Académique Documents
Professionnel Documents
Culture Documents
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/220216335
CITATIONS
READS
76
1,498
2 authors, including:
Shu-Hsien Liao
Tamkang University
97 PUBLICATIONS 2,346 CITATIONS
SEE PROFILE
Graduate School of Management Sciences, Tamkang University, No. 151, Ying-chuan Rd., Tamsui, Taipei 251, Taiwan, ROC
Management Sciences, Tamkang University, No. 151, Ying-chuan Rd., Tamsui, Taipei 251, Taiwan, ROC
a r t i c l e
i n f o
Keywords:
Knowledge management
Organizational learning
Organizational innovation
Mediator
Structural equation modeling
a b s t r a c t
In knowledge economics, enterprises need to adapt and update its knowledge to keep their capability of
innovation. Therefore, the relationship between knowledge management and organizational innovation
is getting an important issue in research and in practical areas. However, without good capability of organizational learning, one organizational cannot retain some important knowledge management practices
in it. This study selects samples based on Common Wealth Magazines Top 1000 manufacturers and Top
100 nancial rms in 2007 by mails. A questionnaire survey was conducted and 327 valid replies were
received. This research analyzes the relationship among knowledge management, as well as organizational learning and organizational innovation utilizing structural equation modeling. The results show
that organizational learning is the mediating variable between knowledge management and organizational innovation. Just like a system, knowledge management is an important input, and organizational
learning is a key process, then organizational innovation is a critical output.
2009 Elsevier Ltd. All rights reserved.
1. Introduction
Facing this rapid change, enterprises should adapt and update its
knowledge to maintain its competitive advantages (Rademakers,
2005). However, past research showed the issues of knowledge
management (KM) are complicated. Some researches are related
to the competitive advantages, and some are the e-business (Lin &
Lee, 2004); some are related to organizational learning, and some
are organizational innovation (Darroch, 2005; Davenport & Prusak,
1998). We found that organizational learning is mixed with KM
(Victor, Francisco, & Antonio, 2006), and the relationship between
knowledge management and organizational learning is not evident.
Reviewing past literatures, many scholars conducted the research to understand the relation among knowledge management,
organizational learning, and organization innovation separately.
We found few papers discussed the practical results and quantitative numbers (Darroch & MaNaughton, 2002). Based on theory,
knowledge management, organizational learning, and organization
innovation should not discuss separately (Goh, 2005). The immediate concern, in the relentless pursuit of innovation within a knowledge enterprise, appears to be more than just identifying and
resolving issues on KM independently.
ment. The results provide a basis for understanding the competitive predisposition of a rm as it enters a program of KM.
Cui, Grifth, and Cavusgil (2005) also mentioned that KM capabilities consist of three interrelated processes: knowledge acquisition, knowledge conversion, and knowledge application (Gold
et al., 2001). Knowledge is not only an important resource for a
rm, but also it serves as a basic source of competitive advantage
(Gold et al., 2001; Grant, 1996; Jaworski & Kohli, 1993). Therefore,
KM capabilities refer to the knowledge management processes in
an organization that develop and use knowledge within the rm
(Gold et al., 2001).
From Gold et al. (2001) and Cui et al. (2005), we nd the completely knowledge management activities form the perspective of
organizational capabilities. They argue that there are three main
processes: acquisition, conversion, and application. Although there
are still many classications of KM, this study prefer the viewpoints of organizational capabilities, and be in favor of these three
dimensions in our study.
2.2. Organizational innovation
The growth innovation literature provides many alternative
conceptualizations and models for the interpretation of observed
data. An innovation can be a new product or service, a new production process technology, a new structure or administrative system,
or a new plan or program pertaining to organizational members.
Therefore, organizational innovation, or innovativeness, is typically
measured by the rate of the adoption of innovations, although a
few studies have used other measures (Damanpour, 1991).
Past research has argued that different types of innovation are
necessary for understanding and identifying in organizations.
