Foster article 6/18/02 10:01 AM Page 20
The Role of Quality Tools
in Improving Satisfaction
with Government
S. THOMAS FOSTER JR., BOISE STATE UNIVERSITY
LARRY W. HOWARD, MIDDLE TENNESSEE STATE UNIVERSITY
PATRICK SHANNON, BOISE STATE UNIVERSITY
© 2002, ASQ
This article presents the results of a study in a city
in the western United States. The authors found that
city employees believed that quality knowledge was
necessary for improving quality. Results show that
departmental leadership was positively associated
with teamwork, process improvement, and employee
satisfaction. Quality knowledge, if followed up with
application, can be effective in improving processes.
Leadership is necessary to the development of quality
tools knowledge. Therefore, both leadership and teamwork are important contextual variables for quality
improvement in the public sector.
Key words: leadership, organizational context, quality
management, quality tools, teamwork
20 QMJ VOL. 9, NO. 3/© 2002, ASQ
INTRODUCTION
Much has been written about infrastructural and environmental variables in quality improvement in business (Adam 1994; Saraph, Benson, and Schroeder
1989; Flynn, Schroeder, and Sakakibara 1995). Most of
this research has focused on antecedents to outcomes
such as market share, return on investment, customer
satisfaction, and self-reported measures of quality
improvement. Interestingly, there is much disagreement among these research models regarding the variables leading to positive quality outcomes. This has led
some authors to adopt the contingency-based view that
organizational quality improvement can occur in a
variety of ways—depending upon organizational context (Mallick, Ritzman, and Safizadeh 1999).
While there is an established quality management
literature in business, there is relatively little relating to
quality improvement in government. Much of the existing literature is anecdotal (Foster and Viano 1996). As
a result, there is little understanding of the variables
leading to quality improvement in government.
There are significant differences in environmental
variables of business vs. government. A primary difference is the lack of profit in government. W. Edwards
Deming (1986) often alluded to the profit issue as a differentiator resulting in necessarily different choices in
quality improvement methods between government and
business. For example, infrastructural, labor-related
practices differ in government. Employees have more job
security in government than in business. To compensate
for this, government wages often lag the private sector.
Government entities often have a difficult time
Foster article 6/18/02 10:01 AM Page 21
The Role of Quality Tools in Improving Satisfaction with Government
Model and Hypotheses
Development
Figure 1 shows a model of quality improvement that
motivated this research. The structure of the model and
the included variables are based upon the literature,
including works by Saraph, Benson, and Schroeder
(1989), Adam (1994), the Malcolm Baldrige National
Quality Award Criteria for Performance Excellence
(2000), Sematech Quality Maturity Grid (1998), and
other sources. These variables are categorized as context
variables, enabler variables, and outcome variables.
Context variables refer to organizational context (Benson,
Saraph, and Schroeder 1991). Organizational context is
the organization’s state of being at the time quality
improvement occurs. Organizational context includes
external and internal factors surrounding the production
system. Internal context variables include leadership and
company organization, such as the extent teamwork and
collaborative decision-making is used for improvement.
Enabler variables make organizational change possible and are necessary for effective improvement. These
are critical factors that affect quality outcomes. For
example, knowledge is a fundamental enabler that all
employees need to do their jobs. Specifically, knowledge
of quality tools is required before these tools can be
applied. The extent that quality tools are then used to
make improvement affects quality outcomes. This could
include understanding and using statistical process control, basic tools, automation, and supplier involvement
in improvement (Benson, Saraph, and Schroeder 1991).
Figure 1 A priori quality improvement model.
Contextual
Enablers
Q-tools
application
H2
Teamwork
H1b
H1a
Leadership
H3
Q-tools
knowledge
H1c
H1d
Outcomes
H4
Process
improvement
H5
Employee
satisfaction
H6a
H6b
Customer
satisfaction
www.asq.org 21
© 2002, ASQ
identifying the customer. In business, the customer often
ends up owning the product. Some authors have posited
that the customer is the “one who pays the bills” (Evans
and Lindsay 1999). However, who is the customer in government? Is it the taxpayer, the elected leader (for example, executive branch), the legislature (they allocate
resources), or the individuals who directly access government services (such as the licensee to a motor-vehicle
division)? In fact, government entities may have a number of customers who cannot be defined with the simple
internal and external designations.
Although there is limited research in government
quality management, the need for more research is
great. There are a number of reasons for this. First,
demands are increasing for government services, while
budgets are stagnant or decreasing. Therefore, process
simplification is needed to respond to increasing
demands. Second, there is increasing competitive pressure on government service providers as pressure
mounts to privatize government services. Third, leaders
in government have moved to improve and reinvent
government. Finally, government employees are internally motivated to provide service that is on par with
the private sector.
It is not clear, however, that quality practices can be
transferred from the private sector to the public sector.
While basic quality tools are used commonly in industry, research has not demonstrated the efficacy of these
tools in improving government service. In fact, Deming
cautioned against applying modern quality management approaches to government (Deming 1986).
