Technovation 24 (2004) 443–453

www.elsevier.com/locate/technovation

The multidimensionality of TQM practices in determining quality

and innovation performance — an empirical examination
Daniel I. Prajogo, Amrik S. Sohal ∗
Department of Management, PO Box 197, Caulfield East, VIC 3145, Australia

Abstract

This paper examines the multidimensionality of TQM in association with organizational performance. The primary proposition
examined in this study is that TQM embodies two different models of practices, mechanistic and organic, with each showing a
different role in the association with two different types of performance, quality and innovation. Using empirical data gathered
from 194 middle/senior managers in Australian firms, the findings support the proposition in pairing the mechanistic elements of
TQM with quality performance and the organic elements with innovation performance. Further results, however, fail to support the
proposition that organizations need to configure TQM practices in different ways for achieving different type of performance.
 2002 Elsevier Ltd. All rights reserved.

Keywords: TQM; Multidimensionality; Mechanistic; Organic; Quality; Innovation; Structural equation modeling

1. Introduction This paper examines the multidimensional aspects of

Prajogo and Sohal (2001) have presented a compre-
hensive literature review on the relationship between
TQM and innovation performance. They argued that the
need to examine this relationship arises from the fact
that, today, the basis for competitive advantage has
shifted from quality to innovation, and given that TQM
principles and practices were developed in the context
of quality management, it is important to evaluate their
suitability for pursuing innovation performance. In
reviewing a body of literature concerning the relation-
ship between TQM and innovation, they identified two
competing schools of thought with one suggesting that
TQM is positively related to innovation performance and
the other group contending that the implementation of
TQM principles and practices could hinder organizations
from being innovative.
As discussed in the next section, a number of scholars
have suggested that TQM actually embodies several
practices whose characteristics are different to each
other, highlighting the multidimensionality of TQM.
∗
Corresponding author. Tel.: +61-3-9903-2033; fax: +61-3-9903-
2979.
E-mail address: amrik.sohal@buseco.monash.edu.au (A.S.
Sohal).
TQM and uses this to resolve the debate concerning the
relationship between TQM and innovation by drawing
on the experience of Australian firms.
2. Literature review
As a theoretical proposition, the multidimensionality
of TQM has been suggested by a number of scholars
(Dean and Bowen, 1994; Lau and Anderson, 1998; Mor-
eno-Luzon and Peris, 1998; Sitkin et al., 1994; Spencer,
1994; Watson and Korukonda, 1995). In particular, Mor-
eno-Luzon and Peris (1998) suggest that the multi-
dimensionality of quality management can be easily
found by examining the various terminologies so far
introduced into the areas of quality control, quality
assurance, total quality control, company-wide quality
control, total quality management, and strategic quality
management. Among these many terminologies, they
argue that quality assurance (QA) and total quality man-
agement (TQM) are the best models in contrasting the
multidimensionality of quality management. This is
because the first is focused on controlling processes and
products to conform to and satisfy established require-
ments, whilst the latter is directed toward involvement
and commitment of management and employees, train-

