Decision Sciences

Volume 35 Number 3
Summer 2004
Printed in the U.S.A.

Evaluating the Deming Management Model

of Total Quality in Services∗
Thomas J. Douglas and Lawrence D. Fredendall†
Department of Management, Clemson University, 101 Sirrine Hall, Clemson, SC 29634-1305,
e-mail: [email protected], [email protected]

ABSTRACT
This article uses the Deming management model developed by Anderson et al. (1994b)
as an initial template to analyze total quality in services. While the literatures addressing
quality management have developed separately for products and services, the founders
of total quality portrayed this management philosophy as universally oriented. Our study
first replicates two earlier studies that tested the Deming management model in manu-
facturing industries. Using hospitals as our unit of analysis, we realized findings similar
to the earlier manufacturing studies. Next, we used contributions from the MBNQA
literature to test an enhanced model. Our subsequent findings support the MBNQA con-
cept that “leadership drives the system that creates results” and provides evidence of the
ubiquitous importance of leadership for ensuring the success of a quality improvement
program. Finally, an anomaly of this study and those published earlier is the inabil-
ity to find support for the relationship between continuous improvement and customer
satisfaction. Integrating the substantial work in the service quality literature focused on
customer satisfaction measurement is recommended to future researchers to help resolve
this issue and further enhance the model.

Subject Areas: Deming Model, Service Quality, and Total Quality

Management.

INTRODUCTION
Services are the dominant segment in the economy of the United States. The per-
centage of workers involved in the service industries in the United States has
increased from 30 percent in 1900 to over 85 percent in 2000 (Bureau of Labor
Statistics, 2002). In addition, many manufacturing companies provide services as
well as products (e.g., automobiles and service dealerships).
Despite services being a large segment of the economy, the concepts of ser-
vice quality are not as well developed as those of manufacturing quality (Ghoba-
dian, Speller, & Jones, 1994). This may be because the manufacturing and service

∗ We thank the special issue guest editor, Kurt Bretthauer, and two anonymous reviewers for their very
constructive feedback. We are also grateful to Joel Ryman for his contribution of key performance data and
to Bill Judge for his insights and thoughtful comments in the design of the original study. An earlier version
of this article was presented at the 2003 Academy of Management Meetings in Seattle.
† Corresponding author.

393
394 Evaluating the Deming Management Model

literatures currently treat quality management differently (Harvey, 1998; Sousa &
Voss, 2002). Most service quality research uses the gap model to examine service
quality (Harvey, 1998). The gap model, which was first proposed by Parasuraman,
Zeithaml, and Berry (1985), considers five gaps between service performance and
customer expectations, but it does not directly consider many of the elements of
total quality.
The views of the founders of total quality management (i.e., Deming, Juran,
and Ishikawa) are prominent in the manufacturing literature, but not in services.
However, the theoretical foundation and methods of total quality support its use
for both products and services (Anderson, Rungtusanatham, & Schroeder, 1994b;
Waldman, 1994). In fact, Deming (1986) devoted an entire chapter of his book,
Out of Crisis, to service industries. In addition, the Malcolm Baldrige National
Quality Awards (MBNQA) program, established by the U.S. Congress in 1987,
encompassed seven categories that could be applied to any organization, whether
in manufacturing or services (Bell & Keys, 1998).
This article is a first step toward integrating the service quality and product
quality literatures. The article uses a conceptually based, empirically tested model
to evaluate potential commonalities between quality management concepts in man-
ufacturing and service environments. First, the study demonstrates that the total
quality or Deming management model developed by Anderson et al. (1994b), and
tested in manufacturing industries (see Anderson, Rungtusanatham, Schroeder, &
Devaraj, 1995; Rungtusanatham, Forza, Filippini, & Anderson, 1998), is theoret-
ically applicable to services. Second, this article tests the Deming management
model with data from hospitals. Third, the study incorporates contributions from
two additional bodies of literature. It uses the MBNQA and the service quality
literatures to suggest further enhancements for the Deming management model.
Fourth, the implications of the empirical findings are examined for both services
and manufacturing.

PRESENTATION AND EVALUATION

OF THE DEMING MANAGEMENT MODEL
The definition of each construct underlying the Deming management model, as for-
mulated by Anderson et al. (1994b), is displayed in Table 1. These seven constructs
were developed from extensive readings of published materials by W. Edwards
Deming as well as other quality experts. Insights from the readings were further
developed via a three-round Delphi study using academic and practitioner experts
to identify the concepts underlying Deming’s 14 points (Deming, 1986). The re-
search team then clustered these concepts into seven constructs to represent the
content of the Deming management method.
To establish the credibility of the seven constructs in the Deming manage-
ment method, each construct was compared to the existing management literature
(Anderson et al., 1994b). All of the constructs were supported by this literature.
The relationships between the constructs were developed into theory, displayed in
Figure 1, using a relations diagram (e.g., Goal/QPC, 1991).
There have been two empirical examinations of the Deming management
model in the literature. First, Anderson et al. (1995), using measures identified from
Douglas and Fredendall 395

Table 1: Constructs underlying the Deming management method (Anderson et al.,

1994b, p. 480).
Visionary Leadership The ability of management to establish, practice, and lead a
long-term vision for the organization, driven by changing
customer requirements, as opposed to an internal
management control role.
Internal and External The propensity of the organization to engage in noncompetitive
Cooperation activities internally among employees and externally with
respect to suppliers.
Learning The organizational capability to recognize and nurture the
development of its skills, abilities, and knowledge base.
Process Management The set of methodological and behavioral practices
emphasizing the management of process, or means of
actions, rather than results.
Continuous The propensity of the organization to pursue incremental and
Improvement innovative improvements of its processes, products, and
services.
Employee Fulfillment The degree to which employees of an organization feel that the
organization continually satisfies their need.
Customer Satisfaction The degree to which an organization’s customers continually
perceive that their needs are being met by the organization’s
products and services.

Figure 1: Proposed theory of quality management for services.

Organizational Process
System Outcomes

Internal and Continuous

External Improvement
Cooperation

Visionary Process Customer

Leadership Management Satisfaction

Learning Employee
Fulfillment

the world-class manufacturing research project (Flynn, Schroeder, & Sakakibara,

1994), found exploratory support for most of the hypothesized relationships.
Rungtusanatham et al. (1998) replicated the first study using manufacturing fa-
cilities located in Italy. Both the Anderson et al. (1995) and Rungtusanatham et al.
(1998) studies supported most of the relationships in the Deming management
model. However, both studies suggested that additional testing of the Deming
management model in other industries was necessary.
396 Evaluating the Deming Management Model

To evaluate and test this model in hospitals, we will first evaluate the con-
structs and relationships of the model to determine their relevance to service indus-
tries in general and to hospitals in the health care industry specifically. In addition,
we will investigate the theoretical and empirical contributions suggested by the
recent studies depicting and evaluating the theoretical framework underlying the
MBNQA criteria and their relationship to the Deming management model. This
is appropriate since the MBNQA criteria were designed to comprehensively cover
overall management requirements from the perspective of the diverse quality com-
munity (Bell & Keys, 1998).

Service and Health Care Literatures

Each construct in the Deming management model, as depicted in Figure 1, is
examined separately in the following subsections to determine its applicability
to the service and health care industries. This is accomplished by reviewing the
existing service quality and health care literatures and the MBNQA criteria.

Visionary leadership
There is mixed support in the service quality literature for the visionary leadership
construct. While Harvey’s (1998) review of service quality does not consider the
role of leadership, Chase (1996) reports evidence that superior leadership leads
to superior results in services. In addition, Foster, Howard, and Shannon (2002)
found that leadership was related to process improvement, teamwork, and employee
satisfaction in their analysis of government services.
Several health care articles addressed the importance of visionary leader-
ship in quality program implementation (e.g., Arndt & Bigelow, 1995; Bender
& Krasnick, 1995; Huq & Martin, 2000; Motwani, Sower, & Brashier, 1996).
Motwani et al. (1996) reviewed the quality implementation literature for hospi-
tals and suggested an implementation model led by top management. Arndt and
Bigelow (1995, p. 7) stated that top management had “the responsibility of imbu-
ing organizational members with a shared sense of purpose aimed at creating both
value for the customer and the committed involvement of organizational employ-
ees.” However, they also argued that top administrators in hospitals might not have
responsibility for some of the key organizational activities, such as those controlled
by physicians.
Finally, the importance of leadership in successfully implementing total qual-
ity in health care is recognized in the MBNQA’s health care criteria for performance
excellence (NIST, 2003). The first award criterion listed is “visionary leadership”
and it outlines the key role of senior administrators in successfully implementing
total quality in health care. Therefore, while the concept of visionary leadership
may be difficult to implement in hospitals due to their organizational structure, it
remains an important element of total quality management.

