive skills (all P > 0.05). No significant difference was found in head circumference between the two groups (Z = 0.56, P = 0.579). In the basic birth survey, no statistically significant differences were found in birth weight, delivery method and gestational age between the two groups (all P > 0.05). As for the investigation of socioeconomic factors, there were no statistical differences in parents' highest education level and family income between the two groups (all P > 0.05).
Table 1
Comparison in demographics, diagnosis and SES between ASD with IQ ≥ 70 and with IQ < 70
|
Total sample
|
With IQ < 70
|
With IQ ≥ 70
|
With IQ < 70 vs With IQ ≥ 70
|
|
N = 151
|
n = 88
|
n = 63
|
t/Z/χ2
|
P
|
Effect Size
|
Age (years)
|
4.20(0.90)
|
4.23 (0.84)
|
4.15(0.98)
|
0.16*
|
0.874
|
0.09
|
Age of diagnosis (months)
|
41.64(12.04)
|
39.85(10.86)
|
44.13(13.22)
|
2.11*
|
0.037
|
0.36
|
Age of first visit (months)
|
35.93(12.01)
|
34.20(10.54)
|
38.35(13.53)
|
1.84*
|
0.068
|
0.35
|
Age when parents firstly noticed developmental abnormalities (months)
|
25.84(8.68)
|
24.65(8.69)
|
27.51(8.47)
|
2.07*
|
0.041
|
0.33
|
Gender (%)
|
|
|
|
0.83†
|
0.363
|
0.23
|
Male
|
127 (84.1%)
|
72 (81.8%)
|
55 (87.3%)
|
|
|
|
Female
|
24 (15.9%)
|
16 (12.2%)
|
8 (12.7%)
|
|
|
|
Head circumference
|
|
|
|
0.56
|
0.579
|
0.08
|
-2SD under
|
6 (4.0%)
|
4 (4.5%)
|
2 (3.2%)
|
|
|
|
-2SD~-1SD
|
20 (13.2%)
|
14 (15.9%)
|
6 (9.5%)
|
|
|
|
-1SD~+1SD
|
97 (64.2%)
|
53 (60.2%)
|
44 (69.8%)
|
|
|
|
+1SD~+2SD
|
22 (14.6%)
|
13 (14.8%)
|
9 (14.3%)
|
|
|
|
+2SD above
|
6 (4.0%)
|
4 (4.5%)
|
2 (3.2%)
|
|
|
|
Gestational age
|
|
|
|
0.07†
|
0.794
|
0.09
|
Premature
|
11 (7.3%)
|
6 (6.8%)
|
5 (7.9%)
|
|
|
|
Term
|
140 (92.7%)
|
82 (93.2%)
|
58 (92.1%)
|
|
|
|
Birth weight (kg)
|
3.20(0.54)
|
3.21(0.52)
|
3.19(0.58)
|
0.09*
|
0.932
|
0.04
|
Delivery method
|
|
|
|
0.03
|
0.858
|
0.03
|
Vaginal
|
85 (56.3%)
|
49 (55.7%)
|
36 (57.1%)
|
|
|
|
Caesarean
|
66 (43.7%)
|
39 (44.3%)
|
27 (42.9%)
|
|
|
|
Paternal highest education
|
|
|
|
0.03
|
0.980
|
0.09
|
Postgraduate
|
20 (13.2%)
|
14 (15.9%)
|
6 (9.5%)
|
|
|
|
Bachelor
|
80 (53.0%)
|
44 (50.0%)
|
36 (57.1%)
|
|
|
|
College
|
27 (17.9%)
|
12 (13.6%)
|
15 (23.8%)
|
|
|
|
High school
|
18 (11.9%)
|
13 (14.8%)
|
5 (7.9%)
|
|
|
|
Junior school/under
|
6 (4.0%)
|
5 (5.7%)
|
1 (1.6%)
|
|
|
|
Maternal highest education
|
|
|
|
1.66
|
0.097
|
0.29
|
Postgraduate
|
13 (8.6%)
|
7 (8.0%)
|
6 (9.5%)
|
|
|
|
Bachelor
|
93 (61.6%)
|
50 (56.8%)
|
43 (68.3%)
|
|
|
|
College
|
39 (25.8%)
|
26 (29.5%)
|
13 (20.6%)
|
|
|
|
High school
|
5 (3.3%)
|
4 (4.5%)
|
1 (1.6%)
|
|
|
|
Junior school/under
|
1 (0.7%)
|
1 (1.1%)
|
0 (0.0%)
|
|
|
|
Family income (RMB/year)
|
|
|
|
0.32
|
0.753