Among numerous typologies of innovation advanced in the relevant
literature, three have gained the most attention. Each centers on a
pair of types of innovation: administrative and technical, product
and process, and radical and incremental. In Wang and Ahmed
(2004), they identied organizational innovation through an extensive literature. A nal 20-item measurement construct is validated
through FAME Database which contains information for companies
in the UK and Ireland. FAME contains information on 3.4 million
companies, 2.6 million of which are in a detailed format. These ve
dimensions are tested from component factors and a three-step approach. They are product innovation, market innovation, process
innovation, behavioral innovation, strategic innovation. Because
this measurement is tested by extensive literature collection, and
precisely statistical testing, this study prefers their work to test
the similar samples in Taiwan to compare the results.
Very little empirical research has specically addressed antecedents and consequences of effective knowledge management
(Darroch & MaNaughton, 2002). The management of knowledge
is frequently identied as an important antecedent of innovation.
Effective KM has been presented in the literature as one method
for improving innovation and performance. While many studies
have reported that KM as antecedents of innovation, none has
explicitly examined the relationship between the two constructs.
In Darroch (2005), we got the result that KM process would positively affect innovation. Therefore, it is fair to conclude that the
relationship between KM and innovation is close related. Thus, this
study propose,
H1: Knowledge management will affect organizational innovation positively.
1097
From literature review, knowledge management will affect organizational learning positively (Garratt, 1990; Su et al., 2004). And
organizational learning will inuence organizational innovation
1098
positively (Calantone et al., 2002; Weerawardena et al., 2006). Darroch (2003) also found that knowledge acquisition had more
indirect than direct inuence on innovation. Therefore, this study
propose,
H4: Organizational learning will be a mediator between knowledge management and organizational innovation.
Therefore, this study utility a perspective of system which takes
knowledge management as an important input, and organizational
learning as a key process, and organizational innovation as a critical output. Fig. 1 shows the perspective of system among these
three constructs.
According to the literatures, this study constructs the research
framework which is shown in Fig. 2.
After reviewing literatures, this study constructs a knowledge
map in Fig. 3. Fig. 3 shows the relationships between any two variables, and the whole picture about our research. It says the relatively positions to each variable. Fig. 2 also shows: (1) Grant
(1996), Gorelick and Tantawy-Monsou (2005), Pilar et al. (2005)
and Ke and Wei (2006) mentioned knowledge management would
Input
Knowledge
Management
Process
Output
Organizational
Learning
Organizational
Innovation
Knowledge Management:
Knowledge Acquisition
Knowledge Conversion
Knowledge Application
H2
H1
H4
Organizational Learning:
Management commitment
System perspective
Openness and experimentation
Knowledge transfer and integration
H3
Organizational Innovation:
Behavior Innovation
Product Innovation
Process Innovation
Market Innovation
Strategic Innovation
Knowledge
Management
1099
Items
Numbers
Percentage (%)
Industry
Manufacturing
Finance
164
163
50.2
49.8
Gender
Male
Female
163
164
49.8
50.2
Education
Senior High
College
University
Graduate School
2
45
203
77
0.6
13.8
62.1
23.5
Position
Production
Management
R&D
Others
9
128
72
118
2.8
39.1
22
36.1
Years in company
1(below)3 years
45 years
610 years
11(above) years
143
57
91
36
43.7
17.4
27.8
11
those different from pilot. The format and content of the questionnaire were initially developed from thorough literature review.
This study adopts three dimensions from Gold et al. (2001).
They are knowledge acquisition, knowledge conversion, and
knowledge application except knowledge protection because it
will decrease knowledge transfer and integration. Organizational
learning is dened as the activities which organizations do in
transformation of learning capability including individuals and
competitors (Pilar et al., 2005). It is considered to be four dimensions: management commitment, system perspective, openness
and experimentation, and knowledge transfer and integration.
Organizational innovation is dened as ve dimensions: they are
behavior, product, process, market, and strategic innovations
(Wang & Ahmed, 2004).