0166-4972/$ - see front matter  2002 Elsevier Ltd. All rights reserved.
doi:10.1016/S0166-4972(02)00122-0
444 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
ing, learning, and internal cooperation or teamwork — in
other words, promoting the human aspects of the system.
When contrasting QA with TQM, they used basic
organizational design variables — formalization, stan-
dardization, and centralization — and proposed that in
general, QA was situated in the area with high levels of
these three variables, whereas TQM was located in the
opposite area. Considering the above description, it can
be concluded that what is designated as QA can be
clearly associated with the mechanistic or control
approach, and that the TQM approach can be linked with
the organic or learning approach. A similar view was
also held by Kekale and Kekale (1995) when differen-
tiating the “behavioristic” approach of TQM practices,
such as systematic measurement, control of work, stan-
dards, and statistical procedures from a “cognitive”
approach that emphasizes “soft” qualitative character-
istics, such as open management style, delegated
responsibility and autonomy.
Watson and Korukonda (1995) affirm that examining
the juxtaposition of different facets of TQM, particularly
the dichotomy between mechanistic and organic
elements, is problematic. This is, nonetheless, important
to facilitate theoretical insights and conceptual clarity of
TQM, as asserted by Watson and Korukonda (1995,
p.105):
Yet, the promoters of TQM are not as enthusiastic
about discussing the mechanistic aspects of TQM as
they are about its organic aspects. This is to be
expected considering the connotations of passivity,
subjugation, and suppression of freewill stirred up by
a mechanistic model. Yet, not to recognize them
would be tantamount to ignoring some basic tenets of
the TQM philosophy
Furthermore, Prajogo and Sohal (2001) have affirmed
that examining the dichotomy of TQM in terms of mech-
anistic and organic models is important for resolving the
controversies in the literature concerning the relationship
between TQM and innovation. In this context, the work
of Sitkin et al. (1994) and Spencer (1994) provide a
theoretical basis to build up the link between the multi-
dimensionality of TQM and innovation. In their argu-
ment, Sitkin et al. (1994) hold that under similar underly-
ing TQM precepts, organizations can apply two different
goals and practices based on two different orientations,
namely TQC (Total Quality Control) and TQL (Total
Quality Learning) with TQC being associated with qual-
ity in terms of conformance, and TQL being related to
innovation. In her seminal work, Spencer (1994) argues
that various practices under the TQM umbrella can be
categorized into several organizational models, including
the mechanistic and the organismic models. For
example, the stated goal of TQM to improve quality is
associated with the mechanistic model, because in prac-
tice the real objective of pursuing quality could well shift
into productivity and efficiency, something on which a
mechanistic organization focuses. On the other hand, the
ideas of employee empowerment and cross-functional
teamwork are closely linked to the organismic model1
A further link can be made between the arguments
of Sitkin et al. (1994) and Spencer (1994) with control-
orientation being strongly associated with the mechan-
istic model and learning-orientation with the organic
model. For example, Sitkin et al. associate the TQC
approach (for example, through the application of SPC
tools) with a cybernetic control system on the basis of
the similarity of two critical requirements: the need for
regulatory standards and the need for activities that are
sufficiently routine to be well understood. At the same
time, Spencer (1994, p. 453) affirms “In the mechanistic
model, stability is prized because it increases pre-
dictability, which, in turn, increases control”.
In this regard, when referring to the mechanistic or
control-oriented model, TQM will focus more on quality
by conformance, and thus appear to meet all negative
arguments concerning its relationship with innovation.
On the other hand, both TQL and an organic model are
more related to innovation. Sitkin et al. suggest that TQL
stresses development of new skills, exploration of new
arenas, and other innovative-like activities, whilst an
organic model has long been identified as instrumental
in supporting innovation in the literature on innovation
(Burns and Stalker, 1961).
Finally, the discussion of the multidimensionality of
TQM is concerned with its application in organizations.
In contrasting QA with TQM, Moreno-Luzon and Peris
(1998) argue that although each of these approaches
come from different conceptions and apply different
tools, they are not exclusive; indeed, TQM arises as an
evolution of the quality assurance (QA) approach and
both perspectives can coexist in the same organization
at different levels. The question then is how these two
contrasting practices co-exist within one organization.
Spencer (1994) argues that organizations that practice
TQM do not necessarily hold strictly to any one of her
three models; rather, they “oscillate” among them. It can
be inferred that under the umbrella of TQM, organiza-
tions can emphasize or promote the exercise of certain
practices over others. In other words, one could expect
various configurations of TQM practices implemented in
different organizations, particularly in the context of the
pursuit of different strategic objectives.
1
The term “organic” has been recognized in the literature on inno-
vation, and therefore from this sentence onward, the term “organic”
will replace the term “organismic”
 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
3. Research framework and hypotheses
In summary, the literature review above has addressed
three issues. First, it highlights the theoretical aspect of
the multidimensionality of TQM; second, it makes a link
between the multidimensionality of TQM with organiza-
tional performance in terms of quality and innovation;
and finally, it stresses the argument that organizations
will show various configurations in implementing TQM
practices in the context of pursuing different strategic
performance. In order to test these propositions, we
developed a research framework aimed at contrasting the
different elements of TQM in association with different
types of organizational performance.
As illustrated in Fig. 1, the research framework intro-
duces TQM as comprising mechanistic and organic
elements and relates these to quality performance and
innovation performance simultaneously. The simul-
taneous relationship between TQM and the two meas-
ures of organizational performance is also aimed at
specifically examining the impact of different practices
on different types of performance. The research ques-
tions of this study are articulated as follows:
앫 Research question 1 — Can TQM practices be
presented in a multidimensional model comprising
both mechanistic and organic structures?
앫 Research question 2 — Can the multidimensionality
of TQM practices be reflected in different associations
between its individual practices with different types
of performance?
앫 Research question 3 — Do organizations implement
TQM in different configurations of practices when
pursuing different types of performance?
The objective of this study is also in line with past
studies on TQM. Whilst most empirical studies on TQM
were aimed at examining the relationship between TQM
practices and organizational performance, particularly
quality, some of these studies (Dow et al., 1999; Flynn
et al., 1995a; Powell, 1995; Samson and Terziovski,
1999) have gone further to investigate which specific
Fig. 1. The research framework.
445
practices significantly determine organizational perform-
ance.
4. Research Instrument
The use of constructs has played an important role in
