Blackwell Publishing Ltd.Oxford, UK and Malden, USAIJTDInternational Journal of Training and Development1360-3736Blackwell Publishing Ltd.

2004December 200484258273ArticlesLearning organization in

mainland China

International Journal of Training and Development 8:4

ISSN 1360-3736

Learning organization in
mainland China: empirical
research on its application to
Chinese state-owned
enterprises
De Zhang, Zhen Zhang and
Baiyin Yang

This paper examines the applicability of the learning organi-

zation concept and its measurement in a Chinese context.
Based on the theoretical framework proposed by Watkins and
Marsick (1993, 1996, 1997), this paper identifies the differences
in seven of the Dimensions of Learning Organization Ques-
tionnaire (DLOQ) between traditional state-owned enterprises
(SOEs) versus independent listed companies and companies in
service versus manufacturing industries in China. Results indi-
cate that the Chinese version of the DLOQ demonstrated
acceptable psychometric properties. Service companies exhibit
better learning practices than manufacturing companies; how-
ever, the independently listed companies failed to show better
learning practices than their unlisted counterparts. Implica-
tions for practice and future research are discussed.

Introduction
China is the leading third-world country in terms of its population and economic
development. China began its economic reform in the late 1970s. Its GNP has increased

r De Zhang, School of Economics and Management, Tsinghua University, Beijing, China 100084.
Email: [email protected]. Zhen Zhang, Carlson School of Management, University of Min-
nesota, 321-19th Ave. S, Minneapolis, MN 55455, USA. Email: [email protected]. Baiyin Yang,
Human Resource Development and Adult Education, University of Minnesota, 1954 Buford Avenue,
Room 325K, St. Paul, MN55108, USA. Email: [email protected].

© Blackwell Publishing Ltd. 2004, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main St., Malden, MA 02148,
USA.

258 International Journal of Training and Development

by an annual average of 9% since then and per capita income has increased more than
fourfold (Cooper and Zheng, 1998). The ownership restructuring in Chinese state-
owned enterprises (SOEs) is the most significant organizational change brought about
by the reform. This structural reform and the economic growth in China have substan-
tially affected both the ownership structure and the human resource development
practice in SOEs and newly founded companies in non-state sectors. During the reform
the non-state sectors have contributed significantly to China’s economic bloom (see
Hussain and Chen, 1999). Although the proportion of SOE output has been reduced
accordingly from three quarters in the early 1980s to less than one-third in the 1990s
(Tung, 1996), the state sector continues to be the most significant part of the economy.
In the 1990s, Chinese SOEs experienced a gradual transformation towards marketi-
zation, corporatization and privatization. In 1992 the corporatization of SOEs began
formally in order to establish a more clearly defined property rights structure. The
shares of existing SOEs are distributed among multiple owners. A few SOEs have been
transformed into shareholding companies, the shares of which are held by the state,
legal entities and employees. A small portion of these shareholding companies have
their shares listed on the Shanghai or Shenzhen stock exchanges. According to the data
released by the China Security Regulatory Commission, 1197 companies had become
listed on the Shanghai or Shenzhen stock markets by August 2002 and the majority of
them are former SOEs (China Security Regulatory Commission, 2002).
Before and after China entered the World Trade Organization (WTO), global com-
petition has brought immense changes to all enterprises in China. Learning has
become a popular topic for all the enterprises in both the private and public sectors.
Advances in technology have made the already challenging competition more severe.
The organization’s ability to learn more effectively and change faster than competitors
(both from domestic and international markets) may be the only source for gaining
sustainable competitive advantage (Goodall and Warner, 1999). In the SOEs, it is no
longer sufficient for the traditional organizational structure, training and development
practices to meet the competition challenges and the mounting customer demands for
high-quality products and services. SOEs, especially the listed ones, are required to
become ‘modern enterprises’ in terms of economic performance and management
structure. They must therefore re-examine continuously employees’ knowledge, skills
and cognitive abilities to guarantee that they can keep up with the competitive en-
vironment. Zhang, Yang and Zhang (2002) suggest that it is imperative for Chinese
SOEs to build learning practices in order to compete in a global market.
The present study aims to examine the applicability of the learning organization
concept and its measurement in a Chinese context. A well-developed instrument,
Dimensions of Learning Organization Questionnaire (DLOQ; Watkins & Marsick,
1993) is applied to, and validated in the Chinese context. By utilizing the DLOQ, this
study tries to investigate the differences in learning practices for selected categories of
Chinese SOEs.

Current status of Chinese SOEs

One issue of great concern in the reforming process is the poor performance of the
state-owner enterprises (SOEs). Indeed, the performance of SOEs deteriorated rapidly
from the early 1990s. Bruton et al. (2000) investigated human resource related problems
for foreign private firms in China and identified various issues including selection,
training, appraisal and reward, among others. It was found that the Chinese workforce
does not appear ready to function in a market environment. Employees were unac-
customed to doing work beyond the job descriptions and they tend to be risk averse.
These observations can also partially explain the overall poor performance of Chinese
SOEs. Gu (1997) observed a dilemma for many SOEs in a transitional environment:
those who should be hired cannot get in (because of many restrictions), those who
should be retained are leaving (for better places such as foreign firms or joint ventures),
and those who should be fired are staying (for reasons of social stability or mandated
by government). Zhang (1999) points out that current training practices in SOEs do