This article presents results from a study performed in a city government. The city in question
had been implementing teams and quality improvement tools over a number of years. Quality tools,
while ubiquitous in the practitioner literature, have
received little attention in research. The primary
research question is, “Were the applications of quality tools effective in improving quality-related outcomes in this city government.” As a result of this
study and analysis, the authors propose a model of
quality tool usage in government. The primary contribution of this article is to examine the role of quality tools in effective implementation of quality
improvement in a government setting.
Foster article 6/18/02 10:01 AM Page 22
The Role of Quality Tools in Improving Satisfaction with Government
Outcome variables represent the desired outcomes
of quality tools application. Outcomes often mentioned
in the literature include process improvement, employee satisfaction, and customer satisfaction. The following paragraphs address the relationships between context variables, enabler variables, and outcome variables
as described in the quality improvement model shown
in Figure 1.
Contextual Variables
Leadership. Leadership is generally regarded as
essential for quality improvement. Leadership provides
the foundation for improvement, as leaders hold both
the positional and monetary authority to oversee
improvement. In a case study of the Office of
Administrative Services, Department of the Interior,
Keck (1996) found leadership to be necessary for successfully completing process improvement projects.
Scully (1993) stated that leadership is needed to initiate
the process of change in government. Rago (1996)
developed a deductive leadership model of government
improvement with leadership enacting purpose, coordination, communication, and empowerment.
Leadership promotes the implementation of teamwork
by providing required resources and assets, and by symbolically communicating top-level commitment to
quality tools application. Leadership is considered in
the literature to be an antecedent to process improvement (Deming 1986). Also, positive leadership is associated with employee satisfaction (Howard and Foster
1999). By inference, perceptions of leadership commitment to quality should also influence the satisfaction of
those affected by satisfied employees and improved
processes—customers.
Hypothesis 1a: There is a positive relationship
between leadership and teamwork.
Hypothesis 1b: There is a positive relationship
between leadership and process improvement.
Hypothesis 1c: There is a positive relationship
between leadership and employee satisfaction.
Hypothesis 1d: There is a positive relationship
between leadership and customer satisfaction.
22 QMJ VOL. 9, NO. 3/© 2002, ASQ
Teamwork. The second contextual variable is
teamwork. As with the manufacturing and services sectors of the private sector, teams have been widely adopted
in government. There are several reasons for this. One of
the main reasons is complexity in the workplace
(Wenger and Snyder 2000). Given the large volumes of
data available to managers, unilateral decision-making
is less effective. Also, businesses are transforming from
“command and control” to collaboration. Collaboration
is needed as complexity drives workers from performing
manual work to knowledge work or work that involves
the development and transmission of knowledge and
information. Knowledge work implies a greater amount
of ambiguity, searching, researching, and on-the-job
learning. As a result, organizations are using teams
more frequently in their normal operations and in their
problem-solving and process improvement efforts. For
the authors’ purposes, a team is defined as a finite number of individuals who are united in a common purpose.
Selander and Cross (1999) view the team component as
essential for business process redesign.
Enabler Variables
Quality tools knowledge. The first enabler variable is
quality tools knowledge. Before quality tools are applied,
training is often provided so employees learn what quality
tools are available and how to use them. The quality tools
referred to in this research include the basic seven tools of
quality (that is, flowcharts, control charts, histograms,
scatter plots, Ishikawa diagrams, run charts, Pareto
charts, and checksheets) and selected advanced tools
(affinity diagrams, surveys). Ceridwen (1992) identified
flowcharting, Ishikawa diagrams, control charts, and
scatter diagrams as the most useful tools for quality
improvement. Foster and Viano (1996) demonstrated how
basic quality tools were used in the Internal Revenue
Service to improve service quality. As teams begin to work
on process improvement, they have more opportunity to
apply quality management tools and to use teamwork to
solve problems. The more the team works together, knowledge of how and when to apply the quality tools is reinforced. Working in teams increases the value of sharing
quality knowledge. It is expected that as people work in
teams, they are more likely to be facilitated by other team
Foster article 6/18/02 10:01 AM Page 23
The Role of Quality Tools in Improving Satisfaction with Government
members in learning about quality tools. Also, some of the
quality tools, such as brainstorming, are specifically
designed to be used in team settings, so the more people
work in teams, the more likely it is they are going to
become familiar with these types of quality tools.
Hypothesis 2: There is a positive relationship
between teamwork and quality tools knowledge.
Quality tools application. Quality tools application refers to the continued use of quality tools after the
training is completed. Knowledge alone will be inadequate for process improvement unless employees actually
apply their knowledge by properly using quality tools. A
long-term commitment is required to improve quality—
one that will result in a change to a culture of improvement. Foster and Franz (1998) proposed the use of quality tools to improve product quality. They also stated that a
method was needed for further expanding the use of these
tools. This required both a method for understanding the
effects of the quality tools and a means for selecting
appropriate quality tools. The more knowledgeable
employees are about quality tools, the more likely they
are to appropriately select and apply those tools.
Hypothesis 3: Quality tools knowledge is positively related to quality tools application.