designing a survey instrument in management research.
In any research concerning behavioral elements, there is
no device that can precisely produce measurement
through a single metric unit, and researchers usually
employ two or more measures to gauge a construct or
scale (Ahire et al., 1996). Given that developing new
constructs or scales of measurement is a complex task,
we followed the suggestion made by Tata et al. (1999)
to, wherever possible, use pre-tested constructs from past
empirical studies to ensure their validity and reliability.
4.1. TQM measures
To measure the level of TQM practices in the organi-
zations, the framework developed by Samson and Ter-
ziovski (1999) was used as a basis and was comp-
lemented by several variables derived from other
models. The first reason is that the model has been used
in the largest study of Australian companies so far con-
ducted. Second, as argued by Samson and Terziovski,
their model constitutes the criteria of the Malcolm Bald-
rige National Quality Award (MBNQA) that has been
accepted to articulate TQM practices by several scholars
such as Juran (1995); Evans and Lindsay (1999); Ahire
et al. (1995), and Dean and Bowen (1994). The MBNQA
consists of six criteria of organizational practices and
one criterion of organizational performance. Third, and
most importantly, we hold that the criteria of the
MBNQA are suitable for demonstrating the dichotomy
of TQM practices in terms of mechanistic and organic
models. The TQM practices embodied in the six criteria
of organizational practices are leadership, strategy and
planning, customer focus, information and analysis,
people management, and process management; the
detailed description of each category can be found in
Samson and Terziovski (1999).
Leadership and people management are practices that
mostly relate to human relations aspects in the organiza-
tion and therefore exhibit practices that are more organic.
This is because these two constructs incorporate such
practices as sharing beliefs and values, providing role
models, empowerment, participative management, and
creating unity between departments, training and devel-
opment, creating a quality work environment, and com-
munication. The instrumental roles of these practices —
that reflect an organic style — in predicting innovation
performance has been well established in the past
empirical studies. Empowerment, for example, should
make people having a certain degree of autonomy less
446 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
constrained by technical or rule-bound aspects, and self-
efficacious in doing their work that will nurture and
develop innovative behaviors (Amabile and Grykiewicz,
1989; Spreitzer, 1995).
Customer focus and process management, on the other
hand, can be associated with a mechanistic or control
model. Spencer (1994) argues that by placing a greater
deal of emphasis on reducing variation, TQM tends to
be a mechanistic. In this respect, both customer focus
and process management are closely related to each
other on the basis that they both represent the major
components of quality assurance which are mainly about
reducing variation. Such practices as surveying customer
needs/expectations, transforming these into product
characteristics and specifications, formalizing and docu-
menting procedures (e.g., through ISO 9000
certification), and the use of statistical process control
(SPC) signify processes or activities that are carried out
under control or in a mechanistic fashion. This prop-
osition has been substantiated by the work of Germain
and Spears (1999) that indicated the positive relationship
between formalization and TQM. However, it is
important to note that in their study, the contents of
TQM construct were dominated by items that were simi-
lar to what were incorporated in process management
constructs in our study.
Finally, strategic planning is linked to information and
analysis in the sense that these two variables constitute
the basic elements of strategic management process. The
classical approach for strategic management process, as
described in the strategic management literatures starts
with external and internal environmental scanning, for-
mulation of the strategic choice, implementation process,
and concludes with control and evaluation (Hill and
Jones, 2001; Thompson and Strickland, 1998). Both stra-
tegic planning and information and analysis practices
well reflect the two edging phases, planning and evalu-
ation, of this sequential process. Given that these two
steps are usually conducted in a formal and systematic
manner (Hill and Jones, 2001), it is suitable to categorize
them into the mechanistic elements of TQM. It is
important to note, however, that they also differ from
the two previous mechanistic elements of TQM.
4.2. Quality performance measures
A quality performance has been reflected and meas-
ured in various ways in past empirical studies on TQM
because, as articulated by Reeves and Bednar (1994),
quality itself can be defined in several dimensions.
Amongst several empirical studies on TQM that have
been reviewed, the construct for measuring quality per-
formance developed by Ahire et al. (1996) was the one
most closely aligned with the purpose of this study. This
construct defined quality performance as comprising of
four items: reliability, performance, durability, and con-
formance to specification, and it has shown good validity
and reliability in their study.
4.3. Innovation performance measures
A review of past empirical studies on organizational
innovation also indicates that there are variations in mea-
suring innovation performance in organizations. For the
purpose of comprehensively capturing the aspects of
innovation performance, this study developed a construct
for measuring product and process innovation on the
basis of several criteria that have been conceptualized
and used in the previous empirical studies of innovation,
such as Cohn (1980); Miller and Friesen (1982);
Deshpande et al. (1993); Karagozoglu and Brown
(1988); Avlonitis et al. (1994); Subramanian (1996);
Hollenstein (1996); and Kleinschmidt and Cooper
(1991). These criteria are the number of innovations, the
speed of innovation, the level of innovativeness (novelty
or newness of the technological aspect), and being the
“first” in the market. The scope of the measures therefore
included areas that could be considered as “radical”
innovation. In regard to the measurement approach, per-
ceptual data were used in which respondents were asked
to evaluate the company’s innovation performance
against the major competitor in the industry to minimize
industry effects. The advantages of this approach are dis-
cussed in detail in the study by Kraft (1990).
5. Source of empirical data
Empirical data was obtained through a random survey
of 1000 middle/senior managers who had knowledge of
past and present organizational practices relating to con-
tinuous improvement and innovation in Australian com-
panies. The sample was selected randomly and encom-
passed various industry sectors, including both
manufacturing and non-manufacturing sectors. The
scope of analysis of this study was limited to one site
(or plant) per organization. A total of 194 managers
responded, whilst 150 questionnaires were returned to
the researchers with RTS (Return to Sender) message
due to the invalid address. By discounting the number
of RTS mails, the final response rate accounted for
22.8%. The proportion of the respondents was nearly
equal between manufacturing and non-manufacturing
sectors (52.5% and 47.5% respectively). Half of the
respondents were quality managers and/or
production/operations managers, one-third of them were
senior managers (General Manager or Managing
Director), and the rest were managers from other areas,
such as marketing, finance, human resource, and admin-
istration. Most companies responding to the survey were