|
0.08
|
More than 500k
|
45 (30.0%)
|
27 (30.7%)
|
18 (28.6%)
|
|
|
|
100-500k
|
95 (63.3%)
|
53 (60.2%)
|
42 (66.7%)
|
|
|
|
Less than 100k
|
10 (6.7%)
|
8 (9.1%)
|
2 (3.2%)
|
|
|
|
† Chi-square test; * t-test; the others were Mann-Whitney U test. |
SD: standard deviation; SES: socioeconomic status; ASD: Autism Spectrum Disorder |
Comparison in adaptive skills, IQ, early signs of developmental abnormalities and symptom severity between ASD with IQ ≥ 70 and with IQ < 70
As shown in Table 2, significant differences were presented in overall adaptive skills and the three domain skills between ASD with IQ ≥ 70 and with IQ < 70 (all P > 0.05). In cognition assessment, ASD children with IQ ≥ 70 exhibited significant superiority in both VAI and NVI compared with ASD children with IQ < 70 (t = 14.32, 15.40, P < 0.001, Cohen's d = 2.56, 2.83). In early signs of developmental abnormalities, age of first single words and first phrases for children with IQ ≥ 70 were significant earlier than children with IQ < 70 (t = 4.15, 4.24; P < 0.001; Cohen's d = 0.61, 0.74), whereas no significant differences in age of first walking unaided were found between the two groups based on ADI-R (t = 0.26, P = 0.799). In symptom assessment, children with IQ ≥ 70 displayed slighter ASD severity compared with IQ < 70 (t = 3.49, P = 0.001, Cohen's d = 0.55)
Table 2
Comparison in adaptive functioning, IQ, early signs of developmental abnormalities and symptom severity between ASD with IQ ≥ 70 and with IQ < 70
|
Total sample
|
With IQ < 70
|
With IQ ≥ 70
|
With IQ < 70 vs With IQ ≥ 70
|
|
N = 151
|
n = 88
|
n = 63
|
t
|
P
|
Effect Size
|
ABAS-II
|
|
|
|
|
|
|
Overall adaptive skills
|
69.52(15.68)
|
61.61(12.51)
|
80.56 (12.77)
|
8.80
|
0.000
|
1.50
|
Conceptual skills
|
68.4(15.80)
|
60.4(12.14)
|
79.57(13.39)
|
9.01
|
0.000
|
1.51
|
Social skills
|
67.02(16.68)
|
59.38(13.83)
|
77.7(14.35)
|
7.57
|
0.000
|
1.30
|
Practical skills
|
72.43(16.93)
|
64.31(13.87)
|
83.78(14.10)
|
8.00
|
0.000
|
1.39
|
WISC-IV
|
|
|
|
|
|
|
FSIQ
|
69.49(19.57)
|
55.91(6.45)
|
88.46(15.52)
|
16.07
|
0.000
|
2.92
|
VAI
|
71.77(21.43)
|
57.57(12.18)
|
91.62(14.71)
|
14.32
|
0.000
|
2.56
|
NVI
|
72.61(19.70)
|
59.07(6.84)
|
91.52(15.81)
|
15.40
|
0.000
|
2.83
|
ADI-R
|
|
|
|
|
|
|
Age of first walking unaided (months)
|
14.11(2.86)
|
14.23(3.12)
|
13.94(2.46)
|
0.26
|
0.799
|
0.10
|
Age of first single words (months)
|
24.95(11.06)
|
27.64(11.56)
|
21.19(9.14)
|
4.15
|
0.000
|
0.61
|
Age of first phrases (months)
|
33.48(10.26)
|
36.47(10.68)
|
29.32(8.03)
|
4.24
|
0.000
|
0.74
|
ADOS-Css
|
7.57(1.53)
|
7.91(1.57)
|
7.10(1.33)
|
3.49
|
0.001
|
0.55
|
