University of Southern California Dissertations and Theses

Core components of the service climate: Linkages to customer satisfaction and profitability

(USC Thesis Other)

Core components of the service climate: Linkages to customer satisfaction and profitability

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. U M I films the
text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor qualify illustrations and
photographs, print bleedthrough, substandard margins, and improper alignment
can adversely affect reproduction.
In the unlikely event that the author did not send U M I a complete manuscript and
there are missing pages, these w ill be noted. Also, if unauthorized copyright
material had to be removed, a note w ill indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning
the original, beginning at the upper left-hand comer and continuing from left to
right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher qualify 6" x 9” black and white photographic
prints are available for any photographs or illustrations appearing in this copy for
an additional charge. Contact U M I directly to order.
Bell & Howell Information and Learning
300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA
U M T
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CORE COMPONENTS OF THE SERVICE CLIMATE:
LINKAGES TO CUSTOMER SATISFACTION AND PROFITABILITY
by
Courtney R. McCashland
A Dissertation Presented to the
FACULTY OF THE SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment o f the
Requirements for the Degree
DOCTOR OF EDUCATION
August 1999
Copyright 1999 Courtney R- McCashland
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UM I Number 9955491
___ ®
UMI
UMI Microform9955491
Copyright 2000 by Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company
300 North Zeet> Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
School o f Education
Los Angeles, California 90089-0031
This dissertation, written by
under the direction o f fAC— D issertation Committee, and
approved by a ll members o f the Committee, has been
presented to and accepted by the Faculty o f the School
o f Education in partiedfulfillm ent o f the requirements fo r
the degree o f
D o c t o r o f Ed u c a t io n
' BSE
Bean
DissertafjonjCommittee
&
Chairperson
/ D - I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CORE COMPONENTS OF THE SERVICE CLIMATE:
LINKAGES TO CUSTOM ER SATISFACTION AND PROFITABILITY
Courtney R_ McCashland
University o f Southern o f California, 1999
Advisor: Dennis Hocevar
Organizations have recently begun to realize the true value o f data through integrating
multiple data points into strategic measurement systems. These strategic measurement
systems can be applied to continually evaluate and improve organizational
development and human resource practices. Theoretical models have been proposed
and tested to determine how the variables in a measurement system interact to explain
business performance; however these models have failed to include a critical variable
impacting the performance o f a business — manager talent. The model proposed and
tested in this dissertation replicates and extends current models to include manager
talent. Through regression and path analyses, this study evaluates the role that
manager talent, employee engagement and customer satisfaction each play in
determining the success o f business in terms o f customer retention and revenue
volume. The interrelationships among these variables are hypothesized a priori based
upon trends in the literature and logical reasoning. These hypothesized paths are
tested across 50 dental business units across the United States.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
upon trends in the literature and logical reasoning. These hypothesized paths are
tested across 50 dental business units across the United States.
Findings generally support the proposed model with each variable contributing
either directly or indirectly to office performance measured by customer retention and
office revenue. O f the variables measured, manager talent — the unique extension o f
this model — demonstrated the strongest direct effect on both customer retention and
office revenue.
The strong, positive relationship between employee engagement and customer
satisfaction also supported the literature; however customer satisfaction exhibited
surprising results as a direct predictor o f customer retention and office revenue. An
unexpected inverse relationship between customer satisfaction and customer retention
was observed and carefully analyzed for interpretation and understanding.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
CHAPTER I
INTRODUCTION 1
CHAPTER II
REVIEW OF LITERATURE 4
Defining Service Climate 4
Linkages Between Employee and Customer Perceptions 5
Employee-Customer Models 7
Employee Commitment/Engagement 15
Service Climate Components Predict Customer Satisfaction 18
Service Climate Components Predict Financial Performance 25
Developing an Integrated Model 30
Talent: The Missing Link 32
Research Questions 37
CHAPTER HI
METHODOLOGY 38
Sample 38
Instruments 39
Analysis 42
CHAPTER IV
RESULTS 45
Descriptives 45
Correlation Matrix 46
Regression Analysis 49
Path Analysis 52
CHAPTER V
DISCUSSION 55
Summary of Linkages 55
Interpretation of Results 60
Conclusions 64
REFERENCES 66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER I
INTRODUCTION
Companies have often depended upon data as input for major decisions. While
financial data has predominately been the report card o f the past, a growing number of
companies today collect and consider other nonfinancial indicators in evaluating their
success. Not until recently however have organizations begun to obtain the true value
from this data through integrating it into strategic measurement systems (Atkinson,
Waterhouse & Wells, 1997; Dempsey, Gatti, Grinnell, & Cats-Baril, 1997; Kaplan &
Norton, 1997; Rucci, Kim & Quinn, 1998). By establishing a model and
understanding the relationships between financial and nonfinancial business measures,
executives can apply what were once disparate data points to create a strategic
navigation system to guide planning and decisions across all levels o f an organization
(Kaplan & Norton, 1998).
This strategic measurement system can be applied to continually evaluate and
improve organizational development and human resource practices (Flynn, 1997;
Harter & Creglow; 1998). The executive who invests time and money to develop a
high-performing service climate must first understand what aspects o f the work
environment are most conducive to increased service quality, customer loyalty and
profitability. Increased competition and a tight labor pool have sent a growing
number o f service organizations searching for a better depth o f understanding o f the
management practices that are consistently found in high-performing business units
and how these practices contribute to business success (The Gallup Organization,
1998). Organizations like Sears, Xerox, Ryder, Bank One, AT&T, Taco Bell, Ford,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Halliburtan and the Bank o f Montreal are among the many companies developing
integrated systems to measure and manage the service climate as it impacts customer
loyalty and business results (Atkinson, Waterhouse & Wells, 1997; Flynn, 1997;
Kaplan & Norton, 1997; Morgan, 1998; Rucci, Kim & Quinn, 1998). By
understanding the consistent patterns that emerge and the magnitude o f these
relationships, business executives are equipped to be more purposeful in replicating
business units with high-performing service climates.
A common theme emerging from the service quality literature is that
organizations must create and maintain a climate for service in order for employees to
effectively deliver excellent service (Schneider, 1990; Schneider & Bowen, 1992).
Research beginning in the early 1980s explored the relationships between employee
and customer perceptions o f service quality and found that employees are more likely
to deliver excellent customer service when the organization expects and rewards such
behavior and establishes practices that facilitate service delivery (Schneider, Wheeler
& Cox, 1992). Although much has been written on the topic, there is very little
research investigating the effectiveness of management practices designed to enhance
service delivery (Johnson, 1996).
In only a few recent studies have researchers examined specific aspects of the
work environment and types o f management practices that are consistently reported by
employees who demonstrate high levels of service quality (Atkins, Marshall & Javalgi,
1996; Hallowell, Schlesinger & Zomitsky, 1995; Harter & Creglow, 1998; Johnson,
1996; Rucci, Kim & Quinn, 1998; Schmit and Allscheid, 1995; Schneider, Ashworth,
Higgs & Carr, 1996). By understanding the consistent findings that these studies
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reveal, organizations may begin to determine how to invest valuable resources in
developing human resource practices that enable employees and enhance service
quality.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
CHAPTER II
REVIEW OF THE LITERATURE
Defining Service Climate
The climate o f an organization is the perception incumbents share about what
is important in the organization, obtained through their experiences on the job and
their perceptions o f the kinds o f behaviors management expects and supports
(Schneider & Bowen, 1995). Research conducted by Schneider and Bowen (1992)
suggests that a positive climate for employee well-being is needed to facilitate quality
service. Survey data from 1,277 employees and 4,269 customers o f a personal lines
insurance organization were analyzed and results demonstrated that service quality can
be improved by the implementation o f a service climate; that is, a climate in which
upper management on down to public contact employees emphasize a commitment to
service and are given the required training, equipment, etc., to facilitate high-quality
service.
Schneider and Bowen (1992) suggest that the climate for service can be
enhanced by hiring people who are willing and able to deliver excellent service,
training them on aspects o f service delivery, and rewarding them for providing
excellent service. They also report that equipment, procedures, and technology should
facilitate service delivery in the eyes of both the employees and customers, and that the
functions of operations management, marketing, and human resources should work
together, rather than compete, to create a cohesive service system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linkages Between Employee and Customer Perceptions
Earlier streams o f research (Albrecht & Zemke, 1985; Parkington & Schneider,
1979; Schneider et al, 1980) pioneered the conceptual integration o f employee and
customer perceptions in the service management literature. This research was
predominately concerned with employee perceptions about the quality o f service as
perceived by customers and generally found a positive relationship between employee
and customer perceptions o f the service climate (Albert and Zemke, 1985; Schneider,
Parkington & Buxton, 1980; Schneider, 1990; Schneider and Bowen, 1992).
Schneider et al. (1980) were the first to link employee perceptions o f service
climate themes to ratings o f customer satisfaction from actual customers by collecting
data from employees and customers in 23 branches o f a large bank. Mean responses
were correlated to survey dimensions across branches to reveal several employee
variables with a significant relationship to customers’ perceptions o f overall service
quality (e.g. managerial functions, enthusiastic service orientation, and commitment to
service quality).