© Blackwell Publishing Ltd. 2004. Learning organization in mainland China 259

not meet the needs nor can be adapted to those needed in the changing marketplace.
A National Committee of Economy and Trade survey reveals that 50–60% of money-
losing enterprises can be attributed to the lack of professional managers. What the
majority of enterprises need most are managers who understand the market economy
and are capable of leading their firms to compete in the market system.
Much of the literature on corporate reforms to SOEs of developing countries argues
that restructuring SOEs into shareholding corporations will make them more efficient
(Baglione and Clark, 1997). Managers of listed companies are expected to be more
disciplined since their firm receives greater scrutiny from shareholders and investors.
However, recent research has shown that contrary to expectations, managers of
Chinese listed SOEs have not changed their behaviours (Freund, 2001; Sheehan et al.,
2000). In addition, becoming listed companies does not guarantee better performance
for SOEs. For example, Young and McGuinnes (2001) observed that Chinese SOEs are
still struggling to generate promised returns after the start of issuing stocks on foreign
trading platforms. One aim of the current paper is to investigate the potential differ-
ence between listed and unlisted SOEs. Following the reasoning of corporate reform
literature, we hypothesize that the listed SOEs would outperform their unlisted coun-
terparts in terms of their learning practices and performance results.
Tung (1996) argues that there are many obstacles hindering the establishment of
modern enterprise systems for traditional SOEs, especially the ones in manufacturing
industries. These firms tend to have a low growth rate and to provide obsolete prod-
ucts. In contrast, the newly founded service-oriented companies continue to outper-
form the traditional manufacturing companies due to the government’s assistance in
terms of financing and favourable policies (Benson and Zhu, 1999).
Most of the service-oriented SOEs were founded after the late 1970s. They are
typically separated from their parent companies and became independent during the
ownership structure reform of the parent companies. In consequence, these service-
oriented companies do not have the burdensome tradition which could impede their
growth. This rapid development and more competitive market make learning an even
more important issue for service industry than the traditional manufacturing industry.
Anecdotal evidence shows that the service industry is more ready to accept modern
management practices and more capable to learn and change. We hypothesize that
companies in the service industry will exhibit more learning practices and outperform
the manufacturing companies in terms of the measures of learning organizations.

Learning organizations
The concept of learning organizations has been one of the principal concepts in man-
agement literature since 1990, and this concept has been introduced gradually to
China. At least two best selling English books on the subject have been translated into
Chinese. One is Senge’s (1990) The Fifth Discipline and the other is Watkins and Mar-
sick’s (1993) Sculpting the Learning Organization: The Art and Science of Systematic Change.
Chinese companies tend to be enthusiastic about this concept, and many companies
started to implement learning organization practices. However, a quick literature
search showed that no empirical research of learning organizations in the Chinese
context has been conducted. Little is known about whether the concept of the learning
organization, which originated in a Western context, is applicable in the Chinese
context.
In a study of Chinese executives’ learning practices, Chwee (1999) discovered that
managers from SOEs were more likely to be reactive instead of proactive in learning,
to learn by rote, and to be risk averse. Given the increasing advocacy of the concept
of the learning organization throughout the world, a series of questions seems mean-
ingful to enterprise reform in China. First, can Chinese SOEs become learning organi-
zations? Second, what are some learning practices or elements of learning culture that
can be identified as key indicators for building learning organizations in the SOEs?
The theoretical foundation for this study is Watkins and Marsick’s conceptualization
of the learning organization (1993, 1996; Marsick and Watkins, 1999). In a recent

260 International Journal of Training and Development © Blackwell Publishing Ltd. 2004.
comprehensive review of literature on learning organizations, Örtenblad (2002) devel-
oped a typology of the idea of the learning organization. It is suggested that there are
four understandings of the idea of the ‘learning organization.’ The first type is ‘old
organizational learning perspective’, which focuses on the storage of knowledge in the
organizational mind. Learning is viewed as applications of knowledge at different
levels. The second type is a perspective of ‘learning at work’, which views the idea of
the learning organization as an organization where individuals learn at the workplace.
The third type is a ‘learning climate’ perspective, which sees the learning organization
as an organization that facilitates the learning of its employees. The fourth type is
a perspective of ‘learning structure’, which regards the learning organization as a
flexible entity. According to Örtenblad (2002), Watkins and Marsick’s (1993) approach
is among few theoretical frameworks that cover all of four understandings of the idea
of the learning organization in the literature.
Watkins and Marsick (1993) define the learning organization as one that cap-
tures, shares and uses knowledge to change the way the organization responds to
challenges. Central to this theoretical framework of the learning organization are
seven complementary action imperatives: (1) create continuous learning opportu-
nities (Continuous Learning); (2) promote inquiry and dialogue (Dialogue and
Inquiry); (3) encourage collaboration and team learning (Team Learning); (4)
empower people toward a collective vision (Empowerment); (5) establish systems
to capture and share learning (Embedded System); (6) connect the organization to
its environment (System Connection); and (7) provide strategic leadership for
learning (Strategic Leadership) (Watkins and Marsick, 1993). They further devel-
oped the Dimensions of Learning Organization Questionnaire (DLOQ) assessing
the learning organization on seven dimensions. The DLOQ has proved to be a
valid and reliable assessment tool that can be used to examine the extent to
which organizations embrace the practices and actions associated with the seven
action imperatives (e.g., Lien et al., 2002; Hernandez, 2000; Yang et al., 2004). A
recent study using the DLOQ has demonstrated a significant correlation between
learning organization practice and firm’s time-lagged actual financial performance
(Ellinger et al., 2002).