Outcome Variables
Process improvement. Process improvement refers
to the extent to which employees and customers perceive
that processes have improved. The importance of process
improvement has long been emphasized by quality proponents (Deming 1986). Process improvement occurs in
a variety of ways, including process redesign, process
simplification, and process elimination. Since Deming
and others have focused on processes and their role as
part of the system, process improvement has received
increased attention by decision makers. Most quality
tools can be used to improve processes. Tools are used for
documenting processes, gathering data about the
processes, and proposing, implementing, and evaluating
improvements to the processes.
Hypothesis 4a: Quality tools application is
positively related to process improvement.
Process improvement has long been cited as a
probable source of employee satisfaction, and quality
improvement has also been shown empirically to be
associated with employee satisfaction (Adam and
Foster 2001). In a study of federal employees, Yuan
(1997) found that organizational characteristics were
significantly related to employee satisfaction. Since
process and quality improvement efforts increase organizational commitment and communication, it is
expected that employee satisfaction will be improved.
Quality and process improvement can be used as a
career anchor (Leavitt 1996) leading to improved
employee satisfaction. Since quality improvement
leads to empowerment of employees and a leveling of
job responsibilities in government organizations,
employees are more satisfied (Stepina and Perrewe
1991). Both the practitioner literature and the authors’
own experience indicate that process improvements
are associated with employee satisfaction.
Hypothesis 5: There is a positive relationship
between process improvement and employee
satisfaction.
Finally, process improvements should be associated
with improvements in customer satisfaction. The focus
on quality improvement in both for-profit industries
and government has been monitored for the last several years using the American Customer Satisfaction
Index (ACSI) (Fornell 1996). Wipper (1994) found a
relationship between organizational improvement and
customer satisfaction through a performance measurement effort at the Oregon Department of
Transportation. Process improvement also promotes a
sense of competence, achievement, and meaning
among employees in the workplace, contributing to
employee job satisfaction. In turn, satisfied employees
are able and predisposed to provide good customer
service (George 1998).
Hypothesis 6a: There is a positive relationship
between employee satisfaction and customer
satisfaction.
Hypothesis 6b: There is a positive relationship
between process improvement and customer
satisfaction.
www.asq.org 23
Foster article 6/18/02 10:01 AM Page 24
The Role of Quality Tools in Improving Satisfaction with Government
The data for this study were drawn from a general attitude survey of employees working for a municipality in
the northwestern United States. The municipal structure
included 11 departmental units: airport, community
development, customer support, fire, legal, library,
mayor’s office, parks, police, public works, and traffic
court. Although all respondents worked within a single
municipality, it should be noted that substantial employee and situational diversity existed. For instance, respondents’ educational attainment ranged from secondary
students to holders of doctoral degrees. Job requirements
also varied widely including technical, nontechnical,
clerical, managerial personnel, and elected and appointed officials. Departments differed significantly, too.
Police and fire personnel were unionized, while other
city employees were not. Some departments operated in
the downtown administrative offices, while others were
located at various field sites. Departments also received
funding from various sources, including federal, state,
and local taxes, and from user fees. The city’s authority
structure is decentralized at the department level. Entire
departments operated as teams, similar to other municipal systems using team structures (for example, Coates
and Miller 1995; Magee 1997). While this study is from
one city, and is thus a limited scope design, the diversity
within and between departments is believed to be sufficient for theory development (Eisenhardt 1989).
The researchers delivered surveys to one coordinator
in each department, who then distributed the surveys to
all employees in their respective departments. Each coordinator collected the completed surveys and returned
them to the researchers for tabulation. Of the city’s 1205
employees, 659 (55.3 percent) full-time employees participated, representing all 11 departmental units of
municipal government. Response rates by department
ranged from 37.8 percent in the mayor’s office to 91.9
percent in the legal department. Nearly one-third (32.1
percent) of the employees held bachelor’s degrees, while
11.3 percent also held graduate college degrees. Table 1
summarizes some of the other key demographic characteristics of the respondents. The distributions for of the
respondents’ genders, ages, tenure, and education levels
matched almost exactly the citywide employee statistics
24 QMJ VOL. 9, NO. 3/© 2002, ASQ
Table 1 Key demographic characteristics
of respondents*.
Age group
Percentage
under 30
11.90%
30-39
26.60%
40-59
47.70%
over 59
13.80
Years with the city
Percentage
less than 1 year
8.40%
1-5 years
33.20%
6-10 years
21%
more than 10 years
32.70
*Totals may not add up to 100 percent due to some
employees not answering certain questions.
provided by the city’s human resources, thereby reducing
expectation of nonresponse bias.
The research relies on self-report measures. The
questionnaire included several survey items that were
not part of this research but were of interest to the city’s
managers. Some of those items were drawn from
employee surveys and training workshops administered
previously. Thirty-nine survey items specific to this
research were developed by the authors based on the
contextual, enabler, and outcome variables included in
the a priori model shown in Figure 1. These items and
the survey format were pretested for face and content
validity using a group of 12 employees and managers.
Feedback resulted in some changes, and the revised
surveys were then further pretested with a second group
of 20 employees and managers, and a consensus was
reached regarding content validity. All measures were
Likert-type scales, using the summated average
of selected items and scored on a range of 1 to 5, with
1 = “strongly disagree” to 5 = “strongly agree.”