certified to ISO 9000.
 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
6. Data analysis
6.1. Data reduction process
The data reduction process was conducted in order to
bring the eight constructs — each consisted of four to
six items — employed in this study into eight composite
variables. Six constructs (leadership, strategic planning,
customer focus, information and analysis, people man-
agement, and process management) constituted TQM
latent variables, and two constructs (product quality and
product innovation) constituted three organizational per-
formance measures. These eight constructs were sub-
jected to validity and reliability tests before a single
score can be calculated to represent each construct. Con-
firmatory Factor Analysis (CFA) using LISREL 8.30
was employed for examining construct validity of each
scale by assessing how well the individual item meas-
ured the scale. During the process, only one item was
dropped due to poor loading on its latent variable (i.e.,
customer focus scale). The Goodness of Fit Index (GFI)
of the eight constructs comfortably exceeded the 0.9 cri-
terion suggested by Hair et al. (1998), hence, estab-
lishing their validity. The reliability analysis following
the construct validity process was conducted by calculat-
ing the Cronbach’s alpha for each scale. The results
show that the Cronbach’s alpha measure for the eight
constructs satisfactorily meet the minimum criterion of
0.7 (Hair et al., 1998), hence, establishing their
reliability. The final results of construct validity and
reliability tests of the eight constructs are reported in
Table 1.
Having met the requirement of construct validity and
reliability, the composite value for each construct can be
calculated. Among several methods suggested by Hair
et al. (1998), mean value was selected to represent the
composite variables due to the simplicity of the method
without forfeiting the accuracy. The result is also
presented in Table 1.
Table 1
Construct validity and reliability
Construct No of items (Final) Goodness of Fit Index
Leadership 4 0.980
Strategic Planning 4 0.998
Customer Focus 5 0.976
Information & Analysis 4 0.991
People Management 5 0.974
Process Management 6 0.978
Product Quality 4 0.983
Product Innovation 5 0.970
447
6.2. The measurement model of the multidimensional
TQM practices
The multidimensionality of TQM practices was exam-
ined by grouping the six variables of TQM practices into
three sub-groups2 based on a theoretical concept as dis-
cussed in the earlier section. A measurement model
using the confirmatory factor analysis (CFA) method
was performed to validate the measurement model of
multidimensional TQM practices using second-order
factor analysis. This method is similar to the work of
Tamimi (1998). As illustrated in Fig. 2, the six variables
of TQM (leadership, strategic planning, customer focus,
information and analysis, people management, and pro-
cess management) were grouped into three TQM sub-
groups (TQM1, TQM2, and TQM3). TQM1 comprises
leadership (“lead”) and people management (“peop”)
variables, TQM2 comprises customer focus (“cust”) and
process management (“proc”) variables, and TQM3
comprises strategic planning (“plan”) and information
and analysis (“info”) variables. The structural equation
links the three latent TQM sub-group variables to the
single second-order latent variable (TQM).
The overall indices show acceptable values. The GFI
and AGFI are well above 0.9, whilst the values of
RMSEA and SRMR are below 0.05. The result also
demonstrates strong links between each observed vari-
able to its respective latent variable (i.e., TQM1, TQM2,
and TQM3), and between the three sub-groups to the
second-order latent variables (i.e., TQM) as indicated by
the significance and coefficients of the paths. Therefore,
in response to Research Question 1, our result supports
the validity of the proposition that TQM can be
presented in a multidimensional model.
2
We chose the term “sub-groups” instead of “groups” to indicate
that the six variables can actually be compounded into one group,
which is TQM.
Mean Standard Deviation Cronbach’s alpha
3.756 0.825 0.8580
3.567 0.901 0.8242
3.918 0.684 0.7853
3.543 0.878 0.7992
3.431 0.802 0.8303
3.601 0.707 0.7922
4.197 0.547 0.8839
3.377 0.697 0.8684
448 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
Fig. 2. The second-order measurement model for multidimensional TQM practices.
6.3. The structural relationship between TQM
subgroups and organization performance
The multidimensionality of TQM practices in terms
of their relationship with quality and innovation per-
formance was analyzed by examining the individual
relationship between each of the three TQM sub-groups
with the two types of performance. From a theoretical
point of view, the mechanistic elements of TQM
(represented by TQM2 and TQM3) were expected to
show stronger relationship to product quality than pro-
duct innovation, whilst the organic element of TQM (i.e.,
TQM1) should exhibit stronger association with product
innovation than product quality.
It is important to note here that we did not directly
impose the structural relationship as hypothesized in the
research framework by setting up the path diagrams
reflecting the relationship illustrated in Fig. 1. Instead,
we followed the exploratory approach along the process
by estimating all possible paths relating the three TQM
subgroups as the independent variables and the two
organizational performance measures as the dependent
variables, as illustrated in Fig. 3.
With three by two relationships, there were six paths
that needed to be estimated. However, the initial findings
showed that not all of these six paths were significant.
Fig. 3. The initial model of the structural relationship between TQM
sub-groups and two organizational performance measures.
The non-significant paths of this initial structural
relationship were then deleted, and every time one of
these paths was deleted the goodness of fit indices of
the model were improved. This process was continued
until the best competing model was established.
The best structural model, presented in Fig. 4, only
has two paths linking the exogenous and the endogenous
variables between TQM1 and product innovation, and
between TQM2 and product quality, whilst TQM3
shows no significant relationship with any of the organi-
zational performance variables, and therefore was
deleted from the model.
Both GFI and AGFI of this model are well above 0.9,
and both RMSEA and SRMR are well below 0.05, indi-
cating the robustness of the model. The significant paths
identified in the finding clearly demonstrate the multi-
dimensionality of TQM practices in terms of their differ-
ent power in predicting quality and innovation perform-
ance, as suggested by the literature. The significant
relationships between certain TQM sub-group and cer-
tain type of organizational performance have empirically
reflected the theoretical proposition outlined in the litera-
ture review concerning the contrast between mechanistic
and organic elements of TQM in relation to quality and
innovation performance where the mechanistic elements
(represented by TQM2) would best predict quality per-
formance, whilst organic elements (represented by
TQM1) would best relate to innovation performance. In
response to Research Question 2, the result therefore
suggests that different practices embodied in TQM show
a different role in predicting different types of organiza-
tional performance.
In particular, there is a concern with regard to the
result showing that the combination of the variables of
strategic planning and information and analysis did not
show a significant relationship with any type of perform-
ance. However, the study by Samson and Terziovski
(1999) — from which the model for TQM construct was
 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453 449