Research questions
The purpose of this study is to determine the applicability of the concept of the
learning organization, based on the theoretical framework proposed by Watkins and
Marsick (1993, 1996, 1997), in the Chinese context and to identify the gaps between
different Chinese SOEs at various stages of enterprise reform (independently-listed
versus unlisted) and in different industries. The study will first examine the psycho-
metric properties of a Chinese version of the DLOQ in the mainland Chinese
context, and then explore the differences in seven dimensions of DLOQ between
traditional SOEs and independent listed companies and companies providing
products versus services. Specifically, this study addresses the following research
questions:
1. Is the Chinese version of the DLOQ a reliable instrument in terms of internal
consistency for the proposed dimensions of the learning organization?
2. Is the seven-dimension framework of the learning organization proposed by
Watkins and Marsick applicable to Chinese organizations?
3. To what extent can the perceived measures of organizational performance be
explained by learning characteristics measured by the Chinese version of the
DLOQ?
4. Are there significant differences in the DLOQ subscales between Chinese
companies of different industries, specifically, manufacturing and service
industry?
5. Are there significant differences in DLOQ subscales between traditional Chinese
SOEs and the independent listed companies?

© Blackwell Publishing Ltd. 2004. Learning organization in mainland China 261

 Method
Instrument
In order to collect the necessary data to test the applicability of the learning organiza-
tion concept in the context of mainland China, we used a Chinese version of the DLOQ
and the survey research method was used. The DLOQ has been translated and used
in different cultural contexts. It has been used in Malaysian organizations (Sta. Maria
& Watkins, 2001). Hernandez (2000) reported findings from translation, validation and
adaptation study of the DLOQ in a Latin American context. It was found that the
Spanish version of the instrument was a reliable and valid instrument. The Chinese
version of the DLOQ was directly translated from the original version by a group of
Chinese researchers and has been empirically validated in a Taiwanese context in the
format of traditional characters or fonts (Lien et al., 2002). The instrument used in this
study is almost identical to the previously reported version except some Chinese
characters were printed in simplified fonts. The written language used in Taiwan and
mainland China is almost the same except for the font system. While the Taiwan region
maintains traditional characters, mainland China has adopted a simplified font
system.
The DLOQ includes 43 items assessing learning activities in the organization. Six
additional items were included to measure Knowledge Performance and another six
items were used to assess Financial Performance. In measuring these two perceptual
outcome measures, participants are asked to indicate their assessment of the organi-
zation’s current performance when compared to the previous year. They assess
financial performance by return on investment, average productivity per employee,
response time for customer complaints, etc. Knowledge performance was assessed by
customer satisfaction, the number of suggestions implemented, the number of indi-
viduals learning new skills, etc. Reliability estimates of these two performance scales
are provided in Table 3 on page 266.

Sample
A convenience sample was used for this study due to the complexity and large size
of Chinese enterprises. We conducted a survey of middle-level managers in six differ-
ent companies. Table 1 lists some organizational characteristics for these six firms. A
total number of 477 responses were collected from the six firms.

Data analysis
To address the first research question, an item analysis procedure was used to assess
the internal consistency for each of seven dimensions of the learning organization in
the context of mainland China. Each of the items included in the DLOQ was examined
in terms of its correlation with the proposed dimension. Cronbach’s alpha was used
to assess the reliability of the scale.
Confirmatory factor analysis (CFA) was used to determine the applicability of the
factor structure of the learning organization in the Chinese context. The CFA was
conducted with the program LISREL 8 (Jöreskog and Sörbom, 1989, 1993a,b). CFA is
a statistical procedure that examines the construct validity of an instrument with pre-
specified dimensions and associated measurement items. In this study, we are par-
ticularly interested in assessing whether the seven-dimension structure proposed in
the DLOQ constitutes an adequate measurement model for the learning organization
concept in the Chinese context. Yang et al. (2004) conducted an extensive series of
exploratory and confirmatory factor analysis and found that a reduced 21-item
measurement model yielded superior fit indices than the original 43-item model.
Consequently, two measurement models were examined to assess the adequacy of the
seven-dimension factor structure of the DLOQ in the new context: one for all 43
learning organization items and another for the reduced set of 21 items.

262 International Journal of Training and Development © Blackwell Publishing Ltd. 2004.
 © Blackwell Publishing Ltd. 2004.
Table 1: Sample firms and their characteristics

Firm Region Number of Main products Listed in Industry Sample size

employees or services stock market orientation

1. Bus LLC Beijing, Capital 600 Public transportation Listed Service 104
2. Petrochemical manufacturer Central China 700 Plastic raw materials Not listed Manufacturing 102
3. Publisher LLC Beijing, Capital 130 Books, audio and video Not listed Service 29
products
4. Pharmaceutical Company Southwest China 650 Pharmaceutics Listed Manufacturing 24
5. Petro-Pipe Line Company Northern China 1500 Gas and Petro-Pipe Line Not listed Service 161
6. Electronics Northern China 720 Communication and Listed Manufacturing 57
network equipments

Learning organization in mainland China 263

 In answering the third research question, canonical correlation was selected to assess
the association between dimensions of the learning organization and perceptual mea-
sures of firm performance. This is a technique for examining the association between
two sets of variables (Stevens, 1996). The underlying principle is to develop a linear
combination of each set of variables (both dependent and independent variables) in a
manner that maximizes the linear correlation between the two sets. The canonical
analysis was conducted by MANOVA procedure using SPSS.
Questions four and five were addressed by utilizing independent sample t-test. The
purpose of this test was to identify differences among different types of organizations.