Schneider and Bowen’s (1985) completed a replication and extension o f this
study using 28 branches from the same bank. Most o f the same relationships were
found, although correlations tended to be lower in the second study. They extended
the first study by including the human resources practices o f supervision, status, career
facilitation, socialization, and work facilitation. All variables were related to customer
ratings of overall service quality, but not as highly as they were related to employee
perceptions of the service climate.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
Albrecht and Zemke (1985) suggested a similar set o f practices and conditions
that they expected to be related to an organization’s ability to deliver high-quality
service. These practices were: a) focusing on the customer and understanding
customers’ wants, needs, and expectations; b) developing and communicating a
service strategy that defines excellent service and how it will be delivered; c) designing
customer-friendly service systems; and d) having well-trained, service-oriented people
at all levels of the organization.
Johnson (1996) tested Albrecht and Zemke’s (1985) four service quality
components in a study conducted with employees and customers o f466 bank branches
across 134 regions in three states. Branch-level data was only obtainable on 57 o f the
total branches responding leaving a somewhat small sample for analysis. Employees
received face-to-face instruction and completed the survey during work time, which
resulted in response rates ranging from 92 percent to 100 percent across regions. The
total number of employees in the 57 branches analyzed was 538. Customers were
randomly sampled within each region. The customer survey was mailed to 77,600
current retail customers and 31,800 were returned for an overall response rate o f 41
percent. The customer sample size for the 57 branches in the study was 7,944. With a
sample size of 57, the power o f a one-tailed test to detect a significant relationship
between variables when the population correlation is .30 is .72 at alpha = .05 and .47
at alpha = .01.
Employee perceptions of the service climate were measured by a survey
comprising 10 service climate dimensions - service strategy, service support, service
systems, information seeking, training, rewards/recognition, management service
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
orientation, employee service orientation, sales/service relationship, estimate o f
customer satisfaction. Eight o f the 10 service climate dimensions rated by employees
had significant correlations with overall customer satisfaction, which was the mean of
all customer survey items used in the study — teller service, branch staff service, phone
representative service, problem solving, convenience, satisfaction with service quality,
perceived quality and bank statements.
The employee dimensions o f information seeking, training, and rewards/
recognition demonstrated the strongest relationships to overall customer satisfaction;
service support, management service orientation, and employee service orientation had
the weakest relationships as measured by the survey. In addition, the overall service
climate variable, which was the mean of all employee survey items included in the
study, was significantly related to all customer dimensions except bank statements
(Johnson, 1996).
Emplovee-Customer Models
While the relationships identified between these employee attitudes and
customer satisfaction suggest important connections between the alternative variable
sets, further conceptual and empirical evidence is needed to establish the links among
all of these variables (Schmit & Allscheid, 1995). According to Schmit & Allscheid
(1995), the relationships between employee attitudes o f the service climate and
customer perceptions of service quality need to be put into the context of an attitudinal
model that links attitudes and outcomes.
Recent studies begin to link attitudes to business outcomes within the context
of a conceptual business model or theoretical framework (Atkinson, Waterhouse &
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wells, 1997; Dempsey, Gatti, GrinnelL, & Cats-Baril, 1997; Kaplan & Norton, 1997;
Rucci, Kim & Quinn, 1998). To understand the evolution o f these models, let us first
look at historical theory surrounding the interplay o f the attitudinal linkages
comprising the left half o f an attitude-outcome equation developed by Bagozzi (1992).
Bagozzi (1992) developed a model o f attitudes, intentions and behavior
relationships which he proposed to replace the deficient attitude models traditionally
popular in psychology (e.g., Ajzen, 1991; Ajzen & Fishbein, 1980). Essentially, his
model elaborates on a more general model developed by Lazarus (1991) that links
greater emotional response with greater coping in a sequential process. Bagozzi
(1992) suggests that the appraisal process involves the assessment o f “outcome-desire
units.” Outcome-desire conflicts occur when an individual fails to meet a goal or
experiences an unpleasant event; outcome-desire fulfillment occurs when a goal is met
or the individual has a pleasant experience. These outcome-desire experiences are
followed by a negative or positive emotional responses respectively. In turn, negative
responses are followed by coping intentions to reduce the conflict, while positive
responses are followed by the coping intent to maintain, increase or share the outcome
(i.e. coping responses). Finally, behavior follows.
Gotlieb, Grewel and Brown (1994) used the Bagozzi (1992) model to explain
the sequential relationship among customers’ perceptions o f quality (i.e., appraisal),
customers’ satisfaction (i.e., an emotional response), and customers’ intentions (i.e.,
coping) to use a service in the future. The model can be applied to the other side of
the customer service-provider relationship. The measurement of the organizational
climates of well-being and service involve employee appraisals of their work situations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
These measures typically involve assessments o f goal support, leader support, role
stress, job challenge, workgroup support (e.g., Burk, Borucki, & Hurley, 1992; James
& James, 1989), all o f which are related to employees’ ability to meet goals and have
positive work experiences.
Schneider and Bowen’s research (1992) supports the theoretical model o f the
employee attitude-customer satisfaction process proposed by Bagozzi’s (1992).
Bagozzi’s (1992) model o f employee attitudes, service quality intentions, and
customer service behavior measures four climate appraisal variables — Management,
Supervisor, Monetary and Service. The four appraisal variables were hypothesized to
represent an underlying emotional response or A ffe c t toward the organization.
Employee attitudes and intentions, and customer satisfaction data from a service-
oriented organization with 160 offices is used to provide an initial test o f the
usefulness of the model. Cross-validations o f the model with the model with and
without common method variance were conducted. The results provide strong
support for the model.
Consistent with the Bagozzi (1992) model, the objective o f research conducted
by Schmit and Allscheid (1995) was to show that “employee appraisals o f climates of
well-being and service are positively related to a latent affective variable, which in turn
affects intentions to provide quality service to customers.” These intentions,
according to Schmidt and Allscheid (1995), should be positively related to actual
customer service behavior.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The study by Schmit and Allscheid (199S) was administered to approximately
9,000 employees of a large security systems company across 361 offices nationwide.
A total o f 5,085 employees completed the survey for a response rate o f approximately
57 percent. Of this original sample, 160 offices had both personnel whose primary
position requirements involved direct customer contact and an office size greater than
three individuals. Thus this smaller sample o f 3,464 employees was used in the study.
A total o f 31,362 customers served by the 160 offices completed a customer
satisfaction survey. The overall response rate for the customer satisfaction survey was
10.7 percent.
Employees rated the quality o f the service climate across Bagozzi’s four
climate appraisal variables - supervisor support, management support, service support
and monetary support - as well as a final subscale labeled perceived service-product
quality measured by sample questions such as “Our company’s products and services
meet the needs of our customers,” and “I’m proud of the quality o f our company’s
products and services.”
Schmit and Allscheid (1995) tested their proposed model against the data using
LISERL 8 (Joreskog & Sorbom, 1993). Four indicators of fit were used to assess the
models tested, including chi-square, Root Mean Square Error of Approximation, the
Normed Fit Index and the Comparative Fit Index. A RMSEA o f .05 has been
suggested as an indicator o f close fit, while .08 suggests a reasonable fit of the model
to the data (Browne & Cudeck, 1993); NFI and CFI values greater than .90 also
suggest a reasonable fit (Bentler & Bonett, 1980). The test o f the structural model fit
the data fairly well, chi-square alpha = 28.23, p< .05; RMSEA = . 12; NFI = .93;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CFI = .95. The standardized path coefficient between the four climate appraisal
variables (A ffect) and Service Intentions (.83; t = 3.71; p < .05) suggests a strong
relationship. A more moderate relationship between Service Intentions and Customer
Service was indicated by the standardized path coefficient (.33; t = 3.71, p < .05). An
indirect outcome o f A ffe c t on Customer Service was also indicated by a moderate
standardized path coefficient (.28; t = 3.61, p < .05). These findings are consistent
with the Bagozzi (1992) model. As predicted, climate appraisals were strongly
associated with an underlying emotional response; this emotional response was
strongly associated with the employees* service intentions and service intentions were
found to be moderately related to behavior, in this case customer service.
Hallowell, Schlesinger and Zomitsky (1995) propose and test a similar model
linking internal service quality, service capability and customer satisfaction. The
relationship between service quality, service capability and ultimate customer
satisfaction is based on the proposition that if an organization delivers good service to
its employees, enabling them to do their jobs well, the employees will feel they have
the ability to serve their customers well and subsequently provide superior levels of
service (Hallowell, Schlesinger & Zomitsky, 1995). Thus, the study proposes that
internal service quality (measured by eight quality components) is linked to customer
satisfaction via service capability.
The data set for the Hallowell, Schlesinger and Zomitsky (1995) study
consisted of anonymous responses to detailed written questionnaires administered to a
census o f front-line employees and front-line managers in life and/or property and
casualty lines at three divisions o f two U.S. insurance companies.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
A total o f 9,475 employees/managers responded to the survey, yielding a
response rate o f more than 75 percent . Preliminary interviews with groups of
employees and external customers derived the content o f the survey o f eight service
quality components comprising tools, policies and procedures, teamwork,
management, goal alignment, training, communication and rewards and recognition.