ABAS-II: Adaptive Behavior Assessment System Version II; WISC-IV: Wechsler Intelligence Scale (the fourth edition); FSIQ: full scale intelligence quotient; VAI: Verbal Acceptation Index; NVI: Nonverbal Index; Autism Diagnostic Interview-Revised (ADI-R); ADOS-Css: Autism Diagnostic Observation Schedule-Calibrated severity score. |
Comparison in IQ-Adaptive skills gap between ASD with IQ≥70 and with IQ<70
We next examined the gap between IQ and ABAS scores, calculated for the composite score and each domain. The interaction of group × age and group× age of diagnosis was firstly examined. There were no interaction of group × age and group × age of diagnosis in model comparisons, except in IQ-social skills gap (seen Table3). In the first level, ASD children with IQ ≥ 70 had significantly higher FSIQ-Overall adaptive skills gap proportion scores compared with IQ < 70 group (F = 35.36, P < 0.001, η²=0.19). Meanwhile, there were still significant discrepancy in VAI-Overall adaptive skills and NVI-Overall adaptive skills gap proportion scores between the two groups (F = 32.51, 22.61; P < 0.001; η²=0.18, 0.13). In the second level, ASD children with IQ ≥ 70 had significantly higher IQ-Adaptive functioning gap proportion scores in all the three domains (all P < 0.001). Significant discrepancies were also exhibited in the gap of VAI /NVI and specific domain between the two groups (all P < 0.001).
Table 3
Comparison in IQ-Adaptive functioning gap proportion scores between ASD with IQ ≥ 70 and with IQ < 70
IQ-adaptive functioning gap proportion scores
|
With IQ < 70
|
With IQ ≥ 70
|
Group*Age
|
Group*Age of diagnosis
|
With IQ < 70 vs With IQ ≥ 70
|
n = 88
|
n = 63
|
F
|
P
|
F
|
P
|
F
|
P
|
Effect Size
|
FSIQ-Overall adaptive skills
|
-10.64(21.18)
|
7.42(16.83)
|
0.83
|
0.44
|
1.11
|
0.332
|
35.36
|
0.000
|
0.19
|
FSIQ-Conceptual skills
|
-8.57(21.08)
|
8.41(17.86)
|
0.85
|
0.429
|
0.82
|
0.441
|
29.17
|
0.000
|
0.17
|
FSIQ-Social skills
|
-6.76(24.18)
|
10.84(17.65)
|
3.70
|
0.027
|
4.54
|
0.012
|
29.65
|
0.000
|
0.17
|
FSIQ- Practical skills
|
-15.51(23.88)
|
3.53(19.19)
|
1.76
|
0.176
|
0.49
|
0.615
|
30.27
|
0.000
|
0.17
|
Verbal Index-Overall adaptive skills
|
-7.67(23.00)
|
12.03(17.51)
|
2.37
|
0.097
|
0.59
|
0.553
|
32.51
|
0.000
|
0.18
|
VAI-Conceptual skills
|
-5.60(22.26)
|
13.02(16.81)
|
2.78
|
0.065
|
0.69
|
0.501
|
33.42
|
0.000
|
0.19
|
VAI -Social skills
|
-3.79(27.48)
|
15.45(20.95)
|
6.37
|
0.002
|
3.26
|
0.041
|
27.96
|
0.000
|
0.16
|
VAI - Practical skills
|
-12.54(24.43)
|
8.14(19.26)
|
3.04
|
0.051
|
0.67
|
0.514
|
30.52
|
0.000
|
0.17
|
NVI-Overall adaptive skills
|
-4.74(22.50)
|
11.14(19.82)
|
0.40
|
0.669
|
0.97
|
0.383
|
22.61
|
0.000
|
0.13
|
NVI-Conceptual skills
|
-2.67(22.46)
|
12.13(21.12)
|
0.46
|
0.629
|
0.71