Results
Table 2 presents the means, standard deviations, scale reliabilities and inter-
correlations for the variables in the study. All of the correlations among seven dimen-
sions of the learning organization are positive and significant at p < 0.01 level, ranging
from 0.38 to 0.75. This result indicates reasonable convergent validity for all of the
proposed seven dimensions as correlations among the seven dimensions as reflections
of the learning organization construct are positive and statistically significant. Further,
the correlations between measures on seven dimensions of learning and two percep-
tion measures of organizational performance are positive and statistically significant
at p < 0.01 level, ranging from 0.15 to 0.43. This result suggests the measures of seven
dimensions of learning have predictive validity in explaining perceived organizational
performance.
An item analysis procedure was used to assess the internal consistency for each of
the seven proposed dimensions, and Cronbach’s alpha was used to assess the reliabil-
ity. Each of the items included in the DLOQ was examined in terms of its correlation
with the proposed dimension. Reliability estimates of the seven dimensions based on
the coefficient alpha are presented in the diagonal of the correlation matrix in Table 2.
The results demonstrate that measures for the seven dimensions of learning are inter-
nally consistent and thus tend to be reliable in the Chinese context (coefficient alpha
ranges from 0.78 to 0.85). Two perceptual measures of organizational performance
also have shown acceptable reliability estimates (coefficient alpha was 0.80 and 0.81,
respectively). Table 3 lists a comparison of the reliability estimates for the measures
included in the DLOQ in this study and those estimates in other studies using the
same scale. The comparison reveals that the Chinese version of the DLOQ has accept-
able reliability estimates that are comparable with the previous studies conducted in
other countries.
An item analysis procedure further revealed useful results for future revision and
modification of the scale in the Chinese context. The reliability estimates for Inquiry
and Dialogue and Team Learning in this study (all were at 0.78) were slightly lower
than those in other studies (ranging from 0.81 to 0.89, and from 0.79 to 0.86, respec-
tively). A careful examination of the item analysis included in each of the measures in
the subscales suggested that there is no particular inadequate item that should be
excluded from that subscale. The coefficient alpha reached a slightly higher level (a =
0.79) when item 13 was deleted from the item analysis of Inquiry and Dialogue.

Factor structure of learning organization

Confirmatory factor analysis (CFA) was used to determine whether the seven-factor
structure of the learning organization construct proposed by Watkins and Marsick is
applicable in the Chinese context. Table 4 presents the CFA results. Six criterion indices
were chosen to evaluate the fit between the proposed measurement model and that
generated from the sample. The indices selected were the traditional chi-square (c2)
test, Jöreskog and Sörbom’s (1989) goodness of fit index (GFI) and goodness of fit index
adjusted for degree of freedom (AGFI), Bentler’s (1990) comparative fit index (CFI),
Bentler and Bonett’s (1980) non-normed fit index (NNFI), Steiger’s (1990) root mean

264 International Journal of Training and Development © Blackwell Publishing Ltd. 2004.
 Table 2: Means, standard deviations, reliabilities and subscale zero-order inter-correlations

© Blackwell Publishing Ltd. 2004.

Variables M SD 1 2 3 4 5 6 7 8 9

1. Continuous Learning 3.74 0.78 (0.80)

2. Inquiry and Dialogue 3.78 0.74 0.70** (0.78)
3. Team Learning 3.77 0.73 0.58** 0.69** (0.78)
4. Embedded System 3.55 0.85 0.55** 0.58** 0.62** (0.82)
5. Empowerment 3.75 0.83 0.45** 0.55** 0.66** 0.61** (0.82)
6. System Connection 4.11 0.94 0.36** 0.39** 0.45** 0.41** 0.70** (0.84)
7. Provide Leadership 4.00 0.91 0.38** 0.38** 0.44** 0.46** 0.68** 0.75** (0.85)
8. Financial Performance 4.40 0.88 0.22** 0.23** 0.15** 0.20** 0.25** 0.25** 0.32** (0.80)
9. Knowledge Performance 4.37 0.95 0.29** 0.26** 0.22** 0.27** 0.33** 0.37** 0.42** 0.67** (0.81)

Note: Sample size = 477 for all analyses, each dimension is measured by 6-point scale.
Internal consistency estimates (coefficient alpha) are presented in the diagonal.
**p < 0.01.

Learning organization in mainland China 265

 Table 3: Reliability estimates for the DLOQ measures

Dimensions America Colombia China China

(Taiwan) (Mainland)

Yang et al. Ellinger et al. Hernandez Lien et al. Current Study

(N = 469) (N = 208) (N = 906) (N = 679) (N = 477)

Continuous Learning 0.79 0.81 0.80 0.72 0.80

Inquiry and Dialogue 0.85 0.86 0.81 0.89 0.78
Team Learning 0.84 0.85 0.79 0.86 0.78
Embedded System 0.80 0.85 0.81 0.71 0.82

266 International Journal of Training and Development

Empowerment 0.75 0.84 0.81 0.75 0.82
System Connection 0.82 0.87 0.80 0.89 0.84
Provide Leadership 0.86 0.89 0.84 0.91 0.85
Financial Performance 0.71 0.75 N/A 0.89 0.80
Knowledge Performance 0.74 0.80 0.82 0.87 0.81

© Blackwell Publishing Ltd. 2004.