Internal and external cooperation

Both the service quality literature and the health care literature support the appli-
cability of the internal and external cooperation construct to the service industry.
The internal cooperation construct is similar to the concept of human resource
Douglas and Fredendall 397

focus as used in service research. Typical items included in human resource fo-
cus are communications, training, recognition of support for quality objectives,
and employee satisfaction (e.g., Cook & Verma, 2002). These are very similar to
the measures used by Anderson et al. (1995), which included communication and
employee involvement in decisions.
The external cooperation concept should be valid for those services that use
significant supplier inputs. For example, Callahan and Moretton (2001) found that
external involvement of sales and marketing with their suppliers shortened soft-
ware development time. Gittell (2002) examined provider-provider relationships
in health care, that is, relationships between service providers who jointly pro-
vide service to the patient, and found that increased internal coordination led to
increased customer satisfaction. For example, Douglas and Ryman (2003) argued
that relationships between hospitals and independent physician groups enabled the
continuum of care provided to patients to be more efficient and effective given the
utilization of complementary resources. Gittell (2002) also concluded that service
operations, which involve reciprocal interdependence, require a high coordination
level between service providers. Carmen et al. (1996) provided an in-depth re-
view and analysis of 10 hospitals and concluded that a culture supportive of group
affiliation and teamwork led to better quality results, including customer satisfac-
tion. Finally, the MBNQA’s health care criteria, in the section on valuing staff
and partners, discuss the importance of building partnerships, both internally and
externally, in a quality program.

Learning
Since most services are delivered through the contact of service personnel with the
customer, employee learning should have a significant impact on service quality.
However, there has been little empirical research about the impact of learning on
service quality. Roth and Jackson (1995) measured the operations capabilities of
service organizations as their organizational knowledge (i.e., institutional knowl-
edge expressed as system design choices and competent knowledgeable staff),
technological leadership, market acuity, and factor productivity. They found that
all of these operations capabilities and market acuity influenced service quality.
Huq and Martin (2000) found that the education and training of the health
care workforce in basic quality principles, statistics, and interpersonal skills, and
the creation of a learning environment were important for successful quality imple-
mentations in hospitals. In addition, Bender and Krasnick (1995) argued that while
organizational learning was a critical aspect of quality management in hospitals,
the typical health care organization did not provide a culture that allowed current
concepts and assumptions to be challenged. Therefore, even though the ability to
learn was important for quality management in hospitals, many hospitals’ cultures
lessened the ability to learn.
The MBNQA’s health care criteria include organizational and personal learn-
ing. These concepts are defined broadly, encompassing continuous learning within
the organization and the ability to adapt to environmental change. The expected out-
comes of successfully embracing this concept include efficiency, responsiveness,
and better performance.
398 Evaluating the Deming Management Model

Process management
Anderson et al. (1994b) viewed process management as being those practices that
focus on managing the means or actions taken in the process and not the results.
The applicability of the process management construct in the Deming management
model to services is supported by both the service and hospital literatures. There are
three major bodies of service process research. The first considers how to design a
service process, using such tools as service blueprinting (e.g., Shostack, 1984). The
second considers how customer contact is managed and how it influences the service
process and the customers’ perception of quality (e.g., Kellogg & Chase, 1995).
The third examines quality management techniques and procedures such as the use
of statistical quality control (e.g., Sureshchandar, Rajendran, & Anantharaman,
2001) and mistake proofing (e.g., Stewart & Chase, 1999) in services.
The hospital literature also recognizes the importance of process management
to service quality. For example, Shortell et al. (1995) included a hospital’s focus on
managing processes and systems in measuring its continuous quality improvement
(CQI)/total quality management (TQM) program, the implementation of which was
related to better perceived patient outcomes. In a study of 227 European hospitals,
Kunst and Lemmink (2000) found that process management, as defined in the
European Quality Awards, was positively related to quality program outcomes
representing both customer and employee satisfaction. The importance of process
management to service quality is recognized by the MBNQA program in health
care, which includes process management as one of its seven criteria.

Continuous improvement
There is limited support for the continuous improvement construct in the gen-
eral service quality literature, but strong support in the health care literature.
Sureshchandar et al. (2001) state that striving for continuous improvement is critical
to the achievement of service quality. In addition, Roth and Jackson (1995) found
an important role for continuous improvement in the firm’s ability to deliver high
service quality. They operationalized continuous improvement as organizational
knowledge, and found that organizational knowledge was the primary determinant
of service quality.
Hospitals often have Continuous Quality Improvement (CQI) programs.
Carmen et al. (1996) argue that the key for hospital success is a commitment
to continuous improvement. Routhieaux and Gutek (1998) measure the effective-
ness of CQI using a framework that includes continuous improvement based on
a customer focus. The MBNQA’s criteria measure the support of the continuous
improvement of processes and, ultimately, systems.
Hospitals receive external pressure for continuous improvement from the
Joint Commission on Accreditation of Health Care Organizations (JCAHO). The
JCAHO conducts periodic reviews, which focus on documented evidence of con-
tinuous improvement efforts (Westphal, Gulati, & Shortell, 1997).

Employee fulfillment
Anderson et al. (1994b) suggest that employee fulfillment be measured by job satis-
faction, job commitment, and pride in their work. The relevance of this construct to
Douglas and Fredendall 399

service quality is strongly supported. Sureshchandar et al. (2001) stress that firms
must focus on employee satisfaction because there is a high correlation between
employee perceptions of well-being and the customers’ perception of service qual-
ity. Hensel (1990) also found that service employee participation leads to employee
ownership of the service being delivered and that the quality challenge for service
companies is to create the correct environment for the employees to function within.
The hospital service literature suggests that employee fulfillment is necessary
for enhanced performance outcomes, including customer satisfaction, but may be
difficult to achieve. Arndt and Bigelow (1995) argued that gaining the committed
involvement of employees may be difficult in hospitals. However, Boerstler et al.
(1996) found that hospitals that ranked high on employee participation, teamwork,
and an adaptable, flexible culture were most successful in the quality management
initiatives. Huq and Martin (2000) echo these findings and add the need to have em-
powered employees with authority and responsibility to ensure successful quality
implementation. Shortell et al. (1995) found that successful quality implementa-
tions led to increased employee relations and satisfaction.

Customer satisfaction
Customer satisfaction is a very important concept in the service quality literature,
addressing how well the service and the service process meet customer expec-
tations (Harvey, 1998). Nilsson, Johnson, and Gustafsson (2001) found that the
internal quality practices of service firms influence customer satisfaction. In ad-
dition, Anderson, Fornell, and Lehmann (1994a) found that, in a service firm,
customer satisfaction influences a firm’s profitability, while Rust, Zahorik, and
Keiningham (1995) found that increased customer satisfaction in services leads to
higher customer retention.
Organizational processes in health care focus on the total satisfaction of
customers, that is, patients (Zabada, Rivers, & Munchens, 1998). Most articles
addressing quality management in health care list multiple outcomes, with patient
satisfaction as a key outcome. For instance, Arndt and Bigelow (1995) measure
improvement in efficiency and effectiveness in terms of cost controls, clinical
outcomes, and satisfaction for all customers including patients, their families, and
all internal constituents. Shortell et al. (1995) and the MBNQA program criteria
measure very similar categories.