|
0.495
|
18.50
|
0.000
|
0.11
|
NVI-Social skills
|
-0.86(25.82)
|
14.56(19.45)
|
2.59
|
0.079
|
4.36
|
0.015
|
20.44
|
0.000
|
0.12
|
NVI- Practical skills
|
-9.61(24.91)
|
7.25(21.65)
|
1.49
|
0.229
|
0.40
|
0.674
|
20.11
|
0.000
|
0.12
|
ASD: Autism Spectrum Disorder; FSIQ: full scale intelligence quotient; VIA: Verbal Acceptation Index; NVI: Nonverbal Index. |
Contribution of socioeconomic factors, IQ, ASD severity and early signs of developmental abnormalities to ABAS overall adaptive skills
The hierarchical regression model was performed to examine how socioeconomic factors, IQ, ASD severity, and early signs of developmental abnormalities impact on overall adaptive skills in sequence (seen Table4). The total model had good interpretation with 61.8% of variance (P < 0.001). Though gender was relevant to overall adaptive skills in total model (β = 0.12, P = 0.038), age and gender accounted for no significant variance in step1 (F = 1.64, P = 0.198). Socioeconomic factors contributed an additional 9.7% of variance in step2 (ΔF = 5.31, P = 0.002), and maternal highest education was positively correlated with overall adaptive skills in total model (β = 0.15, P = 0.022). FSIQ contributed an additional 36.9% variance in step3 (ΔF = 103.14, P < 0.001), manifested that IQ had the largest contribution value to overall adaptive skills. VAI and NVI contributed an additional 6.4% of the variance in step4 (ΔF = 10.12, P < 0.001), and VAI emerged as a specific predictor in the total model (β = 0.34, P = 0.003), rather than NVI (β=-0.37, P = 0.087). ADOS-Css contributed no significant variance in step5 (ΔF = 0.00, P = 0.964), and was found no significant correlation with overall adaptive skills in the total model (β=-0.07, P = 0.233). Early signs of developmental abnormalities accounted for an additional 6.6% of the variance in step6 (ΔF = 7.84, P < 0.001), and age of first walking unaided was negatively relevant to overall adaptive skills in the total model (β=-0.21, P < 0.001).
Table 4
Contribution of SES, IQ, ASD severity and early signs of developmental abnormalities to overall adaptive skills
|
R2
|
ΔR2
|
B
|
SE
|
β
|
t
|
P
|
Step1
|
0.02
|
|
|
|
|
|
0.198
|
Age
|
|
|
1.45
|
1.44
|
0.08
|
1.01
|
0.313
|
Gender
|
|
|
5.77
|
3.53
|
0.14
|
1.64
|
0.104
|
Step2
|
0.12
|
0.10
|
|
|
|
|
0.002
|
Paternal highest education
|
|
|
-0.92
|
1.56
|
-0.06
|
-0.59
|
0.554
|
Maternal highest education
|
|
|
7.27
|
2.06
|
0.32
|
3.53
|
0.001
|
Family income
|
|
|
1.53
|
2.57
|
0.05
|
0.60
|
0.552
|
Step3
|
0.49
|
0.37
|
|
|
|
|
0.000
|
FSIQ
|
|
|
0.50
|
0.05
|
0.62
|
10.16
|
0.000
|
Step4
|
0.55
|
0.06
|
|
|
|
|
0.000
|
VAI
|
|
|
0.32
|
0.08
|
0.44
|
3.90
|
0.000
|
NVI
|
|
|
-0.27
|