Individual’s satisfaction with each of these eight components o f Internal
Service Quality were regressed against the individual’s own perceived Service
Capability. A statically significant relationship (p < .01) was found for all but two of
the components measured — communication and recognition/reward - and the
strongest positive relationships were observed with tools (r = .326),
policies/procedures (r = .163) and teamwork (r = .127) respectively.
The interpretation o f the results of the Hallowell et al (1995) study should
take into consideration that the data source provided only represents the perspective of
internal customers. Actual customer satisfaction data is not provided from the
perspective of the customer. Thus, the relationship between internal service quality
and service capability are from the same data source and the magnitudes of the
correlations observed may be overstated due to the inherent intercorrelation among
items on a survey.
The majority o f the research reported in the literature around these employee-
customer relationships has been conducted within a financial/service industrial
setting. Atkins, Marshall and Javalgi (1996) build upon previously observed
relationships and replicated them in a medical setting through a survey o f more than
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
700 patients and 200 nursing-staff members at a 1,000-bed tertiary care hospital in
the Midwest.
The findings o f Atkins, Marshall and Javalgi (1996) show a strong
relationship between nurse satisfaction and patients’ intent to return in the future or
recommend the hospital to others. The study presents a theoretical “Service Quality
Model” similar to those proposed by Bagozzi (1992); Hallowell, Schlesinger and
Zomitsky (1995); Schmit and Allscheid (1995). Employee satisfaction with the
internal service climate is measured and positioned as a quality indicator for human
resource management staff.
The hypothesized model suggests that management’s ability to ascertain and
meet the needs and expectations o f staff will determine employees’ perception of the
work environment, the way daily tasks are performed, and, ultimately the magnitude
of patient satisfaction ratings (Atkins, Marshall & Javalgi, 1996).
Of the 283 nursing-staff surveys distributed, 126 were returned for a 45
percent response rate. O f the 719 patient satisfaction surveys mailed, 431 were
returned for a 60 percent response rate. Correlational analyses o f the nursing staff
and patient surveys found a strong association between overall job satisfaction and
the likelihood o f patients to recommend both the hospital (r = .535; p < .005) and the
nursing unit (r = .551; p < .005). The Nursing Subscale Summary, which measures
nurse commitment as opposed to basic satisfaction, has a strong correlation with
hospital quality (r = 75; p < .005); but only a moderate correlation with return
behavior (r = .37) and “recommend to others” (r =.37).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Other current research studies have also linked specific aspects o f the service
climate to customer’s perceptions o f service quality. In a recent study reported by
Rucci, Kim and Quinn (1998), Sears, Roebuck and Company expanded past
attitudinal models (Atkins, Marshall & Javalgi, 1996; Bagozzi (1992); Hallowell,
Schlesinger & Zomitsky, 1995; and Schmit & Allscheid, 1995) to include a unit
profitability measure. As part of a radical transformation, Sears developed an
employee-customer-profit chain business model of the company that tracked success
from management behavior through employee attitudes to customer satisfaction and
financial performance. Along with its measurement system, the employee-customer-
profit model was designed to serve as an internal management information system to
be applied for self-assessment and self-improvement with individuals across the
company (Rucci, Kim & Quinn, 1998).
The data set used to develop and test the model consisted of 26 employee
focus groups and the results o f a 70-question survey administered to 80,000
employees annually. The hypothesized employee-customer-profit chain was tested
against the resulting data using causal pathway modeling, a distinct form of
regression analysis, which examines data and observes correlations without
establishing causation. In addition, results were compared across time and place and
linkages were established through the data. Similar to previous studies (i.e.
Johnson, 1996; Hallowell, Schlesinger & Zomitsky, 1995; Schmidt & Allsheid,
1995), results demonstrated that an employee’s ability to see the connection between
his or her work and the company’s strategic objectives was a driver o f positive
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
behavior and that training or business literacy had a direct effect on revenues (Rucci,
Kim & Quinn, 1998).
Employee Commitment/Engagement
Among the employee-to-customer linkages that have been identified through
numerous global studies conducted by The Gallup Organization (1998), the strongest
direct link to service quality as perceived by customers has generally been employee
commitment or engagement, which could be interpreted as an emotional outcome to
the employee resulting from the critical components o f the work place.
In theory, committed employees would tend to demonstrate stronger
psychological ownership and behave in a more favorable manner towards customers
in their daily work. Because o f their engagement with the organization and its
purpose, they would more likely be compelled to go the extra mile in the service they
provide. This psychological ownership is typically measured by questions like I ’ m
proud to w ork here; I p la n to be w orking here fiv e yea rsfro m now; I w ould
recom m end the p ro d u cts a n d services o f th is com pany to m y frie n d s andfam ily; I
woxdd recom m end th is com pany a s a great place to w ork; a n d O verall, I am
sa tisfied w ith th is com pany a s a p la ce to work. This link between employee
commitment and the customer’s overall satisfaction with service quality has been
supported by numerous global studies conducted by The Gallup Organization in
1997 and 1998 — Ryder, Carlson Hospitality, TGI Fridays, Best Buy, and The
Associates First Capital.
Employee commitment or engagement has also been found to have a strong
relationship in Gallup studies to customers’ “intent to repurchase” o r ‘likelihood to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
recommend” a given product (The Gallup Organization, 1998). Gallup’s employee
engagement dimension (1998) incorporates overall employee satisfaction, but does
not measure satisfaction alone. The relationship between employee commitment and
overall employee satisfaction is one that has frequently been observed and reported in
employee survey research. In a meta-analysis o f research on job satisfaction, Blegan
(1993) found that variables that demonstrated the strongest positive mean correlation
(weighted and corrected for measurement error) to job satisfaction were employee
commitment (.526), communication with supervisor (.446), autonomy (.419),
recognition (.415), and peer communication (.358).
The employee engagement dimension is also very similar to Bagozzi’s final
subscale labeled “perceived service-product quality,” which is measured by sample
questions such as O ur com pany's products and services m eet the needs o f our
custom ers; and I ’ m p ro u d o f th e qu a lity o f our com pany’ s products and services.
Bagozzi referenced this “perceived service-product quality” as an emotional response
that demonstrated a strong association with employees’ service intentions, which in
effect had a moderate relationship to customer service.
Similar commitment questions have surfaced as leading indicators to
customer satisfaction in a longitudinal research study conducted by Schneider,
Ashworth, Higgs and Carr (1996). The study included a total sample of
approximately 7,000 employee respondents per administration representing a 55
percent response rate. Customer satisfaction data were collected from approximately
50,000 customers at the regional level. Rucci, Kim and Quinn (1998) likewise found
that the leading correlations to customer satisfaction included questions like W ould
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
y o u recom m end the purchase o f com pany s insurance to a frie n d o r relative: a n d
H ow much opportunity is there fo r yo u to pursue yo u r jo b a tu i career interests.
From this study, four additional questions also emerged with consistently
strong relationships to customer perceptions of the employee-customer interface - 1
fe e l good about the fu tu re o f the com pany; Sears is m aking the changes necessary to
com pete effectively; I understand our business strategy; a n d D o y o u see a connection
between the w ork yo u do a n d the com pany's strategic objectives. The Gallup
Organization (1998) has found similar questions to these that are interrelated and
measure the confidence or commitment o f employees to the company.
In general, these models and studies continue to provide evidence that
validates the positive relationships that exist between employees’ perceptions o f the
service climate and customers’ perceptions of service quality (Hartline & Ferrell, 1996;
Schmit and Allscheid, 1995; Schneider et al., 1980; Schneider & Bowen, 1985;
Schlesinger and Zorintsky, 1992; and Wiley, 1991).
One practical application o f this vein of research is to review the consistent
service climate variables across studies that directly influence customer perceptions of
service quality. By isolating these high impact variables that comprise the service
climate and build employee engagement, an organization can better manage
customers’ perceptions of service quality. This understanding can better focus
corporate investments in creating a service climate that brings optimal return to the
customer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
Service Climate Components Predict Customer Satisfaction/Lovaltv
By examining the composite o f this vein o f research, a core group o f service
climate components have typically been found to demonstrate a relationship to
customer perceptions of service quality: tools/equipment, training, information
seeking, recognition/reward, communication, management commitment, supervisor
care, teamwork, policies and procedures, expectations/goal alignment (Albert and
Zemke, 1985; Hartline & Ferrell, 1996; Johnson, 1996; Rucci, K im & Quinn, 1998;
Schlesinger and Zorintsky, 1992; Schmit and Allscheid, 1995; Schneider, Parkington
& Buxton, 1980; Schneider, 1990; Schneider and Bowen, 1992; Schneider et al.,
1980; Wiley, 1991; and Zeithaml, Parasuraman & Berry, 1990). Although these ten
service climate components, or basic human resource practices, have been linked to
service quality in one or more research studies, the consistency and magnitude of these
relationships vary. Some service quality components - tools/equipment, training,
supervisor care, information seeking, expectations/goal alignment — exhibit a stronger
more consistent link to service quality than others - recognition/reward, teamwork,
communication, policies/procedures, management commitment.