Five items measuring teamwork were examined
with confirmatory factor analysis in an earlier study
(see Howard, Foster, and Shannon 2000). The teamwork scale included five items pertaining to employees
working together and participating on team projects
and on process improvement teams, the extent that
© 2002, ASQ
Procedures and Methods
Foster article 6/18/02 10:01 AM Page 25
The Role of Quality Tools in Improving Satisfaction with Government
teams were used in respective departments, and perceived team success. The five items reflected the teamwork variable quite well. While the chi-square statistic,
which is vulnerable to large sample sizes, was significant (χ2 = 30.16, df = 6, p < .001), other fit indices
confirmed the teamwork factor structure (that is,
normed fit index (NFI) = 1.00, comparative fit index
(CFI) = 1.00, Tucker-Lewis index (TLI) = 0.99, root
mean square resides (RMSEA) < .08). Cronbach’s coefficient alpha, an indicator of internal consistency and
reliability, was α = .81.
The authors submitted the remaining 34 items to
exploratory factor analyses, and rotated the six principle factors to an orthogonal solution. The resulting factor pattern is presented in the appendix. The items are
assigned to the factor for which the factor loading is
shown in bold type. They retained only those items with
factor loadings exceeding .60 on their target factor and
with factor loadings less than .40 on any other factor,
with two exceptions (items 16 and 34). Both of these
items loaded substantially higher on their intended factors than on secondary factors and both items
improved the reliabilities of their respective scales. In
addition, the authors wanted to maintain a minimum
of three items per scale for purposes of justifying the
structural equation model without imposing theoretically irrelevant constraints. Item 34 represented the
third item in its respective scale.
Items 17, 18, and 19 were dropped for cross-loading
on multiple factors. The remaining items were used to
create scales for six variables, three independent variables, and three dependent variables. Brief descriptions
of all measures follow.
Independent variables. The authors collected
data to measure three independent variables (in addition
to teamwork): leadership, quality tools application, and
quality tools knowledge. The leadership scale corresponds to factor 1 in the appendix and consists of a
weighted average of items 1-12 and includes items
reflecting the extent to which leaders listen to ideas, are
long-term oriented, and take an active role in quality
improvement. The coefficient alpha for this measure was
α = .97. Factor 2 corresponds to the variable, quality
tools application, and relates to using a formal process to
determine root causes, measuring and monitoring
quality tools application, and estimating the extent to
which teammates use quality tools. Four items (13-16)
produced a coefficient alpha of α = .84 for this scale.
The quality tools knowledge scale (factor 3) includes six
items (items 20-25) and reflects respondents’ understanding of various quality tools, such as structured
brainstorming, unstructured brainstorming, statistics,
team building, surveys, and flowcharts. Coefficient alpha
was α = .89.
Dependent variables. Three dependent variables are examined: perceived customer satisfaction,
employee satisfaction, and process improvement.
Perceived customer satisfaction is reflected in items
26-28 on factor 4. These items include comparing the
customer satisfaction of one’s own department to that of
other departments, overall customer satisfaction, and
pride in the department’s ability to satisfy customers.
The coefficient alpha for these three items was α = .82.
Items 29-31 under factor 5 form the variable,
employee satisfaction, and reflect self-ratings of
improvement in satisfaction over the prior two years,
feeling of importance to the city, and pride in being a
member of the city government. Coefficient alpha was
α = .79.
Process improvement (factor 6) consists of a
weighted average of items 32-34 relating to whether
work is performed better than it was two years ago,
whether customer response is better than two years ago,
and whether overall customer service is better than two
years ago. This approach is consistent with prior
research in quality management (Adam 1993). Since
continuous improvement methods result in gradual
improvement, it takes time for customer satisfaction
levels to improve (Narasimham, Ghosh, and Mendez
1994). These reflect self-assessments of the efficacy of
process improvement efforts in achieving positive
results. Coefficient alpha was α = .83.
It is important to note that while some studies have
found that employee perceptions of the predictors of
team performance are often not the factors that predict
actual team performance (Gladstein 1984), other studies have found that measures of perceived performance
outcomes correlate positively at moderate-to-strong levels with objective measures of performance (Delaney
and Huselid 1996).
www.asq.org 25
Foster article 6/18/02 10:01 AM Page 26
The Role of Quality Tools in Improving Satisfaction with Government
Results
Table 2 shows descriptive statistics and bivariate correlations for each of the independent and dependent variables. The authors examined hypotheses with correlation coefficients, and then submitted the overall model
to a structural equation path model, using SPSS-AMOS.
The hypothesis test results are discussed in the following paragraphs.
Hypotheses 1a, 1b, 1c, and 1d. These
hypotheses involve the relationships between leadership
and teamwork (1a), process improvement (1b),
employee satisfaction (1c), and customer satisfaction
(1d). As shown in Table 2, the correlations between the
leadership composite variable and the other variables
are significant at r = .62, r = .49, and r = .51, respectively (all p < .0001). This supports the supposition
that leadership is an important antecedent to teamwork, process improvement, employee satisfaction, and
customer satisfaction in government services.