0.18
|
-0.34
|
-1.53
|
0.128
|
Step5
|
0.55
|
0.00
|
|
|
|
|
0.964
|
ADOS-Css
|
|
|
0.03
|
0.63
|
0.00
|
0.05
|
0.964
|
Step6-Total model
|
0.62
|
0.07
|
|
|
|
|
0.000
|
Age
|
|
|
1.58
|
0.98
|
0.09
|
1.60
|
0.111
|
Gender
|
|
|
4.93
|
2.35
|
0.12
|
2.10
|
0.038
|
Paternal highest education
|
|
|
-0.64
|
1.08
|
-0.04
|
-0.59
|
0.555
|
Maternal highest education
|
|
|
3.40
|
1.46
|
0.15
|
2.32
|
0.022
|
Family income
|
|
|
2.12
|
1.78
|
0.08
|
1.19
|
0.237
|
FSIQ
|
|
|
0.47
|
0.19
|
0.59
|
2.45
|
0.016
|
VAI
|
|
|
0.25
|
0.08
|
0.34
|
3.01
|
0.003
|
Nonverbal-Index
|
|
|
-0.29
|
0.17
|
-0.37
|
-1.72
|
0.087
|
ADOS-Css
|
|
|
-0.76
|
0.63
|
-0.07
|
-1.20
|
0.233
|
Age of first walking unaided
|
|
|
-1.15
|
0.32
|
-0.21
|
-3.59
|
0.000
|
Age of first single words
|
|
|
0.03
|
0.13
|
0.02
|
0.19
|
0.850
|
Age of first phrases
|
|
|
-0.27
|
0.14
|
-0.17
|
-1.87
|
0.063
|
Durbin-Watson = 1.82 |
ASD: Autism Spectrum Disorder; FSIQ: full scale intelligence quotient; VIA: Verbal Acceptation Index; NVI: Nonverbal Index; ADOS-Css: Autism Diagnostic Observation Schedule-Calibrated severity score. |
Contribution of socioeconomic factors, IQ, ASD severity and early signs of developmental abnormalities to ABAS conceptual skills
Table 5 displays the model examining how socioeconomic factors, IQ, ASD severity and early signs of developmental abnormalities impact on ABAS conceptual skills. The total model still had good interpretation with 59.5% of variance (P = 0.002). Age and gender accounted for no significant variance in step1 (F = 1.40, P = 0.251), and they did not predict conceptual skills in total model (all P > 0.05). socioeconomic factors contributed an additional 7.8% of variance in step2 (ΔF = 4.16, P = 0.007), while the three factors were not significantly relevant to conceptual skills in total model (all P > 0.05), showed that socioeconomic factors were poor associated with conceptual skills. FSIQ contributed an additional 34.6% variance in step3 (ΔF = 88.88, P < 0.001), manifested that IQ still had the largest contribution value to conceptual skills. VAI and NVI interpreted an additional 10.0% of the variance in step4 (ΔF = 15.40, P < 0.001). Rather than NVI (β=-0.36, P = 0.099), VAI still emerged as a specific predictor in the total model (β = 0.44, P < 0.001). ADOS-Css contributed no significant variance in step5 (ΔF = 1.33, P = 0.252), and did not to predict overall adaptive skills in total model (β = 0.00, P = 0.995). Early signs of developmental abnormalities accounted for an additional 4.8% of the variance in step6 (ΔF = 5.39, P = 0.002), and age of first walking unaided was negatively associated with conceptual skills in the total model (β= -0.15, P = 0.014).