Fig. 4. The best competing model on the structural relationship between TQM subgroups and two organizational performance measures.

derived — also exhibited a similar finding where both

variables did not show any significant relationship with
organization performance. Similarly, in his study, Powell
(1992) re-evaluated the positive notion held by strategic
management scholars about the relationship between
strategic planning and performance and found that stra-
tegic planning did not satisfy the criteria of sustainable
competitive advantage. This therefore provides an
important insight from the point of view of strategic
management that it is the implementation process of the
strategy — in the form of managerial practices — that
will produce real outcomes rather than the planning and
evaluation processes.

6.4. ANOVA test for differentiating high and low

performance in quality and innovation
Fig. 5. Sample grouping based on product quality and product inno-
The final part of data analysis addresses Research vation performance.
Question 3, whether organizations exhibit different con-
figurations of TQM practices (i.e., scoring on certain
practices more than the other) when pursuing different
types of performance. This analysis began by categoriz-
ing the whole sample into four groups on the basis of
dichotomy between high and low performance for pro-
duct quality and product innovation. Based on the distri-
bution and central tendency of the sample, the cut-off
point for differentiating high and low performance was
determined at 4.00 for product quality and 3.25 for pro-
duct innovation. The result of this performance categor-
ization is presented in Fig. 5. Twenty-two firms could
not be used in this analysis due to missing responses for Fig. 6. Group means for TQM practices.
either quality performance or innovation performance.
The mean value of the six TQM practices and the two the highest level of TQM practices. Group 2 and Group
organizational performance measures for each group was 3 fall somewhere between Group 1 and Group 4, and it
then calculated and the result is plotted in Fig. 6. The is evident that Group 3 has a higher level of TQM prac-
plot indicates that all six TQM practices show a consist- tices than Group 2, except in the case of people manage-
ent pattern of variation between the four groups, except ment. Quantitative analysis to examine the significance
for people management. Group 1, as expected, has the of the difference in the level of TQM practices among
lowest level of TQM practices. Similarly, Group 4 shows the four groups was conducted using the ANOVA test.
450 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453

Table 2
Test for homogeneity of variances between the largest and the smallest group

TQM Variables Variance of Group 1 (N = 57) Variance of Group 3 (N = 26) Fmax

lead (Leadership) 0.83 0.23 3.61

plan (Strategic Planning) 0.72 0.50 1.44
cust (Customer Focus) 0.48 0.26 1.85
info (Information & Analysis) 0.83 0.51 1.63
peop (People Management) 0.66 0.35 1.89
proc (Process Management) 0.46 0.33 1.39
qual (Product Quality) 0.16 0.08 2.00
pdin (Product Innovation) 0.21 0.07 3.00

Prior to conducting the ANOVA test, homogeneity of
variances among the four groups needed to be tested to
ensure the robustness of the ANOVA test. Tabachnick
and Fidell (2001) recommend the use of Fmax in conjunc-
tion with sample-size ratios to serve this purpose. Fmax
is the ratio of the largest cell variance to the smallest.
If sample sizes are relatively equal (within a ratio of 4
to 1 or less for largest to smallest cell size), an Fmax as
great as 10 is acceptable (Tabachnick and Fidell, 2001,
p. 80). The result, as presented in Table 2, suggests that
no Fmax value exceeds 10, and therefore the four groups
of the sample meet the criteria of homogeneity of vari-
ance.
The result of the ANOVA test is presented in Table
3. The post hoc pair comparison of mean differences
among the four groups of the sample was conducted
using the Tukey/Kramer method due the unequal sample
size of the groups (Hinkle et al., 1994). The findings
generally indicate that the six practices of TQM (except
people management) exhibit a consistent trend among
the four groups. Significant gaps in the variables were
found between Group 1 and Group 3, between Group 1
and Group 4, and between Group 2 and Group 4. On
the other hand, there is no significant difference in TQM
practices between Group 1 and Group 2 (except for
people management), between Group 2 and Group 3, and
between Group 3 and Group 4 (except for people
management).
In particular, the analysis was focused on comparing
the level of TQM practices between Group 2 and Group
3. This is because these two groups have an antagonistic
performance in terms of product quality and product
innovation. The proposition derived from the literature
review expected that these two groups would exhibit dif-
ferent configurations in their level of TQM practices.
Since Group 3 was significantly higher in quality per-
formance but lower in innovation performance than
Group 2, it should score higher in the mechanistic
elements (i.e., customer focus and process management)
than Group 2. On the other hand, Group 2 should exhibit
higher scores in the organic elements (i.e., leadership
and people management) than Group 3. The result, how-
ever, did not produce such a variation since there was
no significant difference on the six variables of TQM
practices between Group 3 and Group 2. Therefore, in
response to Research Question 3, there is no supporting
evidence to suggest that TQM practices are implemented
in different configurations in pursuing different dimen-
sions of performance. This finding is also supported by
the fact that TQM1 and TQM2, as shown in Fig. 4, are
significantly correlated to each other (0.34).

