Article in Group & Organization Management · January 014 citations reads 13,031 authors: Some of the authors of this publication are also working on these related projects

Table 1. Comparison of Measurement Models for Study Variables.
Model
Description
χ
2
df
Δχ
2
CFI GFI SRMR RMSEA
The baseline 
five-factor 
model
Subordinates’ survey: Self-leadership, 
conscientiousness, job autonomy, 
job satisfaction
Supervisors’ survey: Subordinates’ 
performance rating
840
289
.92 .88
.07
.07
Model 1:
Four 
factors
Self-leadership and conscientiousness 
were combined as one factor
1,241
293
401**
.84 .78
.08
.09
Model 2:
Four 
factors
Self-leadership and job autonomy were 
combined into one factor
1,811
293
971**
.74 .73
.10
.11
Model 3:
Four 
factors
Performance ratings and job 
satisfaction were combined into 
one factor
2,052
293
1,212**
.70 .68
.15
.12
Model 4:
Two factors
Self-leadership, conscientiousness, 
and job autonomy combined as one 
factor and performance ratings and 
job satisfaction as another
3,354
298
2,514**
.47 .56
.17
.16
Model 5:
One factor
All parcels and items were loaded on a 
single factor
4,075
299
3,235**
.34 .47
.16
.17
Note. CFI = comparative fix index; GFI = goodness-of-fit index; SRMR = standardized root mean squared 
residual; RMSEA = root mean square error of approximation.
**p < .01, two-tailed.
 by guest on July 31, 2014
gom.sagepub.com
Downloaded from 

---

402
Table 2.
Means, Standard Deviations, Correlations, and Reliabilities.
Variable
M
SD
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1.
Gender
a
0.39
0.49
—
2.
Age
b
2.41
1.03
−.12*
—
3.
Tenure
4.70
6.09
−.21**
.59**
—
4.
Education
c
3.51
1.66
−.01
−.17**
−.20**
—
5.
Engineering industry
d
0.20
0.40
−.19**
−.02
.16**
.41**
—
6.
Manufacturing 
industry
d
0.21
0.41
.02
−.15**
−.02
−.41**
−.26**
—
7.
Insurance industry
d
0.44
0.50
.22**
−.06
−.30**
.13**
−.44**
−.46**
—
8.
Transportation 
industry
d
0.15
0.36
−.11*
.27**
.26**
−.18**
−.21**
−.22**
−.37**
—
9.
Conscientiousness
4.15
0.66
−.07
.15**
.09
−.12*
−.14**
.42**
−.32**
.12*
(.81)
10.
Self-leadership
3.47
0.47
.04
−.05
−.04
.15**
−.09
.03
.07
−.04
.29**
(.90)
11.
Job autonomy
4.40
0.90
.05
−.05
−.11*
.12*
−.20**
−.14**
.47**
−.26**
.05
.30**
(.94)
12.
Supervisor 
performance rating
4.49
1.07
.05
−.02
.06
.17**
.25**
.11*
−.19**
−.13**
.21**
.21**
.08
(.92)
13.
Objective 
performance 
measure
e
54,481
91,690
−.04
.04
.14
−.03
—
—
—
—
.06
.20*
.28**
.26**
—
14.
Job satisfaction
4.38
0.91
−.05
.11*
.07
−.07
−.13**
−.05
.13*
.02
.23**
.29**
.51**
.16**
.29**
(.90)
Note. N
= 407. Reliability estimates in parentheses.
a
Gender was coded “0” for male and “1” for female.
b
Age was coded 1 = “18-25” to 5 = “56 or above.”
c
Education was coded as follows: “junior high school or below” = 1; “senior high school” =
2; “vocational or technical college” = 3; “associate degree” = 4; 
“undergraduate degree” = 5; “graduate degree or above” = 6.
d
Engineering, Manufacturing, Insurance, and Transportation industry are dummy variables. F
or each dummy variable, the name indicates the type of industry coded 1 
(e.g., for the Insurance industry, the Insurance industry condition was 1 and the other three
industries were 0).
e
Data collection for the objective performance measure was only available for the subsamp
le of this study which consisted of 153 insurance sample agent in Hong Kong. 
Thus, the sample size for all correlations between objective performance measure and oth
er variables is 153 rather than 407 (whole sample size).
*p
< .05. **
p < .1.
 by guest on July 31, 2014
gom.sagepub.com
Downloaded from 

---

Ho and Nesbit 
403
Tests of Hypotheses
We used hierarchical regression analysis (HRA) to test whether self-leadership 
is positively related to supervisor performance rating (H1a), objective work 
performance (H1b), and job satisfaction (H1c) as well as whether job 
autonomy would moderate these relationships (H2a, H2b, H2c). Following 
the method outlined by Aiken and West (1991), we first centered the self-
leadership scores and job autonomy scores around their respective mean. 
Next, we calculated interaction scores by multiplying the mean centered 
self-leadership and job autonomy scores. As shown in Table 3, separate 
analyses were conducted on each of the three dependent variables and each 
HRA consisted of three steps. The control variables were entered in the first 
step. The centered independent variables of self-leadership and job auton-
omy were entered in the second step. The centered interaction term was 
entered in the third step. Table 3 (step 3) shows that self-leadership was posi-
tively and significantly related to supervisor performance ratings (β = .13,