 Table 4: Fit indices for measurement models of the DLOQ

Fit Index 43-items (full version) 21-item (abbreviated version)

c2 3175.94 632.57
df 831 167
c2/df 3.82 3.79
RMSEA 0.077 0.077
RMR 0.065 0.056
GFI 0.75 0.89
AGFI 0.72 0.85
NNFI 0.77 0.85
CFI 0.79 0.88

square error of approximation (RMSEA), and Jöreskog and Sörbom’s (1989) root mean
square residual (RMR). The GFI and AGFI reflect the proportion of joint amount of
data variance and covariance that can be explained by the measurement model being
tested. The CFI is a relative fit index that compares the model being tested to a baseline
model (null model), accounting for the degrees of freedom. The RMR is a measure of
the average of the fitted residuals. The RMSEA represents a real advance in the
evaluation of model fit from both a statistical and a conceptual viewpoint. Browne and
Cudeck (1993) argue that because theoretical models are at best approximations of
reality, the null hypothesis for any measurement/structural equation model (i.e., the
conventional chi-square test that the data fits the model perfectly) will rarely be true.
Rather than testing the null hypothesis of exact fit between the covariance matrix of
sample and that of model for population, RMSEA establishes a hypothesis of close fit
between the model and population. RMSEA values of 0.05 or less indicate a very close
fit between sample and theoretical model, accounting for degrees of freedom. Values
less than 0.08 reflect reasonably well fitting models (Browne and Cudeck, 1993).
The CFA results for the Chinese version of the DLOQ offered further evidence
for the dimensionality of the learning organization proposed by the theorists.
Although the seven-dimension factor structure with 43 items fitted the data
moderately well, the abbreviated 21 items of the same factor structure fitted the data
reasonably well. Nearly 90% of item variance and covariance was explained by the
seven-dimension factor structure (GFI = 0.89). Although neither non-normed fit index
(0.85) nor comparative fit index (0.88) was above 0.90, a critical level that has been
regarded normally as an indication of good model-data fit, indices of RMR and
RMSEA suggested that overall fit between the proposed seven-dimension model and
the data was reasonable (RMR < 0.06 and RMSEA < 0.08). In sum, the seven-dimension
framework of the learning organization proposed by Watkins and Marsick has dem-
onstrated its applicability to the Chinese organizations when their learning activities
were measured by the Chinese version of the DLOQ.

Learning organization practice and organizational performance

A canonical correlation analysis was conducted to examine the relationships between
learning organization practice and organizational performance. Two perceptual mea-
sures of organizational performance (finance and knowledge) were used as a set of
dependent variables, and measures on the seven dimensions of learning organization
were treated as predictors. Table 5 presents the multivariate statistical tests for the
canonical correlation analysis. The results indicate a statistically significant relation-
ship between the learning practice measured on the seven dimensions of the DLOQ
and the two perceptual measures of organizational performance (p < 0.001). Effect size
of the canonical correlation ranged from 0.11 to 0.12, which suggested that more than
ten per cent of the variability in the respondents’ perceptions of organizational per-

© Blackwell Publishing Ltd. 2004. Learning organization in mainland China 267

 Table 5: Multivariate tests of significance for canonical correlation

Test Name Value Approx. Hypoth. Error Sig of Effect

F Df df F Size

Pillai’s Trace 0.225 8.43 14 932 0.000 0.112

Hotelling-Lawley Trace 0.282 9.35 14 928 0.000 0.124
Wilks’ Lambda 0.778 8.89 14 930 0.000 0.118
Roy’s Greatest Root 0.212

Note: F Statistic for Roy’s Greatest Root is an upper bound.

F Statistic for Wilks’ Lambda is exact.

formance could be accounted for by the seven dimensions of the leaning organization.
Comparing these results to the previous studies (Ellinger et al., 2002; Lien et al., 2002;
Yang et al., 1998), the above results showed relatively small correlation between the
learning practice and perceived organizational performance. In addition to explana-
tion of greater measurement errors, this might be explained by the fact that organiza-
tional performance has been affected by much more complicated factors in the case of
Chinese SOEs.

Differences between different types of companies

Independent samples t-tests were used to identify differences between different types
of companies. It was reasoned that service-oriented firms will perform better than
traditional manufacturers. Furthermore, it was also hypothesized that those firms that
have been listed on the stock market, will perform better than those that have not been
listed, because of the implementation of the new enterprise system. Table 6 summa-
rizes the t-tests on seven dimensions of learning organization and two measures of
perceived organizational performance. Among the seven dimensions of learning orga-
nization, service-oriented companies significantly outperformed the traditional manu-
facturers on three dimensions: Continuous Learning (p < 0.01), Inquiry and Dialogue
(p < 0.01), and Embedded System (p < 0.01). However, manufacturers surpassed the
service firms on System Connection. No significant difference was found between the
two types of firms in three areas: Team Learning, Empowerment and Provide Leader-
ship. As anticipated, those in the service industry have perceived significantly better

Table 6: Differences on DLOQ dimensions between service and manufacturing companies

Dimensions Service Manufacturing df t

(n = 294) (n = 183)

M SD M SD

1. Continuous Learning 3.91 0.77 3.47 0.72 475 6.24**

2. Inquiry and Dialogue 3.86 0.76 3.66 0.69 475 2.81**
3. Team Learning 3.76 0.74 3.79 0.73 474 -0.49
4. Embedded System 3.66 0.85 3.38 0.83 475 3.61**
5. Empowerment 3.75 0.80 3.78 0.93 475 -0.51
6. System Connection 3.96 0.89 4.16 0.92 475 -2.31**
7. Provide Leadership 3.96 0.90 4.06 0.93 475 -1.19
8. Financial Performance 4.55 0.82 4.15 0.92 473 4.94**
9. Knowledge Performance 4.44 0.96 4.25 0.92 474 2.13*

*p < 0.05, **p < 0.01.