Hypothesis 2. The correlation between teamwork and quality knowledge is significant and positive
at r = .31 (p < .0001) (see Table 1). This result supports the hypothesis predicting such a relationship.
Hypothesis 3. Hypothesis 3 relates to the relationship between understanding and applying quality tools
knowledge. As shown in Table 2, the correlation between
these two variables is r = .26 and statistically significant
(p < .0001). While this relationship may seem obvious,
the quality tools knowledge gained in this city government was garnered through a structured, long-term
training program. This shows that such an approach is
correlated with the application of quality tools.
Hypotheses 4a, 4b, and 4c. These hypotheses
apply to the relationships between quality tools application and the dependent variables of process
improvement (4a), employee satisfaction (4b), and
customer satisfaction (4c). As reported in Table 2,
these relationships are all significant (all p < .0001),
r = .49, r = .41, and r = .56, respectively. These
results show a positive association between the use of
quality tools and desired quality outcomes, as predicted. While these relationships have been assumed in
much of the quality literature, they had not been previously tested in a public-sector setting.
Hypothesis 5a. The correlation between process
improvement and employee satisfaction is positive at
r = .58 and significant (p < .0001). This supports the
hypothesis that process improvement and employee
satisfaction are linked.
Hypotheses 6a and 6b. The correlations
between customer satisfaction and employee satisfaction
Table 2 Descriptive statistics and correlations for all variables.
N
Mean
S.D.
Leadership
Quality tools
knowledge
Quality tools
application
Teamwork
Process
improvement
Leadership
632
3.36
1.08
(97)
Quality tools
knowledge
659
2.72
0.92
32
(89)
Quality tools
application
628
2.75
0.95
43
26
(84)
Teamwork
662
3.15
0.87
62
31
64
(81)
Process
improvement
581
3.38
0.97
49
23
49
57
(83)
Employee
satisfaction
656
3.57
0.87
51
19
41
62
58
(79)
Customer
satisfaction
663
3.51
0.77
58
20
56
64
58
57
Note: Decimals omitted; numbers on diagonal in parentheses are coefficient alphas; all correlations are significant at p < .001.
26 QMJ VOL. 9, NO. 3/© 2002, ASQ
Employee
satisfaction
Customer
satisfaction
(82)
© 2002, ASQ
Variable
Foster article 6/18/02 10:01 AM Page 27
The Role of Quality Tools in Improving Satisfaction with Government
Figure 2 Post hoc model of quality improvement
in government.
Contextual
Enablers
Q-tools
application
.18
.19
Teamwork
.32
.59
Leadership
Outcomes
.31
.06
Q-tools
knowledge
.13
.61
.21
Process
improvement
.27
.37
Employee
satisfaction
.12
.22
Customer
satisfaction
Note: All coefficients are standardized and significant at
p < .05 or greater; paths not hypothesized are dotted.
quite well. This post hoc model is illustrated in Figure 2,
along with all standardized parameter estimates. The
chi-square statistic, though still significant, was also significantly smaller at χ2 = 65.50, df = 8, p < .001. Other
fit indices were as follows: GFI = .97; AGFI = .91; NFI =
.96; CFI = .96; TLI = .91; RMSEA = .10. In light of the
fact that measures of all variables were taken from the
same sources, presenting the possibility of common
method bias, the authors conducted a second analysis
following the guidelines of Hofmann and Stetzer (1996).
That is, they randomly split their sample into two halves
and used one half to estimate measures for the outcome
variables and the other half to estimate measures for the
contextual and enabler variables. Subsequently, they fit
these modified data to the post hoc model. These data fit
the model nominally better, although direct comparisons are not possible with no difference in degrees of
freedom: χ2 = 21.16, df = 8, p < .05, GFI = .99, AGFI =
.96, NFI = .98, CFI = .99, TLI = .97, and RMSEA = .05.
While this technique compromises the variability in the
data, it also suggests that common method bias was
probably not a serious problem. These results suggest
that the a priori model was under-specified, but with the
addition of three more parameters was a reasonable
approximation to the empirical data.
In summary, the authors found support for their
hypotheses. They found positive correlations between
leadership and teamwork, process improvement,
employee satisfaction, and customer satisfaction, as
www.asq.org 27
© 2002, ASQ
(6a) and process improvement (6b) were positive at
r = .57 and r = .58, respectively and significant (both
p < .0001). While these results are intuitively satisfying,
they should be interpreted cautiously, as all three variables are based on self-reported measures.
Structural equation model. The authors
examined the efficacy of the overall model by submitting the relationships specified in Figure 1 to a structural equation analysis. While no result would provide
definitive proof of a predicted pattern of relationships,
because alternative specifications might explain the
data as well as theirs, this provides a more rigorous test
of the hypotheses as a set. Because SPSS-AMOS will not
produce modification indices when processing structural equation models with missing data, the authors substituted mean scores for items missing data prior to
their aggregation as summated means scales.