Hypotheses
The review of the literature established the validity of the constructs in Figure 1
for services in general and health care in particular. In this section, we examine the
relationships between the constructs and develop formal hypotheses for testing.
Visionary leadership is core to the Deming management model. Leadership
is essential in order to create a service organization that has both internal and
external cooperation. Deming (1986) stressed the need to eliminate fear to improve
the process. Leadership is the primary determinant of the level of fear, which
controls the level of cooperation. This was supported by Anderson et al. (1995)
and Rungtusanatham et al. (1998) for manufacturing plants and the same rationale
seems to apply to services. So, it is hypothesized that:
400 Evaluating the Deming Management Model

Hypothesis 1: Visionary leadership is positively related to internal and

external cooperation.
Leaders control organizational learning by allocating the resources and choos-
ing whether or not to recognize and reward the learning that occurs. This link was
supported by both Anderson et al. (1995) and Rungtusanatham et al. (1998) for
manufacturing and should be true for services as well. So, it is hypothesized that:
Hypothesis 2: Visionary leadership is positively related to learning.
Internal and external cooperation was found to be important to process man-
agement in manufacturing by Anderson et al. (1995) and Rungtusanatham et al.
(1998). Internal and external cooperation should also be essential to service process
management. Internal cooperation should facilitate data sharing, the standardiza-
tion of processes, the visual tracking of defects, and the use of statistical tools
to identify problems, all emphasizing the management of the process. So, it is
hypothesized that:
Hypothesis 3: Internal and external cooperation is positively related
to process management.
The model in Figure 1 also proposes that learning leads to process man-
agement. This link was not supported by either the Anderson et al. (1995) or the
Rungtusanatham et al. (1998) study. However, the scales in their studies seemed
to measure only the amount of task training employees received. This study mea-
sures the total quality training the employees receive and how the organization uses
customer information. It is believed that total quality training allows employees
to learn from and respond to available customer information. This learning from
customer information should lead to process management, which includes making
data available and using data to evaluate performance. So, it is hypothesized that:
Hypothesis 4: Learning is positively related to process management.
The path from process management to continuous improvement was sup-
ported by both the Anderson et al. (1995) and Rungtusanatham et al. (1998) stud-
ies of manufacturing. It should also be significant in services. There is pressure on
services to continuously improve to meet rising customer expectations. It is argued
by many quality management researchers that process management is essential
to improvement. In this context, process management means analyzing current
performance and taking action. Without taking actions based on the analysis, im-
provement is unlikely. So, it is hypothesized that:
Hypothesis 5: Process management is positively related to continuous
improvement.
The path from process management to employee fulfillment was only weakly
supported by Anderson et al. (1995) and not supported by Rungtusanatham et al.
(1998) in manufacturing. Their scales measured employee pride in their work.
We view employee fulfillment in terms of employees being given the authority,
information, and tools necessary to do their job. Since service personnel must
use the processes available to them to respond quickly to and satisfy individual
Douglas and Fredendall 401

customer needs, employees’ fulfillment depends on superior process management.

So, it is hypothesized that:

Hypothesis 6: Process management is positively related to employee

fulfillment.

Figure 1 proposes that customer satisfaction is the result of both continu-

ous improvement and employee fulfillment. The link from continuous improve-
ment to customer satisfaction was not supported by Anderson et al. (1995) but
was marginally supported by Rungtusanatham et al. (1998). It is expected that
continuous improvement in services should lead to customer satisfaction. In our
study, business performance is measured in three ways—customer satisfaction,
perceived financial performance, and the JCAHO audit score. Continuous improve-
ment should lead to both improved financial performance and improved JCAHO
audit scores. Continuous improvement will do this by reducing waste, thus cutting
costs and improving finances. Continuous improvement should also improve audit
scores, since audits are so important to hospitals and top management is under in-
stitutional pressure to do well on the audits (Westphal et al., 1997). The MBNQA
criteria also include similar, multiple measures of business performance (Meyer &
Collier, 2001). So, it is hypothesized that:

Hypothesis 7: Continuous improvement is positively related to busi-

ness performance.

The link from employee fulfillment to customer satisfaction was supported

by Anderson et al. (1995) and not supported by Rungtusanatham et al. (1998)
in manufacturing. This link should be supported for services. Organizations that
allow employees to adapt to the customer’s needs using available processes and
information should perform better. It is believed that employee fulfillment will help
improve financial performance by leading to improved service, which enhances the
hospital’s reputation. This reputation should in turn lead to increased business at
the hospital. Employee fulfillment can improve audit scores, as employees take the
initiative to eliminate obstacles, which might prevent fulfilling the requirements of
the JCAHO audits. So, it is hypothesized that:

Hypothesis 8: Employee fulfillment is positively related to business

performance.

Measuring Framework Constructs and Other Variables

The constructs underlying the Deming management model were operationalized
using previously published scales in the TQM literature. When necessary, multiple
scales were used (e.g., internal and external cooperation). Table 2 lists the source
of the scales used to represent each construct, and displays their reliability scores
based on our study data.
The reliability and validity of these scales was extensively tested by their
creators. Additional factor and reliability analyses were conducted in this study,
since the number of items in most scales was reduced for efficiency reasons and this
is a new data set. The scales and associated items are contained in the Appendix.
402 Evaluating the Deming Management Model

Table 2: Theoretical constructs, measurement scales, sources, and reliabilities.

Reliability
Construct Scale Source (Alpha)
Visionary Leadership Top Management Team Saraph et al. (1989) .92
Involvement (6 items)
Internal and External Quality Philosophy (5 items) Zeitz et al. (1997) .85
Cooperation Supplier Involvement Saraph et al. (1989) .89
(6 items)
Learning Total Quality Training Saraph et al. (1989) .91
(6 items)
Customer Driven (5 items) Zeitz et al. (1997) .87
Powell (1995)
Process Management Management by Fact (6 items) Saraph et al. (1989) .92
Total Quality Methods Zeitz et al. (1997) .87
(5 items)
Continuous Continuous Improvement Zeitz et al. (1997) .90
Improvement (3 items)
Employee Fulfillment Structural exploration Khandwalla (1977) .67
(2 items)

As displayed in Table 2, the scale used in this study to measure visionary

leadership was initially developed by Saraph, Benson, and Schroeder (1989) to
measure top management team involvement. It was modified slightly to ensure
its suitability for hospitals, but is very similar to the scales used by Anderson
et al. (1995) in their study of the Deming management model. The scale fo-
cuses on the roles and participation of top management in leading the quality
initiative.
The scales to measure internal and external cooperation were different than
those used by Anderson et al. (1995). Two scales were used. The first scale, quality
philosophy, is used as a measure of internal cooperation. It measures commitment
and awareness of the firm’s mission and management’s role in preventing prob-
lems. Supplier involvement measures such things as knowledge of the supplier
and whether the hospital offers a long-term relationship to the suppliers. All of the
items assume that we have a cooperative relationship with the supplier. Use of both
scales allows us to capture the compete domain of this construct.
Anderson et al. (1995) measured learning as the amount of training the em-
ployees received. This article uses two different scales to measure learning. The
first looks at the total quality training given to employees. This seems to be a
more appropriate measure in the context of the Deming management model in
terms of providing the skills needed to implement quality. The second scale,
titled “Customer Driven Information,” measures what employees do to learn
about their customers’ needs, increasing their knowledge and ability to meet
these needs. Combined, these scales map well into the learning construct in this
model.
The process management construct was measured with two scales. The first
scale, management by fact, has some similarity to the scale used by Anderson et al.
Douglas and Fredendall 403

(1995), but it also measures who receives and uses quality data and whether the
data are used to manage quality. The second scale, total quality methods, measures
whether the employees use the quality tools to manage their process. This is a scale
developed to measure total quality practices (Zeitz, Johannesson, & Ritchie, 1997)
actually used in an organization, and is an appropriate measure of the extent to
which these are used in the daily process. Both scales focus on the availability of
information and the actions taken to manage processes.
The continuous improvement construct was measured using a scale that was
very similar to that used by Anderson et al. (1995). The employee fulfillment con-
struct was measured by Anderson et al. (1995) as the amount of pride the employee
takes in their work. A different approach was taken in this study. The structural
exploration scale used here to measure employee fulfillment was developed by
Khandwalla (1977) as a measure of organization design. It was felt that employees
who had access to the necessary data and were allowed to adapt as necessary to
complete their task requirements would be fulfilled. This is supported by a model of
empowered behavior that argues that resource and information sharing in the local
work environment leads to psychological empowerment and empowered behavior
(Robbins, Crino, & Fredendall, 2002).
While the Deming management model contains three outcome variables
(continuous improvement, employee fulfillment, and customer satisfaction), we
added two more variables pertinent to hospitals, financial performance and overall
JCAHO audit score. Financial performance was added because it is important for
the survival of hospitals. Survey respondents recorded their hospital’s relative per-
formance over the last 3 years (see Appendix). Obtaining performance information
on a primary basis has extensive precedence in the literature (Powell, 1995). The
five-item measure used in this study was adapted from scales used by Powell (1995)
and Ramanujam, Venkatraman, and Camillus (1986).
The second added outcome measure was the hospital’s overall JCAHO audit
score. This industry expert-rated measure is based on a comprehensive audit of
forty-nine standards related to internal hospital processes conducted once every
3 years. The maximum JCAHO overall score is 100 points, and hospitals are under
considerable institutional pressure to perform well on this audit (Westphal et al.,
1997).
Our measure of customer satisfaction combines three dimensions of the hos-
pital’s perceived performance relative to competitors, patient satisfaction, clinical
outcomes, and average length of stay. Superior performance with respect to all three
of these dimensions should allow the hospital to meet or exceed the expectations
of their customers.
Two control variables were also measured: the number of hospital beds and the
type of ownership. The number of hospital beds is a proxy for organizational size.
This measure is similar to those used in other studies of hospitals (e.g., Ketchen,
Thomas, & Snow, 1993). The number of hospital beds was obtained from the 1995
American Hospital Association’s Annual Survey of Hospitals. Hospital ownership
is a potentially important variable in this industry. For-profit hospitals may have
different organizational goals and unique groups of stakeholders when compared to
their not-for-profit counterparts (Zajac & Shortell, 1989). These data were collected
from the 1996 edition of the Hospital Blue Book.
404 Evaluating the Deming Management Model