In looking at specific patterns across studies, the true relationship o f these
service climate components to customers’ perceptions of service quality becomes more
apparent. Rucci, Kim and Quinn (1998) found that employees’ attitudes towards their
jobs and the company had a greater impact on employee behavior than all other
dimensions of the survey together. The types of questions and salient components of
a service climate discovered in the Rucci et al (1998) study — goal alignment, training,
work efficacy, physical working conditions and treatment by supervisor — were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
consistent with those that have emerged as powerful links to service quality/customer
satisfaction in similar studies (Hartline & Ferrell, 1996; Schneider, 1990; Schneider &
Bowen, 1992; Johnson, 1996).
Contrary to previous research however, Rucci, Kim and Quinn (1998) found
no direct pathway from either the P ersonal Growth a n d D evelopm ent or Em pow ered
Team s dimensions measured on the survey to customers’ perceptions o f service
quality. Other studies have found that teamwork and personal development did impact
customer perceptions o f service quality (Johnson, 1996; Hallowell, Schlesinger &
Zomitsky, 1995; Hartline & Ferrell, 1996). Rucci, Kim and Quinn (1998) attribute
this disparity between their study and the literature to error in measurement:
“Clearly something about the way we measured them was flawed,” (Rucci, Kim &
Quinn, 1998).
Management Support as a component of the workplace is defined loosely
across the literature. Schneider and Bowen (1992) examine overall customer quality as
it relates to Branch Management in terms of the functionary support managers provide
and find a strong relationship exists (r = .54, p < .05). Schmit and Allsheid (1995)
measure M anagem ent Support with seven items including the following: I have
confidence in the fa irn ess o f m anagem ent; M anagem ent is rea lly in terested in the
w elfare o f associates. In this study, M anagem ent Sitpport was one component of
service climate most strongly related to the employees’ underlying emotional response
or latent A ffect towards the organization. This latent A ffe c t variable is linked to the
intention variable labeled Service Intentions. The significant correlation between
M anagem ent Support and A ffe c t was observed in both analyses (i.e. with and without
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
common method variance). Items comprising the M anagem ent Support component
also tend to be more highly related to organizational goal fulfillment than other factors
measured.
Hartline and Ferrell (1996) explore the importance o f manager commitment to
service quality as a predictor o f the quality o f service customers perceive. The
findings of Hartline and Ferrell’s (1996) research conclude that managers who are
committed to service quality are more likely to empower their employees and use
behavior-based evaluation. The use o f empowerment leads to both positive and
negative employee outcomes. Findings suggest that though empowered employees
gain confidence in their abilities, they also experience increased frustration (conflict) in
their attempt to fill multiple roles in the organization. While frustrating to some
employees, this conflict also leads to increased self-efficacy as employees learn to cope
with conflicting role demands. To increase customers’ perceptions o f service quality,
managers must increase employees’ self-efficacy and job satisfaction, and reduce
employees’ role conflict and ambiguity.
M anagem ent Service O rientation was defined by Johnson (1996) as “the
extent to which managers set examples o f excellent customer service, demonstrate
commitment to excellent customer service, and support employees in their service
quality efforts.” In Johnson’s (1996) research, M anagem ent Service O rientation
exhibited a moderate relationship to overall customer satisfaction (r = .20, p < .01).
Johnson’s (1996) overall research results showed that the service climate components
o f Inform ation Seeking (r = .48, p < .01), Training (r = .44, p < .01), a nd
Rew ardsfK ecoffu!don (r = .38, p < .01) demonstrated the strongest relationships to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
overall customer satisfaction; while Service Support, Management Service
Orientation, and Employee Service Orientation had the weakest relationships to
customer satisfaction as measured by the survey.
The implications o f Johnson’s study (1996) lie initially in its ability to once
again reinforce the importance o f workplace attitudes in creating a high-performing
service climate; however, it is the breakdown and close examination of the variables
comprising the service climate dimension that provide the most value to human
resource practitioners. While the results around Training and Information Seeking are
aligned with previous research studies (i.e. Albrecht & Zemke, 1985; i.e. Atkins,
Marshall & Javalgi, 1996; i.e. Schneider & Bowen, 1992), those around
Recognition/Rewards are not.
A finding o f particular interest from Johnson’s (1996) correlation analyses was
the significant positive relationship between employee perceptions o f
rewards/recognition in the workplace to customer perceptions o f service quality.
Although rewarding and recognizing employees who deliver excellent service is
considered an important component of a climate for service (e.g., Schneider & Bowen,
1992), previous studies examining this variable (or similar variables) have shown it to
be relatively unrelated to customer satisfaction (Hallowell, Schlesinger & Zomitsky,
1995; Schneider et al., 1980; Tom ow & Wiley, 1991; Wiley, 1991).
Close examination o f the measures o f rewards/recognition in these studies may
shed some understanding of the inverse finding. Research conducted by Hallowell,
Schlesinger and Zomitsky (1995) did not demonstrate a relationship between
recognition/reward and employee’s perceived service capability; however, the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
questions used to measure recognition/reward, i.e. A t th is com pany, yo u get rew arded
fo r providing g ood service, p ro vid e no tim e constraint. Research conducted by The
Gallup Organization (1998) suggests that work units where employees receive
continuous feedback/recognition (In the la st seven days, [h a v e received praise or
recognition fo r g ood work) demonstrate increased levels o f customer satisfaction.
Perhaps it is the ongoing recognition, as opposed to sporadic feedback or routine
programs that best enables and reinforces employees’ service to customers.
Essential basic needs like materials, tools and equipment are components o f a
quality service climate that consistently demonstrate a strong relationship to customer
service quality across studies o f employee-customer linkages (Schneider &
Bowen, 1992; Schneider, Ashworth, Higgs & Carr, 1996; and Hallowell, Schlesinger &
Zomitsky, 1995). In a replication o f Schneider’s original work (1980), Schneider and
Bowen (1992) validated L o g istics Support/Equipm ent Supply as a leading predictor of
customers’ perceptions of service quality (r = .50, p < .05). Likewise, Hallowell,
Schlesinger and Zomitsky (1995) identified Tools/E quipm ent as the leading
component of service climate predicting service capability (r = .326, p < .01).
The Gallup Organization recently completed an extensive meta-analytical study
across 28 organizations with results supporting the true validity o f similar service
climate components as links to customer satisfaction (Harter & Creglow, 1998). The
Gallup Organization has studied productive work groups and productive individuals
for more than 25 years. In developing measures o f employee perception, researchers
have conducted hundreds o f qualitative focus groups across a wide variety of
industries, which have led to the development of the 13 Core statements found in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
Gallup Workplace Audit (GWA). Quantitative analysis o f these questions has been
combined through meta-analytical research across 105,630 individual employee
responses comprising 2,618 business units.
A meta-analysis is a statistical summarization o f data accumulated across many
different studies and provides powerful information since the findings from one study
will always include measurement and sampling errors and idiosyncrasies. The Gallup
meta-analysis (1998) examined employee perceptions o f the service climate as
measured by the GWA Core items as they directly related to business unit outcomes
including customer satisfaction/loyalty, profitability, productivity and employee
turnover. Following are the GWA Core items used to measure service climate:
O verall, haw sa tisfied are yo u w ith (nam e o f company) a s a place to w ork; I know
w hat is expected o f m e a t w ork; I have the m aterials a n d equipm ent I n eed to do my
w ork right; A t work, I have the opportunity to do w hat I do best every day; In the la st
seven days, I have received recognition or praise fo r doing g o o d w ork; M y
supervisor, o r som eone a t work, seem s to care about m e a s a person; There is
som eone a t w ork who encourages m y developm ent; In the la st six m onths, someone a t
w ork has talked to m e a bo u t m y progress; A t work, m y opinions seem to count; The
m ission/purpose o f m y com pany m akes m e fe e l m y jo b is im portant; M y associates
(fellow em ployees) are com m itted to d oing quality work; I have a best fr ie n d a t work;
a n d This la st year, I have h a d opportunities a t w ork to learn a n d grow .
The meta-analysis with regard to customer criteria found that 11 o f the 13
items had positive 90 percent credibility values across studies: Items with the highest
true validities that appear to generalize across companies include I have a b est frie n d
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
a t work; A t work, I have th e opportunity to do w hat I do best every day; I know what
is expected o f me a t w ork; a n d M y supervisor, o r som eone a t work, seem s to care
about me a s a person. In general, those that measured the employee's B asic N eeds
related most highly to customer satisfaction/loyalty measures.
Business units with employees who know what was expected o f them and had
the materials and equipment they needed to perform their jobs right tended to have
higher customer perceptions o f their business units (Harter &Creglow, 1998). These
findings reinforce individual study results reported over the last few years (i.e.
Schneider, Ashworth, Higgs & Carr, 1996; and i.e. Hallowell, Schlesinger &
Zomitsky, 1995). Management practices as they directly influence expectations,
materials and equipment, relationships and job fit have been consistently found in
business units with higher customer perceptions o f service quality (Albrecht & Zemke,
1985; Johnson, 1996; Schmit, 1995; Schneider & Bowen, 1995).