(2) Relationship between knowledge management and organizational learning: Knowledge management is positively
related to organizational learning, meaning that business
with more knowledge management show higher capability
in enhancing organizational learning.
(3) Relationship between organizational learning and organizational innovation: Organizational learning is positively
related to organizational innovation, meaning that business
with more organizational learning show higher capability in
enhancing organizational innovation.
Correlations can only reveal the degree of relationship between
constructs. To further understand the direct and indirect effects, as
well as mediating effects among the constructs, further analysis by
structural equation model is required.
4. Results
4.2. Measurement model
4.1. Correlation analysis
LISRELs 8.7 maximum likelihood program is implemented to
test the theoretical model proposed, as shown in Fig. 4. This structural equation model approach is characterized by its exible
interplay between theory and data, as well as its bridging of theoretical and empirical knowledge for a better understanding of the
real world (Fornell & Larcker (1981)). Such analysis allows for
modeling based on both latent variables and manifest variables,
which is a property well suited for the hypothesized model, where
most of the represented constructs are abstractions of unobservable phenomena. Furthermore, structural equation modeling
2
.761
.730**
.689**
.576**
.553**
.599**
.561**
.484**
.472**
.581**
.374**
.489**
3.73
.454
.700
.760**
.532**
.507**
.569**
.500**
.490**
.431**
.522**
.358**
.447**
3.68
.450
.807
.577**
.575**
.639**
.610**
.554**
.537**
.596**
.440**
.555**
3.78
.452
.676
.675**
.732**
.670**
.698**
.457**
.583**
.436**
.520**
3.43
.565
.702
.605**
.596**
.576**
.465**
.570**
.415**
.486**
3.48
.600
.656
.675**
.706**
.470**
.606**
.403**
.532**
3.56
.542
.536
.568**
.448**
.512**
.347**
.518**
3.53
.527
.734
.544**
.653**
.522**
.590**
3.42
.582
.710
.705**
.765**
.638**
3.34
.586
10
.697
.658**
.598**
3.50
.513
11
12
.721
.599**
3.12
.595
.439
3.37
.492
1100
considers errors in measurement, variables with multiple indicators, and multiple-group comparisons. Table 2 shows the means,
standard deviations and correlation coefcient of each research
variables, to use as analysis of the signicance level of the relationship that exists between the analyzed aspects.
In terms of the quality of measurement model for the full sample, the constructs display satisfactory levels of reliability, as indicated in Table 1 diagonal from .54 to .81 (Nunnally, 1978). Table 3
indicates the tting index of measurement of each construct. Convergent validity can be judged by considering both the signicance
of the factor loading and t-values. All the multi-items constructs t
this criterion, and the loading is signicantly related to its underlying factor (t-values greater than 1.96) in support of convergent
validity (see Table 4). To assess discriminate validity, a series of difference tests on the factor correlations among all the constructs
Table 3
Index of conrmatory factor analysis.
Index
KM
OL
INN
GFI
SRMR
RMSEA
NNFI
CFI
.86
.06
.077
.94
.95
546.3
186
2.94
.92
.049
.063
.97
.97
225.11
98
2.3
.84
.071
.093
.94
.95
611.63
160
3.82
v2
DF
Normed chi-square
Table 4
Reliability and convergent validity.
Variables
Reliability
t-Value
KM
.890
.75
.73
.76
.44
.46
.42
OL
.884
.79
.71
.73
.70
.38
.49
.46
.51
14.69***
14.27***
15.04***
16.09***
14.06***
14.52***
13.60***
.69
.76
.80
.68
.68
.52
.42
.35
.54
.54
13.64***
15.48***
16.79***
13.23***
13.24***
INN
***
.892
(Anderson & Gerbing, 1988). This was done for one pair of variables
at a time by constraining the estimated correlation parameter between them to 1.0 and then performing a difference test on the values obtained for the constrained and unconstrained models
(Anderson & Gerbing, 1988). The resulting signicant difference
in indicates that the two constructs are not perfectly correlated
and that discriminate validity is achieved (Bagozzi & Phillips,
1982). Therefore, after two-stage analysis, we got the results of
convergent validity and discriminate validity. Based on Tables 4
and 5, all t-values show well convergent validity, and the differences of chi-square are greater than 3.84, where this is a good evidence for the dimensions discriminate validity.