MBNQA Enhancements to the Deming Management Model

Since the first MBNQA awards ceremony in 1988, few research studies have been
conducted to evaluate the underlying framework of the MBNQA framework until
recently (Flynn & Saladin, 2001). The criteria underlying the awards have evolved
over time. While the names of the categories have remained fairly constant, their
definitions and weights have changed somewhat in the process. In addition, an
award specific to health care organizations was introduced in 1995 (Meyer &
Collier, 2001) and first bestowed in 2002.
Comparing the Deming management model to the general model depicted by
the MBNQA framework, Wilson and Collier (2000) argued that the MBNQA model
is more comprehensive and depicts many more causal paths. Since both models
focus on leadership, process information and management, customers, employees,
and results, the recent studies evaluating the MBNQA may contribute important
insights relevant to the Deming management model. Given the comprehensive
construction of the MBNQA framework (Bell & Keys, 1998), enhancements to the
Deming management model based on these insights add to its generalizability.
Meyer and Collier (2001) and Wilson and Collier (2000) argue that the basic
concept underlying the MBNQA model is that “leadership drives the system that
creates results.” Both of these studies, as well as the one by Flynn and Saladin
(2001), confirmed this proposition. While the Deming management model also
begins with leadership, paths from this construct only lead to internal and exter-
nal cooperation and learning. The conceptual foundation supporting the MBNQA
criteria and the results from the initial MBNQA studies suggest that additional
paths from leadership to other intermediate constructs in the Deming management
model exist. This suggestion is also supported by Anderson et al. (1995) in their
discussion of large unexplained effects in their path model and their suggestion
that the possibility of additional paths emanating from leadership may exist.
The MBNQA research also suggests the potential existence of direct paths
from leadership to business results and customer satisfaction. While the MBNQA
framework contains these direct paths, Wilson and Collier (2000) did not find sig-
nificant direct relationships. However, Meyer and Collier (2001), using a sample of
hospitals, did find a significant direct relationship between leadership and organi-
zational performance. In addition, Flynn and Saladin (2001), using the world-class
manufacturing database, found significant relationships between leadership and
both business results and customer satisfaction. Therefore, there exists some evi-
dence from these studies that leadership may be directly related to performance.
Finally, all three of these studies found significant relationships between
process management and business or organizational results. While the Deming
management model does not contain this direct path, Anderson et al. (1994a)
suggested that a direct path may exist between process management and customer
satisfaction. Given that we are using a different database, we will perform an ad
hoc test of the paths suggested by these research studies.

SAMPLE CONSTRUCTION AND DATA COLLECTION

The study was conducted within the General Medical Hospitals (SIC 8062) indus-
try. TQM has been recommended to the members of this industry as a strategy that
Douglas and Fredendall 405

will assist them in dealing with their turbulent environment (Gaucher & Coffey,
1993). Thus, this context was expected to provide an excellent platform on which
to test the subject model.
Data for the analysis were gathered on both a primary and secondary basis.
In 1996 we randomly selected 19 metropolitan areas across the United States and
sent questionnaires to the CEO and the Director of Quality in each hospital in
those Standard Metropolitan Statistical Areas (SMSA). Finally, we combined the
survey data with secondary information available for the responding hospitals. One
hundred ninety-three out of the 512 hospitals in the sampling frame responded with
at least one questionnaire, resulting in an overall response rate of 38 percent.
In an effort to assess the potential for response bias, comparisons were made
across a number of available variables for both the responding and the nonre-
sponding hospitals for the 19 SMSAs using data from the 1995 American Hospital
Association Survey. The mean results of the nonrespondents did not differ signif-
icantly from the responding firms when comparing assets, number of employees,
profitability, or services offered. As a result, there does not appear to be systematic
response bias in the financial and operating characteristics of the hospitals sampled.
Although the database and a subset of the scales were used in a previously published
study (Douglas & Judge, 2001), the conceptual framework, issues addressed, and
methodology are unique to this study.

METHODS
We chose to use structural equation modeling to estimate the relationships in the
path diagram displayed in Figure 1. LISREL 8.5 was the analytical procedure used
to estimate this model. This technique combines path analysis with multiple regres-
sion analysis (Joreskog & Sorbom, 1993) in a manner that matches the theoretical
model displayed in Figure 1. The chi-square test associated with this model is χ 2 =
2,466.7, with 1,466 degrees of freedom ( p = .001). The fit of the model was tested
using the Comparative Fit Index (CFI) suggested by Bentler (1990). While many
fit indices have been developed, the CFI was recommended in a review and evalua-
tion of such indices by Medsker, Williams, and Holahan (1994). Values of the CFI
should realistically range from 0 to 1, with the values closest to 1 representing the
best fit (Marsh, Balla, & McDonald, 1988). The value of the CFI calculated in this
study was .86, suggesting that the model estimated fits the data sufficiently well.

RESULTS
The results will be presented for each hypothesis. Because multiple scales were
used to represent many of the constructs, we will present the results related to each
measure. Table 3 contains the results of the structural equation model estimated
using LISREL.

Hypothesis 1
Our first hypothesis suggested that visionary leadership was related to higher lev-
els of internal and external cooperation. We used two measures to represent in-
ternal and external cooperation, quality philosophy and supplier cooperation. Top
 406

Table 3: Results from the structural model.

Dependent Variables
Explanatory Qual Supplr TQ Cust TQ Cont Struc Cust Financl JCAHO
Variables Phil Coop Training Driven MBF Methods Imprvmt Explor Satisfctn Perf Score
TMT Involvement .69∗∗∗ .49∗∗∗ .78∗∗∗ .57∗∗∗
Qual Phil .44∗∗∗ .23∗
Supplr Coop .00 .02
TQ Training .05 .14∗∗
Cust Driven .48∗∗∗ .73∗∗∗
MBF .29∗∗∗ .66∗∗∗
TQ Methods .37∗∗∗ .32∗∗
Cont Imprvmt .22 .19 1.30
Struc Explor .50∗∗∗ .32∗∗ .33
R2 .78 .40 .53 .64 .64 .70 .60 .52 .30 .13 .05
∗∗∗ ∗∗ ∗
p < .001; p < .01; p < .05; n = 193.
Evaluating the Deming Management Model
Douglas and Fredendall 407

management team involvement, our measure of visionary leadership, was signif-

icantly related to both quality philosophy (t = 10.80, p < .001) and supplier
involvement (t = 7.59, p < .001). Therefore, Hypothesis 1 is supported.

Hypothesis 2
Our second hypothesis, that visionary leadership is related to higher levels of learn-
ing, was tested using two measures of learning, total quality training and customer-
driven information. Top management team involvement was significantly related
to total quality training (t = 9.02, p < .001) and customer-driven information
(t = 9.18, p < .001), indicating support for Hypothesis 2.

Hypothesis 3
Hypothesis 3 suggested that internal and external cooperation was related to pro-
cess management. In addition to the two measures used for internal and external
cooperation, two measures were also used for process management, management
by fact and total quality methods. Quality philosophy was significantly related to
both management by fact (t = 3.98, p < .001) and total quality methods (t = 2.03,
p < .05). However, our second measure of internal and external cooperation, sup-
plier cooperation, was not related to either of the process management measures.
This indicates moderate support for Hypothesis 3.

Hypothesis 4
The fourth hypothesis suggested that learning was positively related to process
management. Our results indicated that of the two measures representing learn-
ing, only customer-driven information (4.35, p < .001) is significantly related to
management by fact. However, both total quality training (t = 2.15, p < .01) and
customer-driven information (t = 5.69, p < .001) are significantly related to total
quality methods. Thus, we found general support for Hypothesis 4.

Hypothesis 5
Hypothesis 5 suggested that process management was positively related to con-
tinuous improvement. Both management by fact (t = 4.42, p < .001) and total
quality methods (t = 5.70, p < .001) were significantly related to continuous
improvement. These results provide strong support for Hypothesis 5.

