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 3. Summary of Hierarchical Multiple Regression Analyses.
Supervisor performance 
rating (n = 407)
Objective performance 
measure (n = 153)
a
Job satisfaction
(n = 407)
Steps and variables
1
2
3
1
2
3
1
2
3
1. Control variables
Gender
.12*
.12*
.12*
−.07
−.02
−.03
−.06
−.04
−.04
Age
−.02
.00
.01
−.05
.00
.00
−.02
.03
.04
Tenure
.09
.06
.06
.17
.11
.09
.09
.02
.01
Education
.15**
.12*
.12*
−.01
−.04
−.02
−.08
−.15** −.15**
Manufacturing industry
b
−.13
−.14
−.14
—
—
—
−.11
−.13
−.13
Transportation 
industry
b
−.28**
−.28**
−.28**
—
—
—
.01
.04
.04
Insurance industry
b
−.30**
−.39**
−.40**
—
—
—
.23** −.07
−.08
Conscientiousness
.22**
.15*
.14*
.07
−.04
−.03
.33** .18**
.17**
2. Main effects
Self-leadership
.13*
.13*
.12
.04
.09*
.09*
Job autonomy
.10
.11
.23*
.26**
.52**
.52**
3. Interaction
Self-leadership × Job 
autonomy
.08*
.17*
.08*
R
2
.16**
.19**
.21**
.03
.10*
.13*
.12** .34**
.36**
Note: The coefficients are standardized β weights.
a
Data collection for the objective performance measure was only available for the subsample of this study 
which consisted of 153 insurance sample agent in Hong Kong. Thus, the sample size is 153 rather than 407 
(whole sample size).
b
Manufacturing, Insurance, and Transportation industry are dummy variables with Engineering the omitted 
category which is the comparison category.
*p <.05. **p < .01.
 by guest on July 31, 2014
gom.sagepub.com
Downloaded from 

---

404 
Group & Organization Management 39(4)
p < .05) and job satisfaction (β = .09, p < .05), but it was not significantly 
related to objective job performance (β = .12, ns). Thus, H1a and H1c were 
supported, but H1b was not supported. Step 3 in Table 3 displays the results 
for the Self-leadership × Job autonomy interactions and shows that the 
interaction was significantly related to supervisor performance ratings (β = 
.08, p < .05), objective work performance (β = .17, p < .05), and job satis-
faction (β = .08, p < .05).
To determine whether the forms of the interactions matched those sug-
gested by H2a, H2b, and H2c, we needed to graphically illustrate the inter-
actions and test the simple slopes for respondents with high job autonomy 
(1 SD above the mean) and respondents with low job autonomy (1 SD 
below the mean; Aiken & West, 1991). In support of H2a, we found that 
self-leadership was more positively related to supervisor performance rat-
ings under high job autonomy (simple slope test: β = .47, ΔR
2
= .02, p < .01) 
than under low job autonomy (simple slope test: β = .13, ns). Furthermore, 
in support of H2b, we found that self-leadership was more positively related 
to objective work performance under high job autonomy (simple slope test: 
β = .18, ΔR
2
= .03, p < .05) than under low job autonomy (simple slope test: 
β = .03, ns). In the same vein, and in line with H2c, simple slope analysis 
indicated that self-leadership related more strongly to job satisfaction with 
individuals with high job autonomy (β = .32, ΔR
2
= .02, p < .01) than for 
individuals with low job autonomy (β = .03, ns). Figures 1, 2, and 3 graphi-
cally depict these findings.