As part of the meta-analysis, Harter and Creglow (1998) completed the same
analysis with profitability as the criterion variable. In this case, profitability for the
business units studied across organizations was defined as a percentage of revenue
figure. Ten (10) o f the 13 items have positive 90 percent credibility values, and the
researchers were able to account for 70 percent o f the variance in validities for nine
items. The mean percent o f variance accounted for across items is 69.21 percent.
Items that appear to generalize across companies and tended to have the highest
validities to the profitability criteria were: O verall, how sa tisfied are yo u w ith (nam e
o f company) as a p la ce to w ork; M y associates (fellow em ployees) are com m itted to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
doing quality w ork; A t work, I have the opportunity to do w hat I do best every day;
and M y supervisor, o r som eone a t work, seem s to care about me as a person.
Service Climate Components Predict Financial Business Unit Performance
Overall, the dimension from Harter and Creglow’s (1998) research that related
most highly to the profit criterion was M anagem ent Support. Those business units
where employees get to do what they do best and feel that their supervisor cares about
them and that there was someone encouraging their development tended to belong to
business units with higher levels of profit (Harter & Creglow, 1998).
It appears that those employees who feel intrinsically valued and appreciated
by the organization will tend to work in ways that produce higher levels o f profit for
their unit. Yet because this analysis was not designed to show causation, one might
also conclude that employees working in units where profit levels are higher will tend
to feel more valued by the organization and thus rate higher on the M anagem ent
Support items. This too would follow sound logic. Speculation around the direction
of this relationship will remain until a longitudinal study is conducted and reported to
include profitability and demonstrate the flow of events. While research to support the
direction of the link between service climate and profitability is inconclusive,
Schneider, Ashworth, Higgs and Carr (1996) did conduct a longitudinal study with
four data points that examined the direction of the relationship that has been proven to
exist between the service climate and the quality of service as perceived by customers.
The pattern of the relationships demonstrated in this study suggests that the causal
arrow runs from employee to customer data, rather than the reverse, over time
(Schneider, Ashworth, Higgs & Carr, 1996).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
Organizations continue to search for measurement and management
approaches to improve their service quality. The service-quality agenda, however, has
now shifted to include other issues (Zeithaml, Berry & Parasuraman, 1996). As
companies invest in their intangible assets to improve their service cultures, the
ultimate question becomes inevitable: What is the impact on my bottom line? While
one might like to think that creating a healthy service climate is the altruistic agenda of
an ambivalent executive, the obvious motive must also be a sustainable improvement
in business performance.
A handful o f studies take a close look at the components of a service climate
that demonstrate a relationship to profitability or other business outcomes (Dempsey,
Gatti, Grinnell & Cats-Baril, 1997; Harter & Creglow, 1998; Kaplan & Norton, 1996;
Schneider, 1991; Rucci, Kim & Quinn,1998; and Zeithaml, Berry & Parasuraman,
1996). Researchers have found that employee and customer perceptions o f service
quality are conditionally related to the profitability o f an organization (Gale, 1992;
Phillips, Chang & Buzzell, 1983; Schneider, 1991; and Harter & Creglow, 1998).
Schneider (1991) reports that for the most part, employee and customer
satisfaction are positively related to the profits o f a business. However, Schneider
points out that there are a number of contingencies that complicate and reduce the
reliability of this relationship. Among these are the age of the service facility, the
tenure of its employees and whether profits are dependent upon facility volume or
repeat business.
Early studies using the Profit Impact o f Marketing Strategy (PIMS) data set
have uncovered significant associations between service quality and profitability.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Findings from these studies show that companies offering superior service achieve
higher-than-normal market share growth (Buzzell and Gale, 1987). According to
Phillips, Chang & Buzzell (1983), the mechanisms by which service quality influences
profits include increased market share and premium prices. In support of this
conclusion, Gale (1992) found that businesses in the top quintile o f relative service
quality on average realize an 8 percent higher price than their competitors (Gale,
1992). This research is reinforced by evidence from companies with multiple outlets
that have reported positive quality-profitability relationship across business units: The
Hospital Corporation o f America found a strong link between perceived quality o f
patient care and profitability across its many hospitals (Koska, 1990); and the Ford
Motor Company has demonstrated that dealers with high service-quality scores have
higher-than-normal profit, return on investment, and profit per new vehicle sold (Ford
Motor Company, 1990).
Attrition o f customers is an obvious cost to an organization. Thus, companies
who keep their customers would logically tend to reduce costs and increase profits. In
an empirical study linking customer satisfaction to profits, Fomell and Wemerfelt
(1988) examine the impact o f complaint-handling programs on customer retention and
conclude that marketing resources are better spent keeping existing customers than
attracting new ones. The longevity of a customer’s relationship favorably influences
profitability. Customers who remain with a firm for a period of years because they are
pleased with the service are more likely to buy additional services and spread favorable
word-of-mouth communication (Zeithaml, Berry & Parasuraman, 1996).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Zeithaml, Berry & Parasuraman (1996) developed and tested a conceptual
model focusing on individual-level behavioral consequences of service quality, then
extrapolated the model to argue its hypothetical impact on profits. Four companies
that provide services to end or business customers were sponsors o f the research
study. A total o f 12,470 questionnaires were mailed to customers o f a retail chain,
automobile insurer, computer manufacturer and life insurer. The overall response rate
was 25 percent for a sample o f 3069.
Results supported a behavioral-intentions model that offers strong empirical
support of the intuitive notion that improving service quality can increase favorable
behavioral intentions and decrease unfavorable intentions. The direct link to profits
was never made in this study, but evidence was provided to support why favorable
behavioral intentions would logically increase profits. In short, the better the
company’s service quality scores, the higher its customer loyalty scores and the lower
its switch costs, which would intuitively lead to higher profit margins (Zeithaml, Berry
& Parasuraman, 1996).
These individual studies slowly begin to factor profitability into the service
quality equation. Only recently have researchers and organizations started to put the
pieces together and consider the interplay o f employee attitudes, customer
satisfaction and profitability across a business. The rapidly emerging view among
practitioners and academicians is that an expanded measurement system is needed for
reliably predicting the future implications o f a firm’s strategies and investments
(Dempsey, Gatti, Grinnell & Cats-Baril, 1997).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In response to a variety o f concerns with current systems for valuing a
company’s worth, the government established the Special Committee o f Financial
Reporting in 1991. Part o f the Committee’s charge was to suggest ways to improve
the nature and extent o f information provided to users o f financial statements. The
final report o f the Special Committee states that in order to meet users’ changing
needs, business reporting must “focus more on factors that create longer-term value,
including nonfinancial measures indicating how key processes are performing” and
must be able to better align information reported externally with information reported
to executives for managing the business (Dempsey, Gatti, Grinnell & Cats-Baril,
1997).
In an effort to explore the linkage between the use o f various performance
measures and their predictive value, Dempsey, Gatti, Grinnell and Cats-Baril (1997)
surveyed 2,751 members from the Association for Investment Management and
Research (AIMR). The final survey instrument included 63 financial and nonfinancial
measures often cited as possible predictors of long-term performance.
The results of the study indicate that financial analysts recognize the value of
many nonfinancial operating measures to assess the strategic performance of the firm.
While the investment community makes heavy use of traditional financial measures to
assess company prospects, the survey results show that analysts use, or are interested
in using, a broad range of nonfinancial information as well (Dempsey, Gatti, Grinnell
& Cats-Baril, 1997). If analysts are to accurately forecast the level and variability of
future earnings of the firm, they must be able to acquire information on key leading
indicators like product quality, employee commitment and customer satisfaction.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Developing An Integrated Model
In an attempt to exploit intangible assets within the context o f traditional
financial indices o f performance, many organizations have moved towards Kaplan and
Norton’s (1996) integrated measurement system called the “balanced scorecard.” The
essence of Kaplan and Norton’s approach is that the drivers o f financial performance
are the relationships a company develops with its customers and that the internal
business processes that it designs and manages to achieve customer satisfaction define
and shape customer relationships. From this theory, Kaplan and Norton (1996)
developed a model of how process results create customer satisfaction that, in turn,
creates owner results.
Sears, BP Chemicals, Halifax, Shell and Xerox are among the companies
reported in the literature for establishing an integrated measurement system to
strategically manage business results. The Gallup Organization currently works with
companies including The Bank of Montreal, Ryder, Caterpillar, Halliburtan, The
Associates First Capital, The Carlson Company and Best Buy that have built or are in
the midst of building an integrated measurement system to refine business strategy and
design needed interventions. Rucci, Kim and Quinn (1998) report the results o f a
longitudinal study to validate the “Employee-Customer-Profit Chain” at Sears. In 18
months from mid-1994 to the end o f 1995, Sears produced a model; refined it three
times, and created an index o f the company as a whole. The organization collects data
continually, assembles information quarterly, and recalculates the impacts o f the model
once a year. The model now shows that a 5-point improvement in employee attitudes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
will drive a 1.3-point improvement in customer satisfaction, which in turn will drive a
.5 percent improvement in revenue growth.
Atkinson, Waterhouse and Wells (1997) extend Kaplan and Norton’s model
and theoretically propose the inclusion o f the role o f the community and consider
performance measurement as a two-way process. The article speaks to the importance
o f the performance-based management systems that can be designed to focus
employees and management on the practices with the highest long-term return. While
Atkinson, Waterhouse and Wells (1997) provide case study examples o f this model in
practice, empirical validation o f the model is not provided.