The fit indices for the proposed model illustrated by
Figure 1 indicate a poor fit to the data. The chi-square
statistic was significant (χ2 = 361.73, df = 9, p < .001).
Other fit indices are: goodness of fit index (GFI) = .89;
adjusted goodness of fit index (AGFI) = .66; NFI = .78;
CFI = .78; TLI = .50; RMSEA = .24. Modification indexes produced along with the output, however, suggested
that the problem was not that the parameters the
authors proposed were improper, but rather that there
were additional parameters they did not propose that
needed to be accounted for. In particular, they added
direct relationships from the teamwork variable to
employee satisfaction and to quality tools application,
and from the leadership variable to quality tools knowledge. (In retrospect, it would have been consistent with
the authors’ a priori model if the authors had proposed
direct relationships between both contextual variables –
leadership and teamwork – and both enabler variables –
quality tools knowledge and quality tools application. In
addition, the authors could have anticipated a direct
relationship between teamwork and employee satisfaction, since studies have previously reported that employees working in teams were more satisfied with their jobs
than employees in the same firms who were not working
in teams (Kirkman and Rosen 1999). Had they done so,
their a priori model would have been identical to the
post hoc model the authors report here.) Subsequent to
adding these three parameters, the model fit the data
Foster article 6/18/02 10:01 AM Page 28
The Role of Quality Tools in Improving Satisfaction with Government
predicted in hypothesis 1. They found positive relationships between teamwork and quality tools knowledge,
and between quality tools knowledge and quality tools
application, as predicted in hypotheses 2 and 3, respectively. As predicted by hypothesis 4, the authors also
found positive relationships between quality tools
application and all three outcome variables, process
improvement, employee satisfaction, and perceived
customer satisfaction. They also found direct relationships between process improvement and employee satisfaction, as predicted by hypothesis 5. Finally, the
authors found positive relationships between employee
satisfaction and customer satisfaction, and between
process improvement and customer satisfaction, as predicted by hypothesis 6. In addition, the structural equation analyses indicated that there were significant
direct relationships between teamwork and quality
tools application and employee satisfaction, and
between leadership and quality tools knowledge, relationships the authors did not predict.
Discussion and Conclusions
This article presents a study of quality improvement in a
city government setting. The research shows that for this
city government, employees believed that quality knowledge was necessary for improving quality. The results
showed that departmental leadership was positively associated with teamwork, process improvement, and
employee satisfaction. Quality knowledge, if followed up
with application, can be effective in improving processes.
These improvements, with teamwork, led to improved
employee satisfaction and customer satisfaction.
From a managerial perspective, the authors find
that for quality tools training to be effective, it should be
followed up by application through team processes.
Leadership is critical to the development of quality tools
knowledge, but teamwork is the vehicle through which
this knowledge is translated into application. Both leadership and teamwork, therefore, are important contextual concerns for quality management in the public
sector. The findings associated with improved employee
satisfaction are important for government agencies
since budget limitations often require nonmonetary
approaches to improve morale. This also suggests that
28 QMJ VOL. 9, NO. 3/© 2002, ASQ
government workers are much like private-sector workers in that they want to perform work effectively and
they feel satisfaction when they achieve positive results.
They also perceive that they are serving the public better
as a result of process improvement.
Since these results were gathered within a single city
government at one point in time, the normal caveats
relative to case studies and cross-sectional data apply.
Care should be taken in generalizing these results, as
they could have been affected by some unique aspect of
this city or its context. The authors’ measures were also
original and lacking validation evidence, and composed entirely of employee self-reports, therefore subject
to common method bias. At the same time as the survey data collection, however, a series of focus groups
was conducted in each of the 11 departments. The
results of these focus group sessions generally validated
the survey results. In addition, results of the exploratory
factor analyses offer support for the independence, convergent validity, and discriminant validity of most
scales, while the split-sample structural equation
analysis suggests that common source bias may not
have been a severe problem. Finally, although the
authors cannot eliminate limitations to the generalizability of their results, they believe that the departmentteam structure of the municipal government organization in this study is fundamentally similar to other
team-based municipal government organizations, and
there were no significant historical events that might
set this city apart from others. It should be noted that
this study was conducted in a city government with an
established quality management program, which is not
always the case. Larger sample studies from a large
group of government agencies are called for to further
validate these findings.
REFERENCES
Adam, E.E. Jr., 1994. Alternative quality improvement practices
and organizational performance. Journal of Operations
Management 12, no. 1: 27-44.
Adam, E. E. Jr., and S. T. Foster Jr. 2001. Quality improvement
approach and performance: Multi-site analysis within a firm.
Journal of Quality Management 5, no. 1: 1-16.
Benson, G., J. Saraph, and R. Schroeder. 1991. The effects of
organizational context on quality management. Journal of
Operations Management 37, no. 9: 1107-1124.
Foster article 6/18/02 10:01 AM Page 29
The Role of Quality Tools in Improving Satisfaction with Government
Ceridwen, J. 1992. Using quality’s tools: What’s working well? The
Journal of Quality and Participation 15, no. 2: 92-98.
Coates, D., and M. Miller. 1995. Self-directed teams: Lessons from
local government. Public Management 77, no. 12: 16-21.