Table 5
Contribution of SES, IQ, ASD severity and early signs of developmental abnormalities to ABAS-II conceptual skills
|
R2
|
ΔR2
|
B
|
SE
|
β
|
t
|
P
|
Step1
|
0.02
|
|
|
|
|
|
0.251
|
Age
|
|
|
1.35
|
1.44
|
0.08
|
0.94
|
0.351
|
Gender
|
|
|
5.35
|
3.55
|
0.13
|
1.51
|
0.134
|
Step2
|
0.10
|
0.08
|
|
|
|
|
0.007
|
Paternal highest education
|
|
|
-0.38
|
1.58
|
-0.02
|
-0.24
|
0.810
|
Maternal highest education
|
|
|
6.53
|
2.10
|
0.29
|
3.12
|
0.002
|
Family income
|
|
|
0.56
|
2.61
|
0.02
|
0.21
|
0.831
|
Step3
|
0.44
|
0.35
|
|
|
|
|
0.000
|
FSIQ
|
|
|
0.49
|
0.05
|
0.60
|
9.43
|
0.000
|
Step4
|
0.54
|
0.10
|
|
|
|
|
0.000
|
VAI
|
|
|
0.41
|
0.08
|
0.56
|
4.88
|
0.000
|
NVI
|
|
|
-0.31
|
0.18
|
-0.39
|
-1.76
|
0.081
|
Step5
|
0.55
|
0.00
|
|
|
|
|
0.252
|
ADOS-Css
|
|
|
0.74
|
0.64
|
0.07
|
1.15
|
0.252
|
Step6-Total model
|
0.60
|
0.05
|
|
|
|
|
0.002
|
Age
|
|
|
1.19
|
1.02
|
0.07
|
1.17
|
0.242
|
Gender
|
|
|
4.64
|
2.43
|
0.11
|
1.91
|
0.058
|
Paternal highest education
|
|
|
-0.44
|
1.12
|
-0.03
|
-0.40
|
0.693
|
Maternal highest education
|
|
|
2.67
|
1.51
|
0.12
|
1.76
|
0.080
|
Family income
|
|
|
1.58
|
1.84
|
0.06
|
0.86
|
0.391
|
FSIQ
|
|
|
0.40
|
0.20
|
0.49
|
1.99
|
0.048
|
VAI
|
|
|
0.33
|
0.09
|
0.44
|
3.85
|
0.000
|
NVI
|
|
|
-0.29
|
0.18
|
-0.36
|
-1.66
|
0.099
|
ADOS-Css
|
|
|
0.00
|
0.65
|
0.00
|
0.01
|
0.995
|
Age of first walking unaided
|
|
|
-0.82
|
0.33
|
-0.15
|
-2.49
|
0.014
|
Age of first single words
|
|
|
-0.04
|
0.14
|
-0.03
|
-0.27
|
0.789
|
Age of first phrases
|
|
|
-0.23
|
0.15
|
-0.15
|
-1.58
|
0.115
|
Durbin-Watson = 1.83 |
ASD: Autism Spectrum Disorder; FSIQ: full scale intelligence quotient; VIA: Verbal Acceptation Index; NVI: Nonverbal Index; ADOS-Css: Autism Diagnostic Observation Schedule-Calibrated severity score. |
Contribution of socioeconomic factors, IQ, ASD severity and early signs of developmental abnormalities to ABAS social skills
The hierarchical regression model was also conducted to examine contribution of socioeconomic factors, IQ, ASD severity, and early signs of developmental abnormalities to social skills (seen Table6). The total model had good interpretation with 49.2% of variance (P = 0.005). Age and gender no significant variance in step1 (F = 1.69, P = 0.188). Though socioeconomic factors contributed an additional 9.5% of variance in step2 (ΔF = 5.16, P = 0.002), they have no significant association with social skills in total model (all P > 0.05). FSIQ contributed an additional 30.2% variance in step3 (ΔF = 74.38, P < 0.001), showed that IQ had the largest contribution value to social skills. Despite ADOS-Css were negatively associated with social skills in the total model (β=-0.15, P = 0.033), the scores accounted for no significant variance in step5 (ΔF = 1.88, P = 0.173). Early signs of developmental abnormalities accounted for an additional 5.0% of the variance in step6 (ΔF = 4.52, P = 0.005), and age of first phrases was negatively correlated with social skills (β=-0.21, P = 0.048), as well as age of first walking unaided (β= -0.16, P = 0.015).