According to Kaplan and Norton (1996), more than 100 organizations have
implemented the balanced scorecard approach. In cases like Sears where an integrated
measurement system has been established, correlational analyses, linear regression and
structural equations modeling may be applied to examine linkages among employees,
customers and profits. The Gallup Organization has conducted much of this research,
which is typically commissioned by a specific organization and held as proprietary
information for competitive advantage.
As a growing number o f organizations create a comprehensive, strategic
measurement system and analyze data both concurrently and longitudinally, more
conclusive results linking to financial performance may be obtained. Today, only a
handful of studies have been reported to validate the linkages among the service
climate (employee), service quality (customer) and financial performance (profit).
Companies have an abundance o f data; it’s time to put the pieces together and
understand how to apply the data for continuous performance improvement. Once we
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
do, we will then be able to drill down to better understand how to manage the leading
indicators of an organization’s future performance.
Talent: The Missing Link
The value o f an integrated measurement model as a management tool for an
organization is fully dependent upon the ability o f the model to capture the most
essential components that predict focal business outcomes. O f the models reported in
the literature, all o f them fail to include a fundamental element of high-performing
organizations - the talents o f the people. Talent for the purposes o f this review will be
defined as the fit o f an individual’s strengths to the task at hand. With the tight labor
market today, organizations clearly have a growing need to increase employees’
productivity, job satisfaction and commitment. Ironically, not one of the reported
strategic measurement models considers the impact o f talent in building a service
climate and managing business performance (Kaplan and Norton,1996). As an
extension to the current research, the model proposed in this study will incorporate
manager talent as an endogenous variable driving the service climate, service quality as
perceived by customers, and ultimate business performance. Below is a brief review of
the limited empirical studies available that link talent, and most specifically
manager/leader talent, to the service climate, customer perceptions of service quality
and overall business performance.
Early research about stress in the workplace visits the importance of job fit to
creating a less stressful work environment (Cox, 1978). This relationship follows the
transactional view that “stress occurs when an individual perceives the external
demands are greater than the individual’s ability to meet them” (Cox, 1978). If there
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
is an improper fit between the individual and the environment, the individual
experiences stress (Harris and Arendt, 1998).
Research conducted by Judge (1992) and Judge and Locke (1993) found that
affective disposition significantly influenced subjective well-being. Affective
disposition in this case was defined as the tendency to respond to classes of
environmental stimuli in a predetermined, affect-based manner. In Locke's study
(1993), the reciprocal relationship between subjective well-being and job satisfaction
was also observed. This evidence suggests that some employees with have a greater
or lesser predisposition to be more or less satisfied and/or engaged in their work place.
Numerous proprietary research studies over the last two decades have
identified that leadership o f an organization affects organizational outcomes and
influences those served by these organizations (Araki 1982; Bass 1985; Glisson and
Durick 1988; Klimoski and Hayes 1980; and Kouzes and Posner 1990). Variables
such as bureaucratic or participative style, locus o f control, risk taking, age and
functional background, and reward/punishment behavior o f the leader have been
examined in relation to strategy formulation and performance (Helmer and Suver,
1988; Nahavandi and Malekzadeh, 1998; Taylor 1978; Williams, Podsakoflf, and
Huber 1992).
Yet while there is research on the effect o f leadership, only one study found by
this researcher in the public domain identifies specific leadership talents or behaviors
that have been quantified in relation to employee productivity, commitment and the
quality of the service climate. McNeese-Smith (1996) conducted this study in two
suburban, not-for-profit hospitals in the Seattle area. Forty-one department managers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and 610 employees who were responsible to one o f the managers were invited to
participate. These managers completed the Leadership Practices Inventory for Self
(LPI-Self) ( Kouzes and Posner, 1990). Employees also evaluated their managers on
the five leadership behaviors using the Leadership Practices Other (LPI-Other)
(Kouzes and Posner, 1992). Each measurement instrument has thirty questions rated
on a five-point Likert-type scale. The results from these assessments were then linked
to productivity, job satisfaction and organizational commitment o f the direct reports to
these managers, and a positive relationship was found between a manager’s use of
specific leadership behaviors, and an employee’s productivity (r = .23, p < .001), job
satisfaction (r = .30, p< .001), and organizational commitment (r = .36, p< .001).
Consistent results have been the result of numerous predictive studies
conducted by The Gallup Organization across mid and front-line managers and
leaders. With a research database of more than 80,000 managers and 10,000 leaders,
The Gallup Organization consistently tracks the performance o f talent assessment
interviews in predicting subsequent performance on the job. Gallup has found that
manager talent has a consistent and significant influence on the productivity,
commitment/engagement, satisfaction and retention o f employee teams.
The purpose o f this investigation is to expand the current knowledge base
around linkages among critical variables in the workplace and business performance.
The study will replicate previous studies (Bagozzi, 1992; Johnson, 1996; Schmit &
Allscheid, 1995) and extend them to include a variable that measures manager talent.
Unlike the currently published integrated measurement models (Kaplan and Norton,
1996; Rucci, Kim and Quinn, 1998), the proposed model tests manager talent as a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
leading indicator o f employee perceptions o f the service climate and customer
perceptions o f service quality, then demonstrates the relationships among all these
variables as they explain the variance in business performance across 50 business units.
Business performance for the purposes o f this study is measured by office revenue
(weighted to provide a measure o f unit profit) and customer retention. In this
hypothesized model, customer retention and office revenue will be tested as criterion
variables with employee engagement, customer satisfaction and manager talent as
predictors (see Figure 1).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission.
FIGURE I
Hypothesized Path Model
Customer
Retention
Customer
Satisfaction
Manager
Talent
Office
Revenue
Employee
Engagement
37
Research Questions
1) Does the mission o f management explain the variability in customer
retention and revenue within an office?
2) Does manager talent help predict the retention o f customers and the size of
revenues within an office?
3) Is customer satisfaction better explained when considering the effects of
manager talent?
4) What is the relationship between employee engagement and customer
satisfaction?
5) How do employee engagement and customer satisfaction relate to
customer retention and revenue within an office?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER m
METHODOLOGY
Sample
Data for the study were collected from 50 dental offices representing
geographic areas o f the United States and Canada. The sample was a random
distribution o f graduates from the L.D. Pankey Institute, a continuing education school
for dentists. For each office, data were obtained from three sources — the dental
manager, office employees (staff) and office customers (patients). Data measuring
office business performance - customer (patient) retention and office revenue - were
collected from each office and compiled by the Pankey Institute.
A census o f dental managers and employees was surveyed along with a
purposeful sample of 50 customers (patients) for each office. The purposeful sample
comprises a target base o f customers the dentists in each office desire to retain. These
customer attitudes define the services and standards the office aspires to provide. The
objective o f each office is to attract and retain more patients like this target population.
A total o f270 employees and 1592 customers responded to the surveys. The
overall employee response rate was 92 percent and the overall customer response rate
was 64 percent. All except three offices had 30 or more customers respond. Dental
Managers for every participating office completed a talent interview resulting in 100
percent cooperation for this measure. Finally, all offices had a per-patient retention
score for those participating in the study and a raw dollar figure reflecting total office
revenue, as well as a weighted dollar figure reflecting total office revenue per dentist
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
per hour. While longitudinal data would be ideal to measure lag effects, all data for
this study were collected concurrently and represent a single period o f time.
Instruments
The engagement of employees for each office was measured using a 25-
question written survey with dichotomous coding on a YES, NO, DK scale. The
survey included questions like
• “I know what is expected of me at work,”
• “My manager or someone at work seems to care about me as a person,”
• “At work, 1 have the opportunity to do what I do best every day,” and
• “In the last seven days, I have received recognition or praise for good work.”
These questions have been tested by the Gallup Organization for reliability and
validity in measuring employee engagement across a number o f industries (Harter &
Creglow, 1998). The internal consistency reliability coefficient (BCR-20) of the overall
employee survey for this study is .86. All offices worked to attain a census of
employee respondents. The number of total employees per office range anywhere
from four to seven members. A strong response rate of 92 percent improved the
accuracy o f results for offices with small sample sizes.
The satisfaction of customers for each office was measured using a 72-question
written survey measuring customers’ perceptions o f service quality. The items on the
survey were primarily coded on a 5-point Likert Scale with 1 being “terrible” and 5
being “excellent.” Customers were asked questions like
• “My dentist is never too busyto answer any questions Ihave,”
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
• “I am confident that my dentist practices all proper sterilization procedures,”
• “I believe that this dental practice is the best available forthe treatment I need,”
• “The doctor and staff members are considerate of my feelings,”
• “My dentist knows a lot about me as an individual,” and
• “My dentist gives me explanations I can understand.”
The reliability and validity o f this instrument have been tracked by the Gallup
Organization across three waves o f customer studies. The internal consistency
reliability coefficient (KR-20) o f customers’ overall perceptions o f service quality for
the baseline study is .75. Between 30 and 50 patients responded per office.