Delaney, J. T., and M. A. Huselid. 1996. The impact of human
resource management practices on perceptions of organizational
performance. Academy of Management Journal 39, no. 4: 949-969.
Deming, W. E. 1986. Out of the crisis. Cambridge, Mass.:
Massachusetts Institute of Technology.
Eisenhardt, K. M. 1989. Building theories from case study research.
Academy of Management Review 14, no. 4: 532-550.
Leavitt, W. M. 1996. High pay and low morale: Can high pay,
excellent benefits, job security, and low job satisfaction coexist in a
public agency? Public Personnel Management 25: 333-341.
Magee, Y. S. 1997. Teams: Avoiding the pitfalls. Public
Management 79, no. 7: 26-28.
Malcolm Baldrige National Quality Award Criteria for Performance
Excellence. 2000. Gaithersburg, M. D.: National Institute of
Standards and Technology.
Mallick, D., L. Ritzman, and M. Safizadeh. 1999. A contingency
approach to quality management. In Proceedings of the 1999
National Conference of the Decision Sciences Institute 3: 1208-1210.
Evans, J., and W. Lindsay. 1999. The management and control of
quality. Cincinnati, Ohio: Southwestern.
Narasimhan, R., S. Ghosh, and D. Mendez. 1993. A dynamic
model of product quality and pricing decisions on sales response.
Decision Sciences 24, no. 5: 893-908.
Flynn, B., R. Schroeder, and S. Sakakibara. 1995. The impact of
quality management practices on performance and competitive
advantage. Decision Sciences 26, no. 5: 659-692.
Rago, W. V. 1996. Struggles in transformation: A study in TQM,
leadership, and organizational culture in a government agency.
Public Administration Review 56: 227-234.
Fornell, C. 1996. The American customer satisfaction index: Nature,
purpose, and findings. Journal of Marketing 60, no. 4: 7-19.
Saraph, J., G. Benson, and R. Schroeder. 1989. An instrument for
measuring the critical factors of quality management. Decision
Sciences 20, no. 4: 810-829.
Foster, S. T. Jr., and C. R. Franz. 1998. On the differences between
information systems users and analysts: Managing Perceptions of
Systems Quality. Journal of Quality Management 3, no. 1: 63-77.
Foster, S.T. Jr., and R. Viano. 1996. Using quality management to
improve customer responsiveness at the internal revenue service.
Production and Inventory Management Journal 37, no. 2: 37-43.
George, J. M. 1998. Salesperson mood at work: Implications for
helping customers. The Journal of Personal Selling 18, no. 3: 23-30.
Gladstein, D. L. 1984. Groups in context: A model of task-group
effectiveness. Administrative Science Quarterly 29: 499-516.
Hofmann, D. A., and A. Stetzer. 1996. A cross-level investigation of
factors influencing unsafe behavior and accidents. Personnel
Psychology 49: 307-339.
Howard, L. W., and S. T. Foster Jr. 1999. The influence of human
resource practices on empowerment and employee perceptions of
management commitment to quality. Journal of Quality
Management 4, no. 1: 5-22.
Howard, L. W., S. T. Foster, and P. Shannon. 2000. Team climate
and teamwork in government: The power of embedded leadership.
In Proceedings of the 4th International Workshop on Teamworking
(IWOT4): 83-119.
Keck, M. E. 1996. Total quality management teams in the office of
administrative services, U. S. department of the interior: A success
story. International Journal of Public Administration 19: 1811-1844.
Kirkman, B. L., and B. Rosen. 1999. Beyond self-management:
Antecedents and consequences of team empowerment. Academy of
Management Journal 42, no. 1: 58-74.
Scully, J. P. 1993. How to really change the federal government.
National Productivity Review 13, no. 1: 29-37.
Selander, J. P., and K. Cross. 1999. Process redesign: Is it worth it?
Government Finance Review 15, no. 4: 23-27.
Sematech Quality Maturity Grid. 1998. Austin, Texas: Sematech.
Stepina, L. P., and P. L. Perrewe. 1991. The stability of comparative
referent choice and feelings of inequity: A longitudinal field study.
Journal of Organizational Behavior 12: 186-200.
Wenger, E. C., and W. M. Snyder. 2000. Communities of practice:
The organizational frontier. Harvard Business Review 78, no. 1:
139-150.
Wipper, L. 1994. Oregon department of transportation steers
improvement with performance measurement. National Productivity
Review 13, 3: 359-368.
Yuan, T. 1997. Determinants of job satisfaction of federal government employees. Public Personnel Management 26: 313-334.
BIOGRAPHIES
S. Thomas Foster is a professor of quality and operations management at Boise State University. He has a doctorate from the
University of Missouri-Columbia. He has been published in journals such as Decision Sciences, International Journal of
Production Research, Journal of Quality Management,
International Journal of Quality and Reliability Management,
Quality Management Journal, and Quality Progress. Foster has
consulted for a number of companies including Hewlett-Packard,
www.asq.org 29
Foster article 6/18/02 10:01 AM Page 30
The Role of Quality Tools in Improving Satisfaction with Government
Trus Joist Macmillan, Cutler-Hammer/Eaton Corp., Heinz Frozen
Foods, Qwest Corporation, Healthwise Corporation, and the
U. S. Department of Energy. Foster served on the 1996 and
1997 board of examiners for the Malcolm Baldrige National
Quality Award. He is the author of Quality Management: An
Integrative Approach. Foster is founder of www.freequality.org
and was awarded the ASBSU 2000 Outstanding Faculty Award.