268 International Journal of Training and Development © Blackwell Publishing Ltd. 2004.
 Table 7: Differences on DLOQ dimensions between listed and non-listed SOEs

Dimensions Independent Non- df T

listed independent
company company
(n = 185) (n = 292)

M SD M SD

1. Continuous Learning 3.52 0.77 3.88 0.75 475 -5.04**

2. Inquiry and Dialogue 3.72 0.75 3.82 0.73 475 -1.52
3. Team Learning 3.69 0.77 3.82 0.71 474 -1.98*
4. Embedded System 3.56 0.95 3.55 0.79 475 0.10
5. Empowerment 3.77 0.89 3.76 0.83 475 0.31
6. System Connection 4.02 0.81 4.05 0.96 475 -0.28
7. Provide Leadership 4.00 0.87 4.00 0.94 475 -0.05
8. Financial Performance 4.65 0.81 4.24 0.89 473 5.13**
9. Knowledge Performance 4.58 0.88 4.23 0.97 474 3.93**

*p < 0.05, **p < 0.01.

Financial Performance (t = 4.94, p < 0.01) and Knowledge Performance (t = 2.13,

p < 0.05) than those from the manufacturing sector. To sum up, the t-test results suggest
that service-oriented companies have demonstrated better characteristics of the learn-
ing organization than manufacturing firms, and they reported better organizational
performance.
Table 7 reports the t-test results of the differences between independent listed and
not independent listed companies on the dimensions of learning organization and two
performance measures. The results indicate that independent listed companies failed
to score significantly higher than those that are listed on the stock market in all seven
areas of the learning organization. On the contrary, firms that are listed have demon-
strated significant lower scores on two dimensions, Continuous Learning (t = -5.04,
p < 0.01) and Team Learning (t = -1.98, p < 0.05), than the unlisted firms. As expected,
those listed companies have reported significantly higher Financial Performance (t =
5.13, p < 0.01) and Knowledge Performance (t = 3.93, p < 0.01) than those from the
manufacturing sector. In summary, the above analyses indicate that although firms
listed on stock market have demonstrated significantly higher financial and knowledge
performance than those that have not been listed, listed firms failed to show better
practice of the learning organization measured on seven dimensions of the DLOQ.

Discussion
This study sought to explore the applicability of the concept and the measure of
the learning organization in the mainland Chinese context and to identify differences
between various types of state-owned-enterprises (SOEs). Using a convenience sample
from six SOEs in various regions and the DLOQ as an assessment instrument, we
tested whether the concept of the learning organization, which originated in Western
industrialized countries, is a valid construct in the Chinese context. We further exam-
ined the differences between different types of firms on the dimensions included in
the DLOQ.
Overall, this study’s findings can be interpreted as lending support to applicability
in a Chinese context of the learning organization concept proposed by Watkins and
Marsick in the DLOQ. When a Chinese version of the DLOQ was used for 477 res-
pondents from various SOEs, the measures of the instrument have demonstrated
acceptable reliability estimates in terms of internal consistency. Few items have been
identified to be less adequate in the Chinese context, suggesting some modifications

© Blackwell Publishing Ltd. 2004. Learning organization in mainland China 269

fitting to the different cultural contexts are needed in future studies. The CFA results
suggested that the factor structure of the seven-dimension measure of learning orga-
nization tend to be applicable in the Chinese context. The concept of the learning
organization was further validated by a significant correlation between learning
culture measured by the DLOQ and two perceived indicators of organizational
performance.
Different types of organization tend to have different learning cultures and practices
assessed by the DLOQ. As expected, service firms have demonstrated significant
higher scores on several dimensions of the DLOQ and organizational performance
measures than manufacturing firms. However, those firms listed on the stock market
failed to show higher learning practices measured by the DLOQ than their unlisted
counterparts, though they tend to enjoy significantly higher performance perception.
At first glance, this result was puzzling as one would reason that listed companies
should perform better than their unlisted counterparts. Nevertheless, this might be
explained by the weak link between firms’ managerial practice and the fact of whether
the firm is listed or not. The primary role of the capital market is its monitoring
function, not the managerial one. Managers in the listed firms are not exposed to public
scrutiny in terms of implementing modern managerial practice. The relatively high
perception of a firm’s performance might be due to the fact that the current stock
market in China is relatively independent from the corporate strategy (Young &
McGuinnes, 2001).
To summarize and answer each individual research question we have raised, based
on our sample of SOEs we found that: (1) the Chinese version of the DLOQ is reliable
in terms of internal consistency, (2) the proposed seven-dimension framework of the
learning organization holds in the Chinese context, (3) learning characteristics can
explain around 11 percent of the variance in the perceived measures of organizational
performance, (4) companies in the service industry outperform manufacturing com-
panies in terms of the two perceptual performance and three out of seven dimensions
of the DLOQ (Continuous Learning, Inquiry and Dialogue, and Embedded System),
and (5) although the listed SOEs have higher ratings on the two perceptual perfor-
mance scales, these independent listed SOEs do not perform significantly better than
the traditional SOEs in terms of the seven dimensions of learning practices.