The instrument used to access the talent of dental m a n a g e rs was a subset of 28
relationship questions from a telephone interview designed and validated by the Gallup
Organization to predict the talent o f Dental Managers (Gallup, 1989). The technical
report supporting the validity and reliability of this instrument is provided in
Appendix 1. The internal consistency reliability coefficient (KR-20) is .74 for the
28-core measuring manager talent as defined by the study.
These interviews were transcribed and analyzed by two certified Gallup
analysts. To obtain certified status, each analyst completed 20 analyses under the
supervision of the researcher who designed the interview. Each o f the 28 interview
questions was scored dichotomously with “1” representing a response that is
predictive of the desired attribute and “0” being nonpredictive o f the desired attribute.
To test the interrater agreement between the two raters, the tw o analysts
independently coded fifty percent o f the total fifty interviews. The average percentage
of agreement between the two analysts for the study was 96 percent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
Four critical talent dimensions are represented by these 28 interview questions:
Mission, Developer, Stimulator and Arranger. Each Dimension is measured using
seven questions that have longitudinal data in Gallup’s interview database to support
their power as predictors o f manager talent. For each o f the seven-question talent
dimensions, the internal consistency reliability coefficients (KR-20) were run.
Reliability for Mission is .36, Developer is .49, Stimulator is .39, and Arranger is .46.
The interview questions per dimension are provided at the end o f the validity study
presented in Appendix 1.
Dentists who are high in the Mission theme will tend to exhibit a high level o f
service orientation in their daily work. These dentists will likely thrive most on helping
and serving others. They are driven by altruistic intentions to better the lives of the
patients they serve. This theme taps into an aspect o f manager talent discussed in
previous research studies involving the workplace (Crozier, 1964; Desatnick, 1987).
The other three themes represent managers’ abilities to understand the
uniquenesses of employees and customers and to adapt their management styles to
stimulate and develop each individual. Similar relationship talents have also been
discussed in prior research studies involving issues in the workplace (Czepiel,
Solomon and Surprenant, 1985). These dimensions each offer unique information in
understanding the overall effectiveness of the dentist in servicing, teaching and
retaining patients, as well as employees.
Customer retention was reported for each customer (patient) who participated
in the survey and then aggregated per office. Office revenue measures were provided
to the Gallup Organization by the Pankey Institute. The total office revenue was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
considered, as well as a weighted measure o f revenue. The weighted measure
considered the number o f patients per doctor per hour. The per hour weighting
incorporated unit productivity as well as service pricing, which provided a stronger
reflection of unit profitability.
Analysis
Data across all offices was compiled to provide an aggregate on each o f the
critical variables measuring the workplace and its performance — employee
engagement, customer satisfaction, manager talent, customer satisfaction and office
revenue. Descriptive statistics were run across each o f these key variables. Because
the data are treated as continuous, a Pearsons Product Moment Correlation Statistic
was applied in measuring the simple zero-order correlations among all variables
included in the study.
Prior to analyzing the key variables in question, the mission dimension - a
subset of the talent composite variable - was isolated and considered as an
independent variable in place o f the complete talent variable. The examination of
manager mission is an important consideration because it is the only aspect of manager
talent reported in the review o f literature that demonstrated a link to employee
attitudes (Crozier, 1964; Desatnick, 1987).
Two Stepwise Regression analyses were run with dependent variables
customer retention and office revenue respectively. Manager mission, employee
engagement and customer satisfaction were the independent variables considered as
measures of the service climate. The results o f these regression analyses address the
first research question proposed at the onset of the study: D oes the m ission o f
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
m anagement explain the variability in custom er retention a n d revenue w ithin an
office?
Stepwise Regression is a regression analysis method by which the computer
enters independent variables into the equation. In Stepwise Regression, PIN and
POUT parameters are set to determine at what significance level an independent
variable is entered into the equation and likewise removed from the equation. A .05
level of significance was required for a variable to be entered into the equation. Thus,
the researcher has 95 percent confidence that the magnitude o f the relationships
presented are not the product o f Type II error.
Finally, a path analysis was conducted to assess the fit o f the data to the
hypothesized model o f relationships in Figure 1. This fully defined model o f expected
path coefficients provides a summary o f all the key relationships under consideration
while accounting for the direct and indirect effects o f variables in explaining the
variance of customer retention and office revenue. The nature and magnitude of these
relationships answer the last four research questions proposed at the onset of this
study: D oes m anager ta lent help predict the retention o f custom ers a nd the size o f
revenues w ithin an o ffice? Is custom er satisfaction b etter explained when considering
the effects o f m anager talent? What is the relationship betw een em ployee
engagem ent an d custom er satisfaction? How do em ployee engagem ent and custom er
sa ti fa c tio n relate to custom er retention a n d revenue w ithin an office?
The hypothesized model is the product o f careful examination of the research
around each of the paths represented. The path analysis is especially revealing because
it captures the anticipated multicollinearity among the variables measuring the quality
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
of the service climate - manager talent, employee engagement and customer
satisfaction. The path model allows us to begin to look at the direct and indirect
effects o f each o f these variables to the proposed criterion variables - customer
retention and office revenue.
In theory, manger talent would be expected to directly impact the engagement
of employees in their daily work and engaged employees would produce more satisfied
customers (Schneider & Bowen, 1992; Schmit & Allscheid, 199S). Likewise,
manager talent would be expected to impact customer satisfaction directly and
customer satisfaction would be expected to have a recursive relationship back to
employee engagement (Johnson, 1996; Schneider & Bowen,1992; Schneider,
Ashworth, Higgs & Carr, 1996; and Hallowell, Schlesinger & Zomitsky, 1995).
Therefore, one would expect that measures o f each these variables - manager talent,
employee engagement and customer satisfaction -- would overlap in explaining the
variance across offices in customer retention and office revenue.
Zero-order correlations and regression analyses provide an initial review of
these hypothesized relationships, and the path analysis allows us to interpret the direct
and indirect relationships o f these variables as they explain office performance.
Following is a summary o f results from the study o f manager talent, employee
engagement, customer satisfaction and office revenue across 50 dental offices.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER IV
RESULTS
Descriptives
The aggregate means and standard deviations for each o f the five critical
variables - manager talent, employee engagement, customer satisfaction, customer
retention and office revenue - is provided in Table 1 below. Different measures and
sample sizes are applied with each variable, but the standard deviation still provides
some information about the dispersion o f scores across respondents for each variable.
The per-participant results for employee engagement and customer satisfaction and
customer retention were aggregated to provide all data at the office unit level.
Manager talent and office revenue were already represented at the unit level.
Table 1 D escriptive S ta tistics fo r C ritical W orkplace V ariables
Variable Cases Mean Standard
Deviation
Manager Talent 50 12.23 4.47
Employee Engagement 270 .86 .10
Customer Satisfaction 1592 .57 .50
Customer Retention 1592 9.38 4.19
Office Revenue 50 475,874.53 230,805.05
These mean scores are all relative to the scales applied in reporting results per
measurement instrument. As explained previously, employee engagement was
measured on a binary scale where “ y e s ” * s coded 1 and “no” is coded 0. While
customer satisfaction was collected using a 5-point scale, these results were analyzed
using the same approach where in this case “5” is coded 1 and everything else is
coded 0. Scientists at The Gallup Organization have found this approach valuable to
clients because it focuses managers and executives on percentage o f “5’s” or “yes’s”,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
which have generally been found to have the strongest relationship to performance
criterion variables like customer retention and sustained revenue (Gallup Employee
Survey Database, 1998). In other words, attitudes that fall into this “top-box”
category best explain employee behaviors as measured by performance.
The mean score for talent reflects the average number o f questions where
predictive responses were given from the 28-question core manager talent interview.
Customer retention represents a per-patient average o f retention measured in months.
Office revenue represents the cumulative average revenue over the last 12 months.
Correlation Matrix
The relationships between pairs of variables in the study were analyzed using
Pearsons Product Moment Correlation Statistic. The zero-order correlations among
variables are presented in the correlation matrix below (see Table 2). Significance was
evaluated using a two-tailed test set at the .05 level.
Table 2 C orrelation M a trix o f C ritical W orkplace V ariables
Variables Manager Employee .Customer Customer Office
Talent Engagement Satisfaction Retention Revenue
Manager 1.00
Talent
Employee .12 1.00
Engagement p = 430
Customer .16 .41 * 1.00
Satisfaction p =.263 p = .002
Customer .22 -.15 -.28* 1.00
Retention p = .133 p = .301 p = .044
Office -.38* -.37* -.10 .21 1.00
Revenue p = .017 p = .013 p = .528 p = .166
p < .05
The results of the simple correlations above reveal some expected and some
unexpected relationships. The magnitude o f the simple correlations between these
paired variables is an important consideration prior to running Multiple Regression
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
analyses with customer retention and revenue as dependent variables. While few
comparable studies have been conducted to include the manager talent variable,
empirical evidence does consistently support the link between employee and customer
attitudes (Albert and Zemke, 1985; Schneider, Parkington & Buxton, 1980; Schneider
and Bowen, 1992).