He can be reached by e-mail at foster@boisestate.edu .
Larry W. Howard is an assistant professor in the management
and marketing department of Middle Tennessee State University’s
Jennings A. Jones College of Business. He received his doctorate
in business administration from the University of Missouri and his
bachelor’s and master’s degrees from Western Michigan
University. Prior to pursuing doctoral studies, Howard was a general manager with two Fortune 100 companies for six years, and
a full-time management consultant for three years. He still consults on occasion with public and private organizations around
the world in team building, organizational change and development, and managing organizational justice. Recently, he has
been involved in a federal government initiative examining public
policy implications of management practices for 21st century
leadership and governance. Howard has presented his research
at professional conferences and has published book chapters and
articles in journals such as Academy of Management Journal,
Journal of Business and Psychology, Journal of Quality
Management, Journal of Education for Business, International
Journal of Organizational Analysis, and others. He can be
reached by e-mail at lhoward@mtsu.edu .
Patrick W. Shannon is a professor of operations management
and department chair of the networking, operations, and information systems department in the College of Business at Boise
State University. He teaches graduate and undergraduate courses
in business statistics, quality management, and production and
operations management, and has received several alumni teaching awards. In addition, Shannon has lectured and consulted in
the statistics, operations management, and quality areas for more
than 20 years. Among his consulting clients are Boise Cascade
Corporation, Hewlett-Packard, PowerBar Inc., Potlatch
Corporation, Woodgrain Millwork Inc., J. R. Simplot Company,
and others. Shannon has coauthored several university-level textbooks including Business Statistics: A Decision Making
Approach, 5th edition; A Course in Business Statistics, 3rd edition; and Introduction to Management Science. He has also published ar ticles in such journals as Business Horizons,
Transpor tation Research Record, Inter faces, Journal of
Simulation, Journal of Production and Inventory Control, Quality
Progress, and Journal of Marketing Research. Shannon has his
bachelor’s and master’s degrees from the University of Montana
and his doctorate from the University of Oregon. He can be
reached by e-mail at pshannon@boisestate.edu .
APPENDIX
Factor Pattern after Varimax Rotation1.
Factor2
1 Dept. head effectively communicates with me.
84
2 Dept. head is willing to change.
83
3 Dept. head takes active role in quality improvement.
83
4 Dept. head inspires employee trust.
83
5 Dept. head is long-term oriented.
81
6 Dept. head has good knowledge of quality concepts.
81
7 Dept. head respects me.
80
8 Dept. head listens to my ideas.
79
9 Dept. head believes in continuous improvement.
78
10 Dept. head is involved in quality planning.
77
11 Dept. head supports improvements in customer service.
75
12 Dept. head supports quality improvement teams.
75
30 QMJ VOL. 9, NO. 3/© 2002, ASQ
2
3
4
5
6
30
32
35
© 2002, ASQ
1
Foster article 6/18/02 10:01 AM Page 31
The Role of Quality Tools in Improving Satisfaction with Government
Factor Pattern after Varimax Rotation1. (continued)
Factor2
1
2
13 We use the tools of quality.
81
14 We use process to determine cause of problems.
77
15 Dept. members are familiar with tools of quality.
31
16 Dept. has system for measuring customer service.
3
4
41
62
48
54
18 Dept. focuses on satisfying customers.
40
53
31
19 Have effective system for resolving customer complaints.
35
48
47
20 I have been taught/used flowcharts.
81
21 I have been taught/used structured brainstorming.
80
22 I have been taught/used unstructured brainstorming.
80
23 I have been taught/used surveys.
79
24 I have been taught/used team building.
72
25 I have been taught/used customer event diagrams.
67
26 Dept. is ahead of other depts. in customer service.
31
68
27 I am proud of the work performed in department.
31
66
45
65
30
29 I feel pride when I say I work for city.
76
34
65
31 I am more satisfied with my job than two years ago.
61
32 We respond more quickly to customer needs than two years ago.
71
33 Customer service is better than two years ago.
39
34 Work is performed better than two years ago.
Variance explained
2
64
43
27.5
12.3
57
11.8
7.8
7.3
6.5
Decimals omitted; loadings lower than 30 omitted; bold identifies scale items.
Factor 1 = leadership; factor 2 = quality tools application; factor 3 = quality tools knowledge; factor 4 = perceived customer
satisfaction; factor 5 = employee satisfaction; factor 6 = process improvement.
www.asq.org 31
© 2002, ASQ
30 I am an important part of city government.
1
6
73
17 Dept. effectively communicates with customers.
28 Overall, my department satisfies our customers.
5
Purchase answer to see full
attachment