Implications
The results of this study provide meaningful implications for human resource man-
agement (HRM) and human resources development (HRD) research and practice. The
study demonstrates that the seven-dimension concept of the learning organization is
applicable in the Chinese context. The positive evidence of psychometric properties of
the Chinese version of the DLOQ suggest that this instrument is reliable and valid,
and thus can be used in organizational studies. On the other hand, a few dimensions
tend to show relatively low internal consistency, suggesting a need to develop more
culturally relevant measurement items for the construct of learning organization.
China is a country with a long and rich civilization, and thus studies in the Chinese
context should offer valuable ideas to the literature. Future studies are needed to
discover and conceptualize indigenous concepts of learning and learning organization.
In the age of globalization, scholars are challenged to provide adequate and commu-
nicative tools on many organizational concepts across different cultures. Consequently,
studies such as the current one can offer better understanding of some abstract con-
cepts, such as learning organization, across cultures.
The current study has several practical implications for HRM and management
practice, particularly for the on-going enterprise reform efforts in mainland China.
Establishing modern enterprise systems for SOEs has been one of the focal points in
the economic reform. Many SOEs, however, especially those in the manufacturing
sector, tend to fall behind during the reform process. They tend to be less market
oriented, maintaining old products or services, maintaining old organizational struc-
tures, and laying off more workers in the face of financial difficulties. Perhaps these

270 International Journal of Training and Development © Blackwell Publishing Ltd. 2004.
firms should learn from the practices that have been demonstrated to be effective by
those service firms. The current study clearly demonstrated the correlation between
strong learning culture and organizational performance. Those service-oriented firms
outperformed the manufacturing ones in financial performance and several areas of
learning practice. This suggests that some adequate organizational practices such as
continuous learning, inquiry and dialogue, can be borrowed and utilized in other
sectors. Those SOEs that have lagged behind their peers can catch up by adopting and
applying practices that have been identified to be adequate predictors of performance
through valid organizational assessment. Instruments like the DLOQ will provide
a valuable tool for such assessments in identifying organizational strengths and
weaknesses.
This study also has implications for firms listed on the stock market. Listing on the
stock market has been deemed as one of the important enterprise reform measures for
many SOEs. The study suggests that some listed companies have failed to demonstrate
desirable learning practices. This implies that a change of enterprise ownership does
not guarantee the realization of adequate management practices. In order to establish
a modern enterprise system, it is imperative to define clearly property rights for all of
the SOEs, but it is more imperative to implement effective organizational practice. The
learning culture cannot be created overnight; instead, building a learning organization
requires long lasting efforts for the SOEs. In order to join the new global competition
and reach prosperity, they must be more open to new knowledge and information

Limitations
Given the complicated nature of Chinese enterprise (during the economic transition
process and different geographic characteristics) and the large number of SOEs, it was
nearly impossible to conduct a random survey to identify these representative com-
panies. Consequently, this convenience sampling method presents some limitations
for the current study. In addition, the sample consists of only middle-level managers
who may not have accurate and comprehensive understanding of the organizational
learning practices of the whole company. Furthermore, we cannot gain access to the
actual financial data of the companies that are not listed on the stock market; thus only
perceptive organizational performance measures were gathered. Although the two
performance scales exhibit acceptable reliabilities, the results would be more informa-
tive if we had measured actual performance.
Care should be taken in generalizing the findings of this study to a larger context
in mainland China. Results of the study may be meaningful to identify practices of
learning organization only for those companies with similar organizational character-
istics as the ones included in this study and that are willing to participate in such data
collection. Future research on learning organizations in a Chinese context could avoid
the above limitations by randomly sampling the participating companies and the
individual participants and measuring actual time-lagged organizational performance
results.

References
Baglione, L. and Clark, C. (1997), ‘A Tale of Two Metallurgical Enterprises: Marketization and
the Social Contract in Russian Industry’, Communist and Post-Communist Studies, 30, 2, 153–80.
Benson, J. and Zhu, Y. (1999), ‘Markets, Firms and Workers in Chinese State-owned Enterprises’,
Human Resource Management Journal, 9, 4, 58–74.
Bentler, P. M. (1990), ‘Comparative Fit Indexes in Structural Models’, Psychological Bulletin, 107,
238–46.
Bentler, P. M. and Bonett, D. G. (1980), ‘Significance Tests and Goodness of Fit in the Analysis
of Covariance Structures’, Psychological Bulletin, 88, 588–606.
Browne, M. W. and Cudeck, R. (1993), ‘Alternative Ways of Assessing Model Fit’, in K. A. Bollen
and J. S. Long (eds), Testing Structural Equation Models (pp. 136–162, Newbury Park, CA: Sage).