Hypothesis 6
The sixth hypothesis stated that process management is positively related to em-
ployee fulfillment. Structural exploration measures employee fulfillment in our
study. Both management by fact (t = 4.37, p < .001) and total quality methods
(t = 2.41, p < .01) were significantly related to employee fulfillment. Thus, support
was found for Hypothesis 6.

Hypothesis 7
Hypothesis 7 suggested that continuous improvement was positively related to
business performance. The Deming management model includes customer satis-
408 Evaluating the Deming Management Model

Table 4: Results from the cash flow margin regression

analysis.
Independent Variables β T
Hospital Beds −.02 −.24
Ownership −.45 −5.47∗∗
Continuous Improvement .17 2.12∗
Adjusted R2 .20
F 11.21∗∗
∗
p < .05; ∗∗ p < .01; n = 122.

faction as a performance variable. We have added financial performance and each

hospital’s JCAHO audit score as additional measures of business performance. In
our study, we did not find significant relationships between continuous improve-
ment and financial performance or customer satisfaction. With respect to the JC-
AHO audit score (t = 1.79, p < .10), marginal significance was found. Therefore,
Hypothesis 7 was generally not supported.
We were able to conduct one additional test of the relationship between
continuous improvement and financial performance. Cash flow margin data were
available archivally for 122 of the 193 hospitals in the sample. We verified the repre-
sentativeness of this subsample by comparing key variables (size, services, JCAHO
score, perceived financial performance) across the remaining hospitals versus those
that dropped out. Cash flow margin is a financial performance variable often used
in empirical studies of hospitals (Douglas & Ryman, 2003). Using the average
cash flow margin for the years 1996–1997, we conducted a regression analysis
using continuous improvement as the independent variable and included the same
control variables mentioned above. The results displayed in Table 4 demonstrate
that continuous improvement is significantly related to cash flow margin (t = 2.12,
p < .05) using the objective data added to our database.

Hypothesis 8
Our final hypothesis suggested that employee fulfillment was positively related to
business performance. We found the structural exploration measure to be signif-
icantly related to both customer satisfaction (t = 2.99, p < .001) and financial
performance (t = 2.49, p < .01). However, it was not significant with respect to
the JCAHO audit score. Thus, Hypothesis 8 is supported in our study for two of
the key measures of business performance.
Finally, neither of the control variables was significant in our study. Neither
size nor ownership type of the hospitals in this database was significantly related
to business performance.

Ad Hoc Tests Suggested by the MBNQA Literature

Results of the ad hoc tests are contained in Table 5. Starting with the Deming man-
agement model results, the LISREL program was used to test the addition of the
paths suggested by the MBNQA studies on a nested basis. Figure 2 overlays the
Douglas and Fredendall 409

Table 5: Ad hoc tests suggested by MBNQA studies.

Critical
Steps df df χ2 χ 2 Value p
Deming Mgt Model 1,466 2,466.73
Step 1: Leadership Paths to 1,462 4 2,435.95 31.23 18.47 <.001
Continuous Improvement,
Process Management, and
Employee Fulfillment
Step 2: Leadership Paths to 1,461 1 2,433.86 2.09 3.84 N.S.
Financial Performance
Customer Satisfaction 1,460 1 2,425.03 8.83 6.63 <.01
Step 3: Process Management 1,458 2 2,422.67 2.36 5.99 N.S.
Paths to
Financial Performance 1,456 2 2,416.71 5.96 5.99 N.S.
Customer Satisfaction

Figure 2: Revised theory of quality management for services.

Organizational Process
System Outcomes

Internal and Continuous

External Improvement
Cooperation

Visionary Process Business

Leadership Management Performance

Learning Employee
Fulfillment

Note: The additional paths suggested by the MBNQA literature that were significant were
included as follows:
Step 1 is represented by
Step 2 is represented by

significant paths onto the conceptual Deming management model. Our first step
was to add the additional paths between leadership and the remaining intermediate
constructs, process management, continuous improvement, and employee fulfill-
ment. This step added four additional paths to the model and resulted in a change
in χ 2 of 31.23, which is significant ( p < .001). This revised model, therefore,
represents a better fit with this database. The second step added paths between
leadership and two of the performance measures, customer satisfaction and finan-
cial performance. Each path was added individually, and only adding the path from
leadership to customer satisfaction resulted in an improved model fit with a change
410 Evaluating the Deming Management Model

in χ 2 of 8.83 ( p < .01). Finally, the third step added paths from process manage-
ment to customer satisfaction and financial performance. Adding these paths did
not result in a better fitting model.

DISCUSSION
The main focus of this research has been a conceptual and empirical evaluation of
quality management in the context of a service industry, initially using the Deming
management model. This model had been previously tested only in manufacturing
industries. In addition, changes to the model, suggested in recent studies focused
on the more general MBNQA framework, were tested and found to be important
enhancements.
This study represents an additional replication of the Deming management
model and contributes to the discussion concerning its generalizability. Rungtu-
sanatham et al. (1998) recommended that structural equation modeling be used
with samples from additional U.S. industries before reaching conclusions regard-
ing the utility of the model. With this in mind, we will compare our findings with
those of the two earlier studies.
Hypotheses 1, 2, 5, and 6 were fully supported. Some support was identified
for Hypotheses 3, 4, and 8 and none for Hypothesis 7 from our LISREL analysis.
These results are similar to those found by the two earlier studies in manufacturing
industries, with a few notable exceptions. First, we used multiple measures to rep-
resent a number of the model concepts. This should allow us to better understand
the relationships central to the model. Second, we found support for the relation-
ship between learning and process management, which each of the earlier studies
failed to do. Third, we added two additional performance variables as suggested by
the MBNQA literature—financial performance and the JCAHO audit score—in
order to better understand the breadth of the outcomes related to implementing
quality management. These additional measures are necessary to fully assess the
effectiveness of TQM (Hackman & Wageman, 1995).
Similarly to the manufacturing studies, we found support for the relationship
between internal and external cooperation and process management. However,
rather than combining both types of cooperation into one measure, we used sep-
arate measures. In our study, the presence of a quality philosophy represented
internal cooperation. We found that this measure was significantly related to pro-
cess management. However, our external measure, supplier cooperation, was not
related to process management. It may be that, in this industry, close cooperation
with suppliers has little relevance to hospital processes. Or, it is possible that the
respondents to the survey were not in a position to evaluate the level of cooperation
with suppliers. The survey respondents were located at individual hospitals, while
many of the supplier contacts were made at a central location in the hospital’s
corporate structure.
In this study, learning was represented with two measures, total quality train-
ing and customer-driven information. We also included two measures of process
management—management by fact and total quality methods. All relationships
were significant except the one between total quality training and management
by fact. Neither Anderson et al. (1995) nor Rungtusanatham et al. (1998) found a
Douglas and Fredendall 411

significant relationship between learning and process management. The significant

relationship between learning and process management is important. While it has
been argued by many that learning is a key concept within quality management
(Anderson et al., 1994a; Hackman & Wageman, 1995; Huq & Martin, 2000), little
supportive evidence has existed.
The path in the model from process management to employee fulfillment was
supported. There were positive relationships between both process management
measures and employee fulfillment measures. Since exploration reflects the ability
of organizational members to use readily available information for decision making,
this significant path indicates that supplying and utilizing quality data is critical to
employee decision making.
The reason that Anderson et al. (1995) and Rungtusanatham et al. (1998) did
not find a significant relationship between process management and employee ful-
fillment may be due to the operationalization of the employee fulfillment variable.
Anderson et al. (1994a, p. 480) defined employee fulfillment as “exemplified by
job satisfaction, job commitment, and pride of workmanship.” However, Anderson
et al. (1995) and Rungtusanatham et al. (1998) used three and two items, respec-
tively, which seemed to measure only pride of workmanship.
The structural exploration items in this study measured the level of informa-
tion sharing and employee adaptation, which are expected to result in job satis-
faction and commitment (Robbins et al., 2002). The lack of support for employee
fulfillment in earlier studies may be attributable to the lack of a metric that cor-
relates well with job satisfaction or job commitment. Future work should explore
more complete operationalizations of this key construct.
With respect to the relationship between employee fulfillment and business
performance, we found significant relationships between structural exploration
and both financial performance and customer satisfaction. Anderson et al. (1995)
also found a strong relationship between employee fulfillment and customer sat-
isfaction. Rungtusanatham et al. (1998) did not, but attributed the problem to the
culture in the Italian manufacturing plants in their study. Our finding with re-
spect to structural exploration suggests additional studies may be needed to better
understand or even revise the role of employee involvement in quality implemen-
tation, especially in service organizations. Higher levels of exploration were re-
lated to higher levels of customer satisfaction and financial performance relative to
competitors.
In the manufacturing studies, Anderson et al. (1995) found no relationship be-
tween continuous improvement and customer satisfaction while Rungtusanatham
et al. (1998) found marginal significance. Our study did not find a significant
relationship between continuous improvement and any of the three performance
measures. This is surprising, because there is strong theoretical support for this re-
lationship. One possible explanation is that the hospitals’ continuous improvement
efforts may be focused on cost control or cost reductions, which would not impact
customer satisfaction directly. Our additional test using the subset of the sample for
which objective financial results were available did find a significant relationship
between continuous improvement and financial performance. This ad hoc finding
suggests that a relationship may exist and that further research is needed to better
investigate this relationship and the relevant constructs.
412 Evaluating the Deming Management Model