Table 3
ANOVA test among companies with different levels of quality and innovation performance

Mean values Means are

Group 1 Group 2 Group 3 Group 4 different at
TQM variables (N = 57) (N = 37) (N = 26) (N = 52) p ⬍ 0.05

lead (Leadership) 3.29 3.58 3.76 4.22 1–3; 1–4; 2–4

plan (Strategic Planning) 3.16 3.43 3.71 3.98 1–3; 1–4; 2–4
cust (Customer Focus) 3.54 3.83 4.01 4.21 1–3; 1–4; 2–4
info (Information & Analysis) 3.09 3.45 3.64 3.91 1–3; 1–4; 2–4
peop (People Management) 2.95 3.36 3.32 3.91 1–2; 1–3∗; 1–4; 2–4; 3–4
proc (Process Management) 3.20 3.35 3.72 4.03 1–3; 1–4; 2–4
qual (Product Quality) 3.73 3.89 4.62 4.68 1–3; 1–4; 2–3; 2–4
pdin (Product Innovation) 2.77 3.81 2.90 3.96 1–2; 1–4; 2–3; 3–4

∗
significant at p ⬍ 0.1
 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
7. Discussion
The results of the SEM model, in general, provide
support for the multidimensionality of TQM in the sense
that TQM embodies two different models of practices,
mechanistic and organic, with each showing a different
role in association with two different types of perform-
ance, quality and innovation. Furthermore, the findings
support the proposition in pairing the mechanistic
elements of TQM with quality performance and the
organic elements with innovation. The contrasting roles
of different TQM practices in predicting performance are
also consistent with the results of previous studies. For
example, the study conducted by Ahire et al. (1996)
found that among several constructs under TQM prin-
ciples, customer focus had the strongest correlation with
product quality. Similarly, the work of Atuahene-Gima
(1996) concluded that customer focus was associated
with product conformance (quality) but not with product
newness (innovation). From a theoretical point of view,
this result is instructive in understanding the underlying
uniqueness of the role different TQM practices play in
determining different types of performance despite the
conventional wisdom that introduces TQM as a “pack-
age” or a unidimensional set of practices. Concerning
this issue, Dow et al. (1999) have urged that discretion
needs to be applied as to why the practices are being
adopted and what the likely benefits are.
However, we strongly suggest that one needs to put
in a precaution when drawing implications of this result.
For example, while leadership and people management
practices did not show a significant relationship with
product quality, we believe that it will certainly contra-
dict theoretical views if one simply concludes that
organizations can ignore these two aspects and only con-
centrate on customer focus and process management
practices for pursuing quality performance. It is absol-
utely clear from the literature that TQM has strongly
advocated the important role of leadership and people
for achieving a high level of quality performance
(Deming, 1986; Imai, 1986; Lawler, 1994). It is there-
fore important to reconcile these contradictory results.
The plausible explanation is that the relationship
between TQM practices (variables) and organizational
performance may not take place as a simple linear and
simultaneous model tested in this study; rather, it could
work in a more complex interrelationship among the
elements of TQM practices as examined in several stud-
ies, such as Anderson et al. (1994) and Flynn et al.
(1995b). No less important is the likelihood of moderat-
ing or mediating interrelationships among TQM prac-
tices in explaining variance of organizational perform-
ance that will enable us to identify indirect relationships
as well as direct relationships between TQM variables
and organization performance measures (Ahire et al.,
1996; Dow et al., 1999).
451
Further results in this study indicate that although the
multidimensionality of TQM has been established, there
is no supporting evidence to suggest that organizations
will emphasize certain practices more than others when
pursuing different strategic performances. This suggests
that these two different models of TQM practices coexist
within organizations, meaning that organizations usually
implement TQM in a holistic manner rather than in a
piecemeal fashion as consistently demonstrated in pre-
vious studies (Ahire et al., 1996; Dow et al., 1999;
Grandzol and Gershon, 1998; Samson and Terziovski,
1999). Also, our results do not indicate that the coexist-
ence of the mechanistic and organic elements of TQM
causes a situation where one undermines the other, or
vice versa. This finding therefore supports the argument
for the juxtaposition of different elements of TQM in an
organization even though these elements are contradic-
tory in nature (Moreno-Luzon and Peris, 1998). The
coexistence of the mechanistic and organic models
implies that TQM calls for a synthesis of these antagon-
istic elements within an organization, leading to the con-
cept of an “ambidextrous organization”, which is neces-
sary for advancing the research on TQM in the future
(Watson and Korukonda, 1995).
This issue becomes even more important when it is
viewed in relation to the fact that in today’s highly com-
petitive environment, organizations need to pursue a
more and more complex or multidimensional aspects of
performance, including quality, cost, delivery, flexibility,
responsiveness, and innovation. More importantly, these
different aspects of performance have become more and
more interrelated with each other, even accumulative
(Bolwijn and Kumpe, 1990; Corbett and Van Wassen-
hove, 1993). Therefore, it is plausible to consider that
organizations today need to pursue these multiple
aspects of performance, including quality and inno-
vation, simultaneously. The results reported in this paper
suggest that each of these two types of performance (i.e.,
quality and innovation) requires a different model of
practices; hence, it is necessary for organizations to
maintain practices that are even antagonistic in their nat-
ure, as mentioned earlier. Indeed, maintaining paradoxes
can be one of the biggest challenges for organizations
in today’s competitive environment (Thompson, 1998).
Finally, the coexistence of multidimensional TQM
practices can be explained from the contingency per-