Table 6
Contribution of SES, IQ, ASD severity and early signs of developmental abnormalities to ABAS-II social skills
|
R2
|
ΔR2
|
B
|
SE
|
β
|
t
|
P
|
Step1
|
0.02
|
|
|
|
|
|
0.188
|
Age
|
|
|
-0.69
|
1.52
|
-0.04
|
-0.45
|
0.651
|
Gender
|
|
|
6.33
|
3.74
|
0.14
|
1.69
|
0.093
|
Step2
|
0.12
|
0.10
|
|
|
|
|
0.002
|
Paternal highest education
|
|
|
0.44
|
1.65
|
0.03
|
0.27
|
0.789
|
Maternal highest education
|
|
|
7.01
|
2.19
|
0.29
|
3.21
|
0.002
|
Family income
|
|
|
0.46
|
2.72
|
0.02
|
0.17
|
0.865
|
Step3
|
0.42
|
0.30
|
|
|
|
|
0.000
|
FSIQ
|
|
|
0.48
|
0.06
|
0.56
|
8.63
|
0.000
|
Step4
|
0.43
|
0.02
|
|
|
|
|
0.166
|
VAI
|
|
|
0.17
|
0.10
|
0.22
|
1.73
|
0.086
|
NVI
|
|
|
-0.11
|
0.21
|
-0.12
|
-0.50
|
0.619
|
Step5
|
0.44
|
0.01
|
|
|
|
|
0.173
|
ADOS-Css
|
|
|
-1.03
|
0.75
|
-0.09
|
-1.37
|
0.173
|
Step6-Total model
|
0.49
|
0.05
|
|
|
|
|
0.005
|
Age
|
|
|
0.03
|
1.20
|
0.00
|
0.03
|
0.978
|
Gender
|
|
|
5.24
|
2.88
|
0.12
|
1.82
|
0.071
|
Paternal highest education
|
|
|
0.96
|
1.33
|
0.06
|
0.73
|
0.468
|
Maternal highest education
|
|
|
3.50
|
1.79
|
0.15
|
1.96
|
0.052
|
Family income
|
|
|
0.49
|
2.18
|
0.02
|
0.23
|
0.821
|
FSIQ
|
|
|
0.46
|
0.24
|
0.55
|
1.96
|
0.052
|
VAI
|
|
|
0.11
|
0.10
|
0.14
|
1.05
|
0.296
|
NVI
|
|
|
-0.18
|
0.21
|
-0.22
|
-0.89
|
0.376
|
ADOS-CSS
|
|
|
-1.67
|
0.77
|
-0.15
|
-2.16
|
0.033
|
Age of first walking unaided
|
|
|
-0.96
|
0.39
|
-0.16
|
-2.46
|
0.015
|
Age of first single words
|
|
|
0.10
|
0.16
|
0.07
|
0.62
|
0.538
|
Age of first phrases
|
|
|
-0.35
|
0.17
|
-0.21
|
-1.99
|
0.048
|
Durbin-Watson = 1.86 |
ASD: Autism Spectrum Disorder; FSIQ: full scale intelligence quotient; VIA: Verbal Acceptation Index; NVI: Nonverbal Index; ADOS-Css: Autism Diagnostic Observation Schedule-Calibrated severity score. |
Contribution of socioeconomic factors, IQ, ASD severity and early signs of developmental abnormalities to ABAS practical skills
Table 7 showed the hierarchical regression model examining how socioeconomic factors, IQ, ASD severity and early signs of developmental abnormalities impact on ABAS practical skills. The total model also had good interpretation with 56.0% of variance (P < 0.001). Age and gender still accounted for no significant variance in step1 (F = 1.46, P = 0.237). Socioeconomic factors contributed an additional 8.1% of variance in step2 (ΔF = 4.32, P = 0.006), and maternal highest education was positively associated with practical skills in total model (β = 0.14, P = 0.043). FSIQ contributed an additional 31.2% variance in step3 (ΔF = 75.94, P < 0.001), demonstrated that IQ still had the largest contribution value to practical skills consisting with conceptual skills. VAI and NVI contributed an additional 7.2% of variance in step4 (ΔF = 9.85, P < 0.001). VAI also emerged as a specific predictor to practical skills similar with conceptual skills in the total model (β = 0.40, P = 0.001), rather than NVI (β=-0.32, P = 0.159). Though ADOS-Css were negatively correlated with practical skills in the total model (β=-0.14, P = 0.041), the scores accounted for no significant variance in step5 (ΔF = 0.63, P = 0.428). In the final step, age of first walking unaided still exhibited as a unique predictor to practical skills (β=-0.23, P < 0.001).