In addition, there are two other possible explanations suggested by the lit-
erature for the lack of a significant relationship in this study between continuous
improvement and customer satisfaction. First, neither this study nor the prior two
studies of the Deming management model directly measured customer satisfac-
tion. Second, it is possible that service quality is an independent construct that
is an antecedent to customer satisfaction. Both explanations are involved in the
debate about the appropriate use of instruments such as SERVQUAL to measure
customer satisfaction (Van Dyke, Kappelman, & Prybutok, 1997) and are discussed
in greater detail below.
The ad hoc tests based on the MBNQA literature are an important contribution
to the literature. The additional four paths from leadership to the other variables
shown in Figure 2 significantly increased the model fit and suggest that, as originally
conceptualized, the Deming management model may be too parsimonious. The
significance of these paths does support Deming’s stress on the importance of
leadership to business improvement. These ad hoc tests suggest that leadership has
a direct effect on every subsequent variable in the model, strongly supporting the
underlying concepts associated with the MBNQA model. Because the derivation
of the MBNQA model has a broader base than the initial focus on Deming’s work
taken by Anderson et al. (1994a), the enhancements suggested for the Deming
management model may prove more useful going forward.
While the founders of total quality stressed its application to services, most
service research has ignored many of the elements of the quality management
model. The significant findings of this research are an important contribution to the
quality and service literatures in general and to health services in particular. These
findings suggest that it is important that service quality researchers integrate more
elements of quality management into their research.
In addition, it is important that future research incorporate insights from
the service quality literature into the Deming management model. One of the key
questions for investigation is whether service quality is a distinct variable preceding
customer satisfaction as suggested by Dabholkar, Shepherd, and Thorpe (2000),
or is service quality already contained in the customer satisfaction construct? A
second, related question is what instrument is most appropriate to use to measure
customer satisfaction?
Dabholkar et al. (2000) viewed customer satisfaction as an overall judgment
of service quality and suggested that it be measured separately from service quality.
They defined service quality as a set of dimensions such as reliability and respon-
siveness that preceded customer satisfaction. They found the service quality and
customer satisfaction constructs highly correlated, but distinct in the retail indus-
try. These issues have not yet been examined in great depth outside of the retail
industry.
SERVQUAL is a frequently used instrument measuring customer satisfac-
tion in the service quality literature (Harvey, 1998; Kettinger & Lee, 1997). The
definition of service quality underlying SERVQUAL (Parasuraman, Zeithaml, &
Berry, 1988) is equivalent to the definition of customer satisfaction associated with
the Deming management model that is displayed in Table 1. Therefore, it would be
appropriate to use SERVQUAL to measure whether customers perceive that their
Douglas and Fredendall 413

needs are being met. Many researchers support the continued use of SERVQUAL to
measure customer satisfaction, but recommend that more work be done to improve
its scales (Kettinger & Lee, 1999).
The suggestion by Dabholkar et al. (2000) presented above, that service
quality is conceptually different from customer satisfaction, is based on a different
definition of service quality. According to Voss, Roth, Rosenzweig, Blackmon, and
Chase (2004), service quality, as discussed in the operations management literature,
depends on meeting or exceeding service standards, which are established within
the provider organization. The question of whether a distinct, internal measure of
service quality exists and whether it should be added to the Deming management
model is an important topic for future research.
In summary, two important issues that need to be resolved are whether service
quality, as defined in the operations management literature, should be included in
the Deming management model as a separate construct, and what is the most
appropriate method to use to measure customer satisfaction? Given reliable and
valid measures of service quality, if it is a necessary and distinct construct, and
customer satisfaction, their use in the enhanced Deming management model may
lead to a better understanding of the outcomes of quality management for both
services and manufacturing.

Implications for Managers

The findings from this study should also be useful to managers and administrators
interested in a continued focus on quality improvement. Since TQM/CQI has not
been consistently adopted across hospitals (Westphal et al., 1997) and the prac-
tices underlying these concepts are embedded in more recent programs such as
Reengineering, Six Sigma, and Lean Production, the implications of our findings
may help focus managerial attention on a more complete implementation of the
relevant management model. It is also worthy to note the importance of including
the MBNQA enhancements, since the MBNQA model has been used as a model
for establishing awards in many of the individual states in the United States and
in countries abroad, including the European Quality Award and those in Brazil,
Egypt, Japan, and Mexico (Flynn & Saladin, 2001).
The importance of leadership, emphasized in the findings with respect to
the initial paths to cooperation and learning and the secondary paths suggested by
the MBNQA, should be of paramount importance for hospital administrators and
managers within other service industries. This finding also directly addresses a key
issue raised by Arndt and Bigelow (1995). These authors questioned the ability
of the CEO and other top administrators to adequately provide the needed vision
and guidance for a quality implementation in a hospital environment. Our findings
suggest that these managers can and do have a significant impact on all aspects of
a quality system and need to continue to take an active part to ensure success.
Our study also provides empirical evidence that total quality training and
the availability and use of customer-driven information are closely related to im-
proved process management. This finding supports efforts of hospital managers
to develop a learning culture. As Bender and Krasnick (1995) have pointed out,
414 Evaluating the Deming Management Model

few hospitals exhibit such a culture and yet developing one is important for a total
quality implementation in this industry.
The finding that process management is significantly related to continu-
ous improvement suggests that managers in service industries need to empha-
size the use of basic process management tools such as management by fact
and total quality management methods to enhance their continuous improvement
efforts. As pointed out by Hackman and Wageman (1995), many TQM organi-
zations have focused little attention on using these scientific methods in their
programs.
Overall, managers in service organizations should be aware that each of the
constructs underlying the Deming management model is important. In addition,
leadership from the top of the organization is critically related to attaining higher
levels in each area. Limited managerial attention to the quality system jeopardizes
a successful implementation of TQM.

Limitations
This study shares some of the same limitations as the Anderson et al. (1995)
and Rungtusanatham et al. (1998) studies. The most significant limitation is that
the scales used were not originally developed for the constructs in the Deming
management model. As stated earlier, this may be a partial explanation of why
some paths in the model are not significant or are only weakly supported. However,
where possible we used multiple measures for most constructs in order to broaden
our understanding of the relationships within the model. In addition, the measures
were obtained from the management literature where they had been evaluated for
reliability and validity. It would be desirable for future studies to evaluate the
measures used here, the ones from the manufacturing studies, those associated
with SERVQUAL, and others that may be appropriate to further our understanding
of the Deming management model.
One advantage of this study, that all respondents were from the same industry,
also limits the generalizability of the findings to other service industries. Future
research that applies the enhanced Deming management model to other industries
is needed. As Reeves and Bednar (1994) suggested that a concept as complex as
service quality management can only be understood through cumulative analyses
of multiple industries.
In addition, while the causality implied by the model is supported by the
theory, the study design does not allow us to directly test the proposed causality
because the data were gathered at a point in time. Future studies of a longitudinal
nature would be helpful to address this issue.