© Blackwell Publishing Ltd. 2004. Learning organization in mainland China 271

Bruton, G. D., Ahlstrom, D. and Chan, E. S. (2000), ‘Foreign Firms in China: Facing Human
Resources Challenges in a Transitional Economy’, SAM Advanced Management Journal, 65, 36,
4–11.
China Security Regulatory Commission (2002, August), Monthly report. Retrieved September 5,
2002, from <http://www.csrc.gov.cn/CSRCSite/tongjiku/199911/default.html>.
Chwee, W. W. (1999), ‘Individual and Organizational Learning of Chinese Executives at
Compaq-China’, Advances in Developing Human Resources, 4, 69–82.
Cooper, C. and Zheng, E. (1998), ‘State-owned Enterprise Reform in China’, International Busi-
ness, 11, 22–23.
Ellinger, A. D., Ellinger, A. E., Yang, B. and Howton, S. W. (2002), ‘The Relationship Between
the Learning Organization Concept and Firms’ Financial Performance: An Empirical Assess-
ment’, Human Resource Development Quarterly, 13, 1, 5–21.
Freund, E. M. (2001), ‘Fizz, Froth, Flat: The Challenge of Converting China’s SOEs into Share-
holding Corporations’, Policy Studies Review, 18, 1, 96–112.
Goodall, K. and Warner, M. (1999), ‘Enterprise Reform, Labor-management Relations, and
Human Resource Management in a Multinational Context’, International Studies of Management
and Organization, 29, 21–36.
Gu, M. (1997), ‘Changing Compensation System is the Key to Retain Talents’, Postsecondary
Education and Human Resource Development, 15, 3. (Chinese)
Hernandez, M. (2000), ‘Translation, Validation, and Adaptation of an Instrument to Assess
Learning Activities in the Organization: The Spanish Version of the Modified Dimensions of
the Learning Organization Questionnaire’, in K. P. Kuchinke (ed.), 2000 Academy of Human
Resource Development Conference Proceedings (pp. 254–59, Baton Rouge, LA: Academy of
Human Resource Development).
Hussain, A. and Chen, J. (1999), ‘Changes in China’s Industrial Landscape and Their Implica-
tions’, International Studies of Management and Organization, 29, 3, 5–20.
Jöreskog, K. G. and Sörbom, D. (1989), LISREL 7: A Guide to the Program Applications (2nd. ed,
Chicago: SPSS Inc).
Jöreskog, K. G. and Sörbom, D. (1993a), LISREL 8: User’s Reference Guide (Chicago: Scientific
Software International).
Jöreskog, K. G. and Sörbom, D. (1993b), PRELIS 8: User’s Reference Guide (Chicago: Scientific
Software International).
Lien, B. Y., Yang, B. and Li, M. (2002, March), ‘An Examination of Psychometric Properties
of Chinese Version of the Dimensions of Learning Organization Questionnaire (DLOQ)
in Taiwanese context’, in T. M. Egan (ed.), 2002 Academy of Human Resource Development
Conference Proceedings (pp. 854–61, Bowling Green, OH: Academy of Human Resource
Development).
Marsick, V. J. and Watkins, K. E. (1999), Facilitating Learning Organization (Cambridge, Great
Britain: Gower Publishing Limited).
Örtenblad, A. (2002), ‘A Typology of the Idea of Learning Organization’, Management Learning,
33, 2, 213–30.
Sheehan, J., Morris, J. and Hassard, J. (2000), ‘Redundancies in Chinese State Enterprises: A
Research Report’, Industrial Relations, 39, 3, 486–501.
Sta. Maria, R. F. and Watkins, K. (2001), ‘Perception of Learning Culture and Concerns about the
Innovation on Use of Innovation: A Question of Level of Analysis’, in O. Aliaga (ed.), 2001
Academy of Human Resource Development Conference Proceedings (pp. 871–77, Baton Rouge, LA:
Academy of Human Resource Development).
Steiger, J. H. (1990), ‘Structural Model Evaluation and Modification: An Interval Estimation
Approach’, Multivariate Behavioral Research, 25, 173–80.
Stevens, J. (1996), Applied Multivariate Statistics for the Social Sciences (3rd ed. Hillsdale, NJ:
Lawrence Erlbaum Associates).
Tung, R. (1996), ‘Obstacles Hindering the Establishment of a Modern State-owned Enterprise
System in Mainland China’, Issues and Studies, 32, 2, 32–51.
Watkins, K. E. and Marsick, V. J. (1993), Sculpting the Learning Organization: The Art and Science
of Systematic Change (San Francisco: Jossey-Bass).
Watkins, K. E. and Marsick, V. J. (1996), In Action: Creating the Learning Organization (Alexandria,
VA: American Society for Training and Development).
Watkins, K. E. and Marsick, V. J. (1997), Dimensions of Learning Organization (DLOQ) [Survey]
(Warwick, RI: Partners for the Learning Organization).
Yang, B., Watkins, K. E. and Marsick, V. J. (1998), ‘Examining Construct Validity of the Dimen-
sions of the Learning Organization Questionnaire’, in R. J. Torraco (ed.), 1998 Academy of
Human Resource Development Conference Proceedings (pp. 83–90, Baton Rouge, LA: Academy of
Human Resource Development).

272 International Journal of Training and Development © Blackwell Publishing Ltd. 2004.
Yang, B., Watkins, K. E. and Marsick, V. J. (2004), ‘The Construct of Learning Organization:
Dimensions, Measurement, and Validation’, Human Resource Development Quarterly, 15, 31–55.
Young, M. N. and McGuinnes, P. B. (2001), ‘The Missing Link: Why Stock Markets Have Been
Ineffective in Chinese SOE Reform’, Business Horizons, 44, 4, 55–62.
Zhang, D., Yang, B. and Zhang, Y. (2002), ‘Challenges and Strategies of Developing Human
Resources in the Surge of Globalization: A Case of the People’s Republic of China’, in T. M.
Egan (ed.), 2002 Academy of Human Resource Development Conference Proceedings (pp. 193–200,
Bowling Green, OH: Academy of Human Resource Development).
Zhang, Y. (1999), ‘On the Current Status and New Thinking of Human Resource Development
in China’, Postsecondary Education and Human Resource Development, 24, 3, 35–42. (Chinese)

© Blackwell Publishing Ltd. 2004. Learning organization in mainland China 273