4.3. Structural model
Structural equation modeling of the LISREL 8.7 is implemented
to assess the robustness of the results and the stability of the models. For the structural model, Table 6 illustrates the parameter estimates and GFI indicators. The results indicated that this structure
t the data well, v2(51, n = 327) = 148.18, p < :01, CFI = .98,
NNFI = .97, RMSEA = .076.
Table 7 shows the structural model with the standardized coefcients for the research sample. The result reported in Table 6 provided sufcient support for hypothesis 1. Knowledge management
is signicantly and positively related to organizational innovation,
c1 :26, t(51) = 2.67, p < :05. And the numbers in Table 5 provided
support for hypothesis 2 and 3. Knowledge management is significantly and positively related to organizational learning, c2 :78,
t(51) = 11.49, p < :05. Organizational learning is signicantly and
positively related to organizational innovation, b1 :62, t(51) =
5.88, p < :05.
However, the result reported in Table 7 provides path analysis
showing the direct and indirect effect of each constructs. After
analysis, we nd the direct effect of knowledge management
and organizational innovation, c1 :26, t(51) = 2.67, p < :01, is
Table 5
Discriminate validity.
Model
v2
DF
Dv2
KMOL
KMINN
OLINN
148.18
227.74
258
238.27
51
52
52
52
79.56*
109.82*
90.09*
1101
Standardized coefcients
H1
H2
H3
Paths
.26
.78*
.62*
GFI = .93
NNFI = .97
t-Value
Result
2.67
11.94
5.88
CFI = .98
RMSEA = .076
Supported
Supported
Supported
Table 7
Direct and indirect relationship.
Variables
Endogenous
Organizational learning
Exogenous
Knowledge management
Direct
Indirect
Total
Organizational innovation
Effect
t-Value
Effect
t-Value
.78***
11.94
.26**
.48***
.74***
2.67
5.62
10.51
.62***
.61***
5.88
5.83
Endogenous
Organizational learning
Direct
Indirect
Total
*
**
***
Lin and Lee (2004), this paper contribute that knowledge management affect organizational learning directly. Via empirical evidence, knowledge management will affect organizational learning.
Thirdly, the results indicate that there is sufcient evidence to
support a relationship between organizational learning and organizational innovation. This empirical evidence implies that organizational learning has affected organizational innovation in this
study. This study considers that the organizational learning could
lead to this solution. This result concurs with Weerawardena
et al. (2006) and Liao et al. (2008) who shows that the more organizational learning, the more organizational innovation. Accordingly, this study encourages organizational learning including
organizations and members to each other, to increase organizational innovation. Organizations and members should learn by active, novelty ways instead of learning inertia. Once, learning
inertia exist, it will has negative effect on organizational learning
(Liao, Fei, & Liu, 2008).
Fourthly, the main theoretical contribution of this study to the
organizational learning is its mediation of knowledge management
and organizational innovation. Empirical evidence shows that the
relationship between knowledge management and organizational
innovation is signicant with a direct effect smaller than indirect
effect. Accordingly, this paper contribute that organizational learning triggers the relationship between knowledge management and
organizational innovation. In other words, in order to increase
organizational innovation, knowledge must execute via organizational learning. Organizational innovation will show out if business
can implement organizational learning in knowledge-intensive
industry.
Traditionally, many authors mixed the knowledge management
and organizational learning (Victor et al., 2006). Therefore, authors
ignore the importance of organizational learning. This study rst
considers organizational learning as a mediator variable. And after
empirical evidence analysis, knowledge management is seemed as
1102
an input, and organizational learning is a kind of process, organizational innovation will appear as an outcome.
learning. By thoroughly doing organizational learning, KM implementation will lead to organizational innovation.
References
1103