CONCLUSIONS
An important contribution of this research is the extension of the Deming manage-
ment model to services. The strong theoretical and empirical support for many of
the relationships in the Deming management model tested within the health care
industry suggests that it is important for researchers to incorporate these variables
Douglas and Fredendall 415

into their research about service quality. The findings here suggest that the Deming
management model is as applicable to services as it is to manufacturing.
An additional contribution relates to the enhancements to the Deming
management model suggested by the development and testing of the MBNQA
framework. The MBNQA criteria were established on a broader basis than the
Deming model and better represent the quality implementations in organizations.
The importance of having visionary leadership directly involved in all facets of
quality programs proved important in this dataset and represents a key message for
managers and scholars alike.
Significant questions remain unanswered with respect to performance mea-
sures within the enhanced model. While conceptual arguments are strong with
respect to hypothesized relationships between continuous improvement and out-
come variables representing customer satisfaction and operational and financial
performance, only marginal evidence was found to support the arguments. The
hope is that the findings here will encourage research integrating the rich service
quality literature into the model. Future research should investigate whether cus-
tomer satisfaction mediates the relationship between service quality and business
performance. As stated earlier this will require research into the distinction be-
tween the service quality and customer satisfaction constructs. It will also require
a determination of whether it is better to measure customer satisfaction by directly
using an instrument such as SERVQUAL.
Quality management continues to be an important research topic, especially
given its embeddedness in current managerial programs such as Six Sigma and
Lean Production. Use of the enhanced Deming management model, combined
with the customer satisfaction work contained in the service quality literature, in
future research may lead to better-informed literatures focusing on both manu-
facturing and service performance. [Received: February 2003. Accepted: January
2004.]

REFERENCES
Anderson, E. W., Fornell, C., & Lehmann, D. R. (1994a). Customer satisfaction,
market share, and profitability: Findings from Sweden. Journal of Marketing,
58(July), 53–66.
Anderson, J., Rungtusanatham, M., & Schroeder, R. (1994b). A theory of qual-
ity management underlying the Deming management method. Academy of
Management Review, 19, 472–509.
Anderson, J., Rungtusanatham, M., Schroeder, R., & Devaraj, S. (1995). A path
analytic model of a theory of quality management underlying the Deming
management method: Preliminary empirical findings. Decision Sciences, 26,
637–658.
Arndt, M., & Bigelow, B. (1995). The implementation of total quality management
in hospitals: How good is the fit? Health Care Management Review, 20(4),
7–14.
416 Evaluating the Deming Management Model

Bell, R., & Keys, B. (1998). A conversation with Curt W. Reimann on the back-
ground and future of the Baldrige award. Organizational Dynamics, 26(4),
51–61.
Bender, A., & Krasnick, C. (1995). Total quality management: Lessons learned
from the medical practice environment. Health Care Supervisor, 13(4), 70–
76.
Bentler, P. (1990). Comparative fit indexes in structural models. Psychological
Bulletin, 107, 237–246.
Boerstler, H., Foster, R., O’Connor, E., O’Brien, J., Shortell, S., Carmen, J., &
Hughes, E. (1996). Implementation of total quality management: Conven-
tional wisdom versus reality. Hospital & Health Services Administration,
41(2), 143–159.
Bureau of Labor Statistics. (2002). Current employment statistics survey (Na-
tional). Washington, DC: U.S. Department of Labor.
Callahan, J., & Moretton, B. (2001). Reducing software product development time.
International Journal of Project Management, 19, 59–70.
Carmen, J., Shortell, S., Foster, R., Hughes, E., Boerstler, H., O’Brien, J., &
O’Connor, E. (1996). Keys for successful implementation of total quality
management in hospitals. Heath Care Management Review, 21(1), 48–60.
Chase, R. B. (1996). The mall is my factory: Reflections of a service junkie.
Production and Operations Management, 5(4), 298–308.
Cook, L. S., & Verma, R. (2002). Exploring the linkages between quality system,
service quality, and performance excellence: Service providers’ perspectives.
Quality Management Journal, 9(2), 44–56.
Dabholkar, P., Shepherd, C. D., & Thorpe, D. I. (2000). A comprehensive frame-
work for service quality: An investigation of critical conceptual and mea-
surement issues through a longitudinal study. Journal of Retailing, 76(2),
139–173.
Deming, W. (1986). Out of the crisis. Cambridge: Massachusetts Institute of Tech-
nology Press.
Douglas, T., & Judge, W. (2001). Total quality management implementation
and competitive advantage: The role of structural control and exploration.
Academy of Management Journal, 44, 158–169.
Douglas, T., & Ryman, J. (2003). Understanding competitive advantage in the
general hospital industry: Evaluating strategic competencies. Strategic Man-
agement Journal, 24, 333–347.
Flynn, B. B., Schroeder, R. G., & Sakakibara, S. (1994). A framework for quality
management research and an associated measurement instrument. Journal of
Operations Management, 11, 339–366.
Flynn, B. B., & Saladin, B. (2001). Further evidence on the validity of the theoretical
models underlying the Baldrige criteria. Journal of Operations Management,
19, 617–652.
Douglas and Fredendall 417

Foster, S. T., Howard, L. W., & Shannon, P. (2002). The role of quality tools in
improving satisfaction with government. Quality Management Journal, 9(3),
20–31.
Gaucher, E., & Coffey, R. (1993). Total quality in healthcare. San Francisco:
Jossey-Bass.
Ghobadian, A., Speller, S., & Jones, M. (1994). Service quality: Concepts and
models. International Journal of Quality and Reliability Management, 11(9),
43–66.
Gittell, J. H. (2002). Relationships between service providers and their impact on
customers. Journal of Service Research, 4(4), 299–311.
Goal/QPC. (1991). The Memory Jogger Plus+ Videotape Series. Methuen, MA:
Goal/QPC with Amatulli & Associates, Inc.
Hackman, J., & Wageman, R. (1995). Total quality management: Empirical, con-
ceptual, and practical issues. Administrative Science Quarterly, 40, 309–342.
Harvey, J. (1998). Service quality: A tutorial. Journal of Operations Management,
16, 583–597.
Hensel, J. S. (1990). Service quality improvement and control: A customer-based
approach. Journal of Business Research, 20, 43–54.
Huq, Z., & Martin, T. (2000). Workforce cultural factors in TQM/CQI implemen-
tation in hospitals. Health Care Management Review, 25(3), 80–103.
Joreskog, K., & Sorbom, D. (1993). LISREL 8: Structural equation modeling with
the SIMPLIS command language. Chicago: Scientific Software International,
Inc.
Kellogg, D. L., & Chase, R. B. (1995). Constructing an empirically derived measure
for customer contact. Management Science, 41(11), 1734–1749.
Ketchen, D., Thomas, J., & Snow, C. (1993). Organizational configurations and
performance: A comparison of theoretical approaches. Academy of Manage-
ment Journal, 36, 1278–1313.
Kettinger, W. J., & Lee, C. C. (1997). Pragmatic perspectives on the measurement
of information systems service quality. MIS Quarterly, 21(2), 223–240.
Kettinger, W. J., & Lee, C. C. (1999). Replication of measures in information
systems research: The case of IS SERVQUAL. Decision Sciences, 30, 893–
899.
Khandwalla, P. (1977). The design of organizations. New York: Harcourt Brace
Jovanovich, Inc.
Kunst, P., & Lemmink, J. (2000). Quality management and business performance
in hospitals: A search for success parameters. Total Quality Management, 11,
1123–1133.
Marsh, H., Balla, J., & McDonald, R. (1988). Goodness-of-fit indexes in confirma-
tory factor analysis: The effect of sample size. Psychological Bulletin, 103,
391–410.
418 Evaluating the Deming Management Model

Medsker, G., Williams, L., & Holahan, P. (1994). A review of current practices
for evaluating causal models in organizational behavior and human resources
management research. Journal of Management, 20, 439–464.
Meyer, S., & Collier, D. (2001). An empirical test of the causal relationships in the
Baldrige health care pilot criteria. Journal of Operations Management, 19,
403–425.
Motwani, J., Sower, V., & Brashier, L. (1996). Implementing TQM in the health
care sector. Health Care Management Review, 21(1), 73–82.
National Institute of Standards and Technology (NIST). (2003). Malcolm Baldrige
National Quality Program Health Care Criteria for Program Excellence.
Gaithersburg, MD.
Nilsson, L., Johnson, M. D., & Gustafsson, A. (2001). The impact of quality prac-
tices on customer satisfaction and business results: Product versus service
organizations. Journal of Quality Management, 6, 5–27.
Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1985). A conceptual model of
service quality and its implications for future research. Journal of Marketing,
49(Fall), 41–50.
Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A multiple-
item scale for measuring consumer perceptions of service quality. Journal of
Retailing, 64(1), 12–37.
Powell, T. (1995). Total quality management as competitive advantage: A review
and empirical study. Strategic Management Journal, 16, 15–37.
Ramanujam, V., Venkatraman, N., & Camillus, J. (1986). Multi-objective assess-
ment of effectiveness of strategic planning. Academy of Management Journal,
29, 347–372.
Reeves, C., & Bednar, D. (1994). Defining quality: Alternatives and implications.
Academy of Management Review, 19, 419–445.
Robbins, T. L., Crino, M. D., & Fredendall, L. D. (2002). An integrative model
of the empowerment process. Human Resource Management Review, 12,
419–433.
Roth, A. V., & Jackson, W. E. III (1995). Strategic determinants of service quality
and performance: Evidence from the banking industry. Management Science,
41(11), 1720–1733.
Routhieaux, R., & Gutek, B. (1998). TQM/CQI effectiveness at team and depart-
ment levels. Journal of Quality Management, 3, 39–62.
Rungtusanatham, M., Forza, C., Filippini, R., & Anderson, J. (1998). A replication
study of a theory of quality management underlying the Deming management
method: Insights from an Italian context. Journal of Operations Management,
17, 77–95.
Rust, R., Zahorik, A. J., & Keiningham, T. L. (1995). Return on quality (ROQ):
Making service quality financially accountable. Journal of Marketing, 59(2),
58–70.
Douglas and Fredendall 419