spective. When contrasting TQC from TQL, Sitkin et al.
(1994) emphasize the role of the underlying situational
difference that drives the contingent suitability of each
of the two models. Especially, they suggest that the level
of uncertainty surrounding the implementation of TQM
would determine the effectiveness and the applicability
of these two different models, and the different level of
uncertainty existing in both the external environment
(i.e., industry and market characteristics) and the internal
environment (i.e., different functions) of the organiza-
452 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
tions. Innovation scholars have also agreed on the need
for different types of approaches in different innovation
stages, with the organic-type approach being required
during the innovation initiation or development stages,
and mechanistic-type approach being suitable for the
implementation stage (Watson and Korukonda, 1995).
We present a parallel argument by contending that the
dichotomy (mechanistic versus organic) of TQM can be
found along the value-chain processes with the organic
approach being suitable for such functions as research
and development (R&D), whilst the mechanistic model
is more appropriate for the production area. This study
could not scrutinize down to that level since its scope
was at organizational or firm level. However, this has
indicated opportunities for further research on the issues
of the multidimensionality of TQM.
8. Conclusion
In conclusion, in response to the first two research
questions, the results have verified the proposition that
both mechanistic and organic types of practices can
coexist under the umbrella of TQM, hence, establishing
the multidimensionality of TQM. The result has also
demonstrated the different role of each type of practice
in determining different measures of performance with
customer focus and process management practices being
associated with product quality whilst leadership and
people management being related to product innovation.
However, further results indicate no variation in the con-
figurations of TQM practices even when organizations
pursue different types of strategic performance. This
suggests that all TQM practices are equally important
although each of them has a different role in determining
different types of performance.
Further research should focus on identifying and
examining several key factors that drive the application
of the multidimensionality of TQM from both internal
and external perspectives. For example, how TQM prac-
tices are implemented in various industry sectors with
different levels of uncertainty and dynamism. It is also
important to explore the practices of TQM in different
functions or departments within the same organization;
for example, between marketing and production. Given
that, to date, most of the major research on TQM has
been focused at the organization level, such a study will
provide a significant contribution in developing a better
understanding of the flexibility and multidimensionality
of TQM applications in organizations.
References
Ahire, S.L., Golhar, D.Y., Waller, M.W., 1996. Development and Vali-
dation of TQM Implementation Constructs. Decision Sciences 27
(1), 23–56.
Ahire, S.L., Landeros, R., Golhar, D.Y., 1995. Total Quality Manage-
ment: A Literature Review and an Agenda for Future Research.
Production and Operations Management 4 (3), 277–306.
Amabile, T.M., Grykiewicz, N.D., 1989. The Creative Environment
Scales: Work Environment Inventory. Creativity Research Journal
2, 231–253.
Anderson, J., Rungtusanatham, M., Schroeder, R., 1994. A theory of
quality management underlying the Deming management method.
Academy of Management Review 19 (3), 472–509.
Atuahene-Gima, K., 1996. Market Orientation and Innovation. Journal
of Business Research 35 (2), 93–103.
Avlonitis, G.J., Kouremenos, A., Tzokas, N., 1994. Assessing the
Innovativeness of Organizations and its Antecedents: Project
Innovstrat. European Journal of Marketing 28 (11), 5–28.
Bolwijn, P.T., Kumpe, T., 1990. Manufacturing in the 1990s — Pro-
ductivity, Flexibility and Innovation. Long Range Planning 23 (4),
44–57.
Burns, T., Stalker, G.M., 1961. The Management of Innovation. Social
Science Paperbacks, London.
Cohn, S.F., 1980. Characteristics of Technically Progressive Firms.
Omega 8 (4), 441–459.
Corbett, C., Van Wassenhove, L., 1993. Trade-Offs? What Trade-
Offs? Competence and Competitiveness in Manufacturing Strategy.
California Management Review 35 (4), 107–122.
Dean, J.W., Bowen, D.E., 1994. Management theory and Total Qual-
ity: Improving Research and Practice through Theory Develop-
ment. Academy of Management Review 19 (3), 392–418.
Deming, W.E., 1986. Out of the Crisis. MIT Press, Cambridge, MA.
Deshpande, R., Farley, J.U., Webster, F.E. Jr, 1993. Corporate Culture,
Customer Orientation, and Innovativeness in Japanese Firms: A
Quadrad Analysis. Journal of Marketing 57 (1), 23–27.
Dow, D., Samson, D., Ford, S., 1999. Exploding the myth: Do all
quality management practices contribute to superior quality per-
formance? Production and Operations Management 8 (1), 1–27.
Evans, J.R., Lindsay, W.M., 1999. The management and control of
quality. South-Western College Publishing, Cincinnati, OH.
Flynn, B.B., Schroeder, R., Sakakibara, S., 1995a. Determinants of
quality performance in high- and low-quality plants. Quality Man-
agement Journal 2 (2), 8–25.
Flynn, B.B., Schroeder, R.G., Sakakibara, S., 1995b. The Impact of
Quality Management Practices on Performance and Competitive
Advantage. Decision Sciences 26 (5), 659–691.
Germain, R., Spears, N., 1999. Quality management and its relation-
ship with organizational context and design. International Journal
of Quality & Reliability Management 16 (4), 371–391.
Grandzol, J.R., Gershon, M., 1998. A survey instrument for standardiz-
ing TQM modeling research. International Journal of Quality
Science 3 (1), 80–105.
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multi-
variate Data Analysis. Prentice-Hall International Inc, Upper
Saddle River, NJ.
Hill, C.W.L., Jones, G.R., 2001. Strategic Management Theory — An