Table 7
Contribution of SES, IQ, ASD severity and early signs of developmental abnormalities to ABAS-II practical skills
|
R2
|
ΔR2
|
B
|
SE
|
β
|
t
|
P
|
Step1
|
0.02
|
|
|
|
|
|
0.237
|
Age
|
|
|
2.14
|
1.55
|
0.11
|
1.38
|
0.170
|
Gender
|
|
|
4.57
|
3.81
|
0.10
|
1.20
|
0.232
|
Step2
|
0.10
|
0.08
|
|
|
|
|
0.006
|
Paternal highest education
|
|
|
-1.65
|
1.70
|
-0.10
|
-0.97
|
0.334
|
Maternal highest education
|
|
|
7.25
|
2.25
|
0.29
|
3.22
|
0.002
|
Family income
|
|
|
2.30
|
2.80
|
0.08
|
0.82
|
0.412
|
Step3
|
0.41
|
0.31
|
|
|
|
|
0.000
|
FSIQ
|
|
|
0.50
|
0.06
|
0.57
|
8.72
|
0.000
|
Step4
|
0.48
|
0.07
|
|
|
|
|
0.000
|
VAI
|
|
|
0.39
|
0.10
|
0.50
|
4.08
|
0.000
|
NVI
|
|
|
-0.21
|
0.20
|
-0.24
|
-1.03
|
0.306
|
Step5
|
0.49
|
0.00
|
|
|
|
|
0.428
|
ADOS-Css
|
|
|
-0.58
|
0.73
|
-0.05
|
-0.80
|
0.428
|
Step6-Total model
|
0.56
|
0.07
|
|
|
|
|
0.000
|
Age
|
|
|
2.18
|
1.14
|
0.12
|
1.91
|
0.058
|
Gender
|
|
|
4.04
|
2.72
|
0.09
|
1.48
|
0.141
|
Paternal highest education
|
|
|
-1.14
|
1.26
|
-0.07
|
-0.91
|
0.364
|
Maternal highest education
|
|
|
3.46
|
1.70
|
0.14
|
2.04
|
0.043
|
Family income
|
|
|
2.45
|
2.06
|
0.08
|
1.19
|
0.238
|
FSIQ
|
|
|
0.38
|
0.22
|
0.44
|
1.68
|
0.095
|
VAI
|
|
|
0.32
|
0.10
|
0.40
|
3.31
|
0.001
|
NVI
|
|
|
-0.28
|
0.20
|
-0.32
|
-1.42
|
0.159
|
ADOS-Css
|
|
|
-1.51
|
0.73
|
-0.14
|
-2.06
|
0.041
|
Age of first walking unaided
|
|
|
-1.40
|
0.37
|
-0.23
|
-3.78
|
0.000
|
Age of first single words
|
|
|
-0.01
|
0.15
|
-0.01
|
-0.05
|
0.959
|
Age of first phrases
|
|
|
-0.23
|
0.17
|
-0.14
|
-1.38
|
0.172
|
Durbin-Watson = 1.90 |
ASD: Autism Spectrum Disorder; FSIQ: full scale intelligence quotient; VIA: Verbal Acceptation Index; NVI: Nonverbal Index; ADOS-Css: Autism Diagnostic Observation Schedule-Calibrated severity score. |