Saraph, J., Benson, P., & Schroeder, R. (1989). An instrument for measuring
the critical factors of quality management. Decision Sciences, 20, 810–
829.
Shortell, S., O’Brien, J., Carman, J., Foster, R., Hughes, E., Boerstler, H., &
O’Conner, E. (1995). Assessing the impact of continuous quality improve-
ment/total quality management: Concept versus implementation. Health Ser-
vices Research, 30, 377–401.
Shostack, G. L., (1984). Designing services that deliver. Harvard Business Review
(January–February), 133–139.
Sousa, R., & Voss, C. A. (2002). Quality management re-visited: A reflective
review and agenda for future research. Journal of Operations Management,
20, 91–109.
Stewart, D. M., & Chase, R. B. (1999). The impact of human error on deliv-
ering service quality. Production and Operations Management, 8(3), 240–
263.
Sureshchandar, G. S., Rajendran, C., & Anantharaman, R. N. (2001). A conceptual
model for total quality management in service organizations. Total Quality
Management, 12(3), 343–363.
Waldman, D. (1994). The contributions of total quality management to a theory of
work performance. Academy of Management Review, 19, 510–536.
Westphal, J., Gulati, R., & Shortell, S. (1997). Customization or conformity? An
institutional and network perspective on the content and consequences of
TQM adoption. Administrative Science Quarterly, 42, 366–394.
Wilson, D., & Collier, D. (2000). An empirical investigation of the Malcolm
Baldrige National Quality Award Causal Model. Decision Sciences, 31, 361–
383.
Van Dyke, T. P., Kappelman, L. A., & Prybutok, V. R. (1997). Measuring infor-
mation systems service quality: Concerns on the use of the SERVQUAL
questionnaire. MIS Quarterly, 21(2), 195–209.
Voss, C. A., Roth, A. V., Rosenzweig, E. D., Blackmon, K., & Chase, R. B.
(2004). A Tale of Two Countries’ Conservatism, Service Quality, and Feed-
back on Customer Satisfaction. Journal of Service Research, 6(3), 212–
230.
Zabada, C., Rivers, P., & Munchus, G. (1998). Obstacles to the application of total
quality management in health-care organizations. Total Quality Management,
9, 57–66.
Zajac, E., & Shortell, S. (1989). Changing generic strategies: Likelihood, direc-
tion, and performance implications. Strategic Management Journal, 10, 413–
430.
Zeitz, G., Johannesson, R., & Ritchie, J. (1997). An employee survey measuring
total quality management practices and culture: Development and culture.
Group & Organization Management, 22, 414–434.
420 Evaluating the Deming Management Model

APPENDIX
Questionnaire Items Used in This Study‡
TQM Practices: Scale represents extent to which items are practiced in your orga-
nization (1 = very low to 5 = very high)

1. Top Management Team Involvement

• The top health care organization executive assumes responsibility for
quality performance.
• The major department heads participate in the quality improvement
process.
• The organization’s top management (top administrator and major
department heads) has objectives for quality performance.
• The goal-setting process for quality within the health care organization
is comprehensive.
• Importance is attached to quality by the organization’s top management
in relation to cost objectives.
• Quality issues are reviewed in the organization’s top management
meetings.
2. Quality Philosophy
• There is a strong commitment to quality at all levels of this organization.
• People in this organization are aware of its overall mission.
• Members of this organization show concern for the need for quality.
• Continuous quality improvement is an important goal of this
organization.
• Managers here try to plan ahead for changes that might affect our
performance.
3. Total Quality Training
• Quality-related training is given to hourly employees throughout the
organization.
• Quality-related training is given to managers and supervisors throughout
the organization.
• Training is given in the “total quality concept” (i.e., philosophy of
company-wide responsibility for quality) throughout the organization.
• Training is given in the basic statistical techniques (such as histograms
and control charts) in the organization as a whole.
• The organization’s top management is committed to employee training
for quality.
• Resources are provided for employee training in quality.
4. Customer Driven Information
• Associates know who their customers are.
• Associates attempt to measure their internal customers’ needs
(customers inside this organization).
Douglas and Fredendall 421

• Associates attempt to measure their external customers’ needs

(customers outside this organization).
• The organization uses customer requirements as the basis for Quality.
• Our organization is more customer focused than our competitors.
5. Continuous Improvement
• Associates in the organization try to improve the quality of their service.
• Associates in the organization believe that quality improvement is their
responsibility.
• Associates in the organization analyze their work products to look for
ways of doing a better job.
6. Management by Fact
• Quality data (defects, complaints, outcomes, time, satisfaction, etc.) are
available.
• Quality data are timely.
• Quality data are used as tools to manage quality.
• Quality data are available to hourly workers.
• Quality data are available to managers and supervisors.
• Quality data are used to evaluate supervisor and managerial
performance.
7. Total Quality Methods
• Associates use the basic statistical techniques (such as histograms and
control charts) to study their work processes.
• Associates analyze the time it takes to get the job done.
• Associates keep records and charts measuring the quality of work dis-
played in their work area.
• Statistical techniques are used to reduce variation in processes in the
organization.
• TQM procedures (such as brainstorming, cause-and-effect diagrams,
Pareto charts) are used to analyze information for process improvement.
8. Supplier Involvement
• Suppliers are selected based on quality rather than price.
• The organization’s supplier rating system is thorough.
• The organization relies on reasonably few, but dependable suppliers.
• The organization provides education to its suppliers.
• Longer term relationships are offered to suppliers.
• Clear specifications are provided to suppliers.

Perceived Financial Performance: Scale represents the organization’s relative per-

formance to competitors over the last three years (1 = much worse, 5 = much
better).
r Growth in earnings
r Growth in revenue
422 Evaluating the Deming Management Model

r Changes in market share

r Return on assets
r Long-run level of profitability
Perceived Customer Satisfaction: Scale represents the organization’s relative per-
formance to competitors over the last three years (1 = much worse, 5 = much
better).
r Patient satisfaction
r Average length of stay
r Clinical outcomes
Structural Exploration: Scale asks the respondent to identify the operating manage-
ment philosophy actually used in their health care organization. A “1” represents
the expression on the left-hand side and a “7” represents the expression on the
right-hand side, with “4” as the mid-point standing for a combination of the two.

Highly structured channels of 1 2 3 4 5 6 7 Open channels of communication

communication and highly with important financial and
restricted access to operating information flowing
important financial and quite freely throughout the
operating information organization
A strong emphasis on holding 1 2 3 4 5 6 7 A strong emphasis on adapting
fast to true and tried freely to changing
management principles circumstances without too
despite any changes in much concern for past
business conditions practices
‡
A complete copy of the questionnaire can be obtained from the lead author.

Thomas J. Douglas is assistant professor of management at Clemson University.

He received his PhD in strategic management from the University of Tennessee.
He has more than 25 years of industry experience with SBC Communications.
He has published or has forthcoming work in the Academy of Management Jour-
nal, Strategic Management Journal, Journal of Business Venturing, and Interfaces,
among others. His research interests are in the areas of competitive advantage, en-
trepreneurship, total quality management, and sustainable environmental strategies.
He is a member of the Academy of Management and the Strategic Management
Society.

Lawrence D. Fredendall is associate professor of management at Clemson Univer-

sity. He received his MBA and his PhD in operations management from Michigan
State University. His most recent book is titled Basics of Supply Chain Management
and was published by The St. Lucie Press/APICS Series. His work has appeared in
journals such as the Journal of Operations Management, International Journal of
Production Research, European Journal of Operational Research, and Production
and Operations Management.