Integrated Approach. Houghton Mifflin Company, Boston, New
York.
Hinkle, D.E., Wiersma, W., Jurs, S.G., 1994. Applied statistics for the
behavioral sciences. Houghton Mifflin, Boston, MA.
Hollenstein, H., 1996. A composite indicator of a firm’s innovative-
ness — An empirical analysis based on survey data for Swiss
manufacturing. Research Policy 25 (4), 633–645.
Imai, M., 1986. Kaizen: The Key to Japan’s Competitive Success. Ran-
dom House, New York.
Juran, J.M. (Ed.), 1995. A History of Managing for Quality. ASQC
Quality Press, Milwaukee, WI.
Karagozoglu, N., Brown, W.B., 1988. Adaptive Responses by Con-
servative and Entrepreneurial Firms. Journal of Product Innovation
Management 5 (4), 269–281.
Kekale, T., Kekale, J., 1995. A mismatch of cultures: a pitfall of
 D.I. Prajogo, A.S. Sohal / Technovation 24 (2004) 443–453
implementing a total quality approach. International Journal of
Quality & Reliability Management 12 (9), 210–220.
Kleinschmidt, E.J., Cooper, R.G., 1991. The Impact of Product Innov-
ativeness on Performance. Journal of Product Innovation Manage-
ment 8 (4), 240–251.
Kraft, K., 1990. Are product and process innovations independent of
each other? Applied Economics 22 (8), 1029–1038.
Lau, R.S.M., Anderson, C.A., 1998. A three-dimensional perspective
of total quality management. International Journal of Quality &
Reliability Management 15 (1), 85–98.
Lawler, E.E., 1994. Total Quality Management and employee involve-
ment: Are they compatible? Academy of Management Executive
8 (1), 68–76.
Miller, D., Friesen, P.H., 1982. Innovation in Conservative and
Entrepreneurial Firms: Two Models of Strategic Momentum. Stra-
tegic Management Journal 3 (1), 1–25.
Moreno-Luzon, M.D., Peris, F.J., 1998. Strategic approaches, organi-
zational design and quality management — Integration in a fit and
contingency model. International Journal of Quality Science 3 (4),
328–347.
Powell, T.C., 1992. Strategic planning as competitive advantage. Stra-
tegic Management Journal 13, 551–558.
Powell, T.C., 1995. Total Quality Management as Competitive Advan-
tage: A Review and Empirical Study. Strategic Management Jour-
nal 16, 15–37.
Prajogo, D.I., Sohal, A.S., 2001. The relationship between TQM prac-
tices and innovation performance: a literature review and research
framework. Technovation 21 (9), 539–558.
Reeves, C.A., Bednar, D.A., 1994. Defining quality: alternatives and
implications. Academy of Management Review 19 (3), 419–445.
Samson, D., Terziovski, M., 1999. The relationship between total qual-
ity management practices and operational performance. Journal of
Operations Management 17 (4), 393–409.
Sitkin, S.B., Sutcliffe, K.M., Schroeder, R.G., 1994. Distinguishing
control from learning in Total Quality Management: A contingency
perspective. Academy of Management Review 19 (3), 537–564.
Spencer, B.A., 1994. Models of Organization and Total Quality Man-
agement: A Comparison and Critical Evaluation. Academy of Man-
agement Review 19 (3), 446–471.
Spreitzer, G.M., 1995. Psychological empowerment in the workplace:
dimensions, measurement, and validation. Academy of Manage-
ment Journal 38 (5), 1442–1465.
Subramanian, A., 1996. Innovativeness: Redefining the Concept. Jour-
nal of Engineering Technology Management 13, 223–243.
Tabachnick, B.G., Fidell, L.S., 2001. Using multivariate statistics.
Allyn & Bacon, Nedham Heights, MA.
Tamimi, N., 1998. A second-order factor analysis of TQM factors.
International Journal of Quality Science 3 (1), 71–79.
Tata, J., Prasad, S., Thorn, R., 1999. The influence of organizational
structure on the effectiveness of TQM programs. Journal of Mana-
gerial Issues 11 (4), 440–453.
Thompson, A.A., Strickland, A.J., 1998. Strategic management: con-
cepts and cases. Irwin/McGraw-Hill, Boston.
Thompson, K.R., 1998. Confronting the paradoxes in a total quality
environment. Organizational Dynamics 26 (3), 62–74.
Watson, J.G., Korukonda, A.R., 1995. The TQM jungle: a dialetical
analysis. International Journal of Quality & Reliability Manage-
ment 12 (9), 100–109.
453
Amrik S. Sohal is a Professor in the Depart-
ment of Management and Associate Dean
(Research Degrees) in the Faculty of Business
and Economics at Monash University, Mel-
bourne, Australia. He is also Co-Director of the
Australian Consortium for Effective Organis-
ations (ACEO). From 1991 to 2000, Professor
Sohal was Director of the Quality Management
Research Unit and from 1993 to 1997 served as
an Associate Dean (Research Development) and
Associate Dean (Graduate Teaching) for the
Faculty of Business and Economics at Monash
University. He holds a PhD in Manufacturing/Operations Management
from the University of Bradford Management Centre in the UK, as well
as a BEng (Hons) and an MBA, also from the University of Bradford.Pro-
fessor Sohal is Associate Editor for the journal Technovation and Asia
Pacific Editor of the International Journal of Quality and Reliability Man-
agement. He is a member of the Editorial Board of a number of journals
in the area of quality management, technology management and operations
management. Professor Sohal has authored or co-authored over 100 papers
published in refereed journals, as well as three books and a number of
chapters contributed to books. His current research interests are in
manufacturing/operations strategy, technology/information management,
quality management, supply chain management, lean/agile production sys-
tems and electronic business. He has received research grants from the
State and Federal Governments, the Australian Research Council and
Monash University. In 2001, Professor Sohal received the Vice-Chancel-
lor’s Award for Postgraduate Supervision.
Daniel Prajogo is a PhD student in the Depart-
ment of Management at Monash University. His
bachelor’s degree is in electrical engineering
and he completed his Master of Engineering in
Quality Management at RMIT University, Mel-
bourne, in 1996. He has been working as a lec-
turer in the Department of Industrial Engineer-
ing at Petra Christian University in Surabaya,
Indonesia since 1992, and he is currently on
study leave. His areas of research interest
include quality management, service quality,
strategic management and innovation.