It is not surprising that in this study, employee engagement and customer
satisfaction are significantly correlated (r = .41, p< .05). This supports widely-held
theory and previous research indicating that the attitudes of employees are passed
along to customers during the service interaction (Schlesinger and Zomitsky, 1991;
Schneider, 1980; and Schneider and Bowen, 1985) or perhaps it is the attitudes o f
customers that rub off on the employees who service them. In either case, there is
clearly a positive relationship between the two. The magnitude of this relationship
suggests that both variables may explain a similar aspect o f the service climate and
thus overlap in explaining select dependent variables in the Regression Formula. In
effect, employee engagement and customer satisfaction are more correlated with each
other than they are with either of the designated dependent variables - customer
retention or office revenue.
The third independent variable - manager talent - however, is not significantly
correlated with employee engagement or customer satisfaction ( r = .12 and .16
respectively, p > .05) and therefore appears to explain a unique element o f the service
climate. Manager talent is also more highly correlated with the criterion variable,
revenue, than it is with either employee engagement or customer satisfaction. Thus,
manager talent appears to have the most influence as a predictor of office revenue and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
customer retention. By running a Multiple Regression analysis, we can separate and
analyze the interconnections among variables with more clarity and accuracy.
One o f the four dimensions measured by manager talent is mission — the
altruistic desire to service customers with a motive to make their lives better in some
way. The mission o f the manager has been found in early studies to have an impact on
employee and customer retention (Crozier, 1964; Desatnick, 1987). To investigate the
unique contribution o f the mission dimension, mission was isolated as a subset o f the
talent dimension and correlated with employee engagement, customer satisfaction,
customer retention and office revenue. O f these correlations, mission was found to
have a significant relationship with only one variable: customer retention. The
Pearson Product Moment Correlation Statistic between mission and customer
.retention is r = .30 at the .05 level of significance. To better analyze this relationship,
mission will be isolated and treated independently as part of the regression analyses.
Two counter intuitive results were revealed in the correlation matrix. First,
employee engagement was found to have a significant inverse relationship with office
revenue (r = -.37; p < ,05). This may be explained by large, high volume offices that
work employees hard and provide little incentive, recognition or reward for their
efforts. The ideal examination of this relationship would be repeated measures over
time. The question that arises is whether offices that rush high volumes o f business
through their offices provide the same quality o f service and ultimate sustainability of
customers. Will offices with high service volumes tend to lose customers and suffer in
producing strong profit margins over some designated time period? While additional
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
analysis can provide insight around this relationship, this study is not longitudinal and
does not provide the opportunity to analyze any type o f lag effects.
The second startling result found through conducting simple correlations is the
relationship between customer satisfaction and customer retention. A significant
inverse correlation (r = -.28; p< .05) suggests that the more satisfied the customers the
more likely they are to leave. This is perhaps the most peculiar result found from this
study and merits further analysis and review. It is quite likely that there may be an exit
restriction for some dental practices in smaller communities with only one alternative
for service. If this is the case, separating large and small regions would allow us to
control for any restriction that may be impeding a customer’s ability to seek care of
treatment elsewhere.
Regression Analysis
For the Stepwise Regression analysis, the entry of independent variables into
the equation was determined by PIN and POUT parameters. PEN and POUT were set
at .05 and .10 respectively. This entry method o f regression analysis allows us to
observe the interrelationships among the independent variables in the equation as they
are entered into the equation. The object o f these Stepwise Regression analyses is to
isolate the mission dimension of manager talent and to answer the fifth o f the proposed
research questions for this study: D oes the m ission o f m anagem ent explain variability
in custom er retention and revem te w ithin an office?
In analyzing customer retention, mission was entered first into the equation
because it has the highest zero-order correlation with customer retention o f the three
variables considered - mission, employee engagement and customer satisfaction. As a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
significant predictor o f customer retention (Beta = .30, Sig T = .037), manager
mission explains 9 percent o f the variance in customer retention (R Square = .09).
None of the remaining independent variables account for unique variance in explaining
customer retention; therefore, they were not entered into the equation. Consequently
according to these results, neither employee engagement nor customer satisfaction was
found to be a significant predictor o f customer retention. While only 9 percent o f the
variance in retention is explained, these results summarized in Table 3 support the
hypothesis posed in research question 5 regarding the impact of mission on customer
retention.
Table 3 Stepwise R egression w ith Customer R etention a s the D V a n d
M anager M ission, Em ployee Engagem ent and C ustom er S a t as I Vs
Multiple R .301
R Square .091
Adjusted R Square .071
Standard Error 4.1108
F = 6.94518 Signif F = .037
A separate Stepwise Regression Analysis was conducted with office revenue as
the dependent variable and manager mission, employee engagement and customer
satisfaction as possible independent variables. In this analysis, none o f the independent
variables considered entered into the equation as significant contributors to office
revenue. For a closer look at the manager mission dimension, a forced-entry
regression analysis was completed and manager mission demonstrated a weak, non
significant relationship to office revenue (Beta = .14, Sig T = .391).
When manager mission is removed from the equation and substituted with the
composite manager talent variable, the contribution is increased considerably (Beta =
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
.37, Sig T = .017). This suggests that manager talent overall is a significant predictor
of office revenue, accounting for 14 percent o f its variance (R square = .135). These
results also provide strong support for research question 4 regarding the impact of
manager talent on office revenue. This relationship will be explored further in the
forthcoming path analysis. A. summary o f the contribution of manager talent to office
revenue is provided in Table 4.
Table 4 Stepwise R egression w ith O ffice Revenue a s the D V a n d
M anager Talent, Em ployee Engagem ent and C ustom er S a t as TVs
Multiple R .367
R Square .135
Adjusted R Square . 113
Standard Error 204,608.01
F = 6.94518 Signif F = .017
One limitation in this study was the inability to report an accurate measure of
profitability that is comparable across offices. Depending upon how the office is
managed, personal costs may or may not be incorporated into business costs. The
variability of these costs confounds the accuracy of bottom line statistics on the overall
office. For this reason, this study did not explore profitability as a performance
measure, but studied only the service climate in relation to office revenue and
customer retention. Two distinct conditions contribute to the size o f revenue produced
by an office - number of total patients and the charge for services. Without clarity
around what is driving revenues up - patient volume or prices - office revenue may be
misleading. To accommodate in part for this limitation, an alternate measure of
revenue was also considered - revenue per doctor per hour.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
Revenue per doctor per hour controls for the number o f doctors in the office
and gives some indication o f the relative profitability o f the service volume. While this
measure may primarily reflect how the dentist prices services, it does provide some
indication o f relative profit margins assuming that actual service costs per procedure
are fairly consistent across offices.
Results from a Stepwise Regression Analysis with revenue per doctor per hour
as the dependent variable found manager talent once again to be the single significant
predictor o f those considered - employee engagement, customer satisfaction and
manager talent. In addition, the predictive power o f manger talent is even greater in
explaining revenue per doctor per hour (R = .40; p < .05). In short, manager talent
accounts for approximately 16 percent of the variability in revenue per doctor per hour
(R Square = .158).
Path Analysis
Path analysis was applied in order to operationalize the constructs contained in
the theoretical model proposed in Figure 1. This model summarizes the relationships
that have already been discussed while accounting for the direct and in-direct effects of
the variables in the service climate — manager talent, employee engagement and
customer satisfaction - as they explain the variation in performance across offices -
customer retention and office revenue.
Path analysis is a type o f causal analysis that provides estimates o f unknown
coefficients in a set o f linear structural equations. In its most general form, a path
model assumes a specified causal structure among a set o f latent variables, some of
which are dependent or endogenous and others which are independent or exogenous.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The hypothesized relationships in Figure 1 represent a fully defined model in
which all paths are specified. All variables are in some way expected to influence
others in the model. The manager talent variable is the only exogenous variable
proposed because there is no path that runs into it. Subsequent variables in the model
- employee engagement, customer satisfaction, customer retention and office revenue
- are considered endogenous variables that may both influence and/or be influenced by
other variables.
The results o f this analysis are provided in Figure 2. The path coefficients
presented the first four questions posed at the onset o f this study: D oes manager
talent help p red ict the retention o f custom ers and the size o f revenues w ithin an
o ffice ? Is custom er satisfaction b etter explained when considering the effects o f
m anager talent? W hat is the relationship between em ployee engagem ent and
custom er satisfaction? How do em ployee engagem ent and custom er satisfaction
relate to custom er retention a n d revenue within an office?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
FIGURE 2
Path Analysis Results
.2826
Customer
Satisfaction
Customer
Retention
-.2825
.1600.
.1042
Manager
Talent
.4100
.0050
.1200
Employee
Engagement
Office
Revenue -.3419
.4763
CHAPTER V
DISCUSSION
Summary of Linkages
The results o f the Path Analysis provide a clear picture of how the data fit the
proposed model o f relationships. To summarize the individual effects o f variables
within the model, standardized path coefficients provide the magnitude of the direct
and indirect relationships to customer retention and office revenue.
Research Q uestion ~2: D oes manager talent help p red ict the retention o f
custom ers and the size o f revem ies w ithin an office?
As expected, manager talent plays a meaningful role in the model with a strong
positive direct influence on both customer retention and office revenue. Because the
manager is the dentist who interacts regularly with the patients, the robust relationship
of the manager talent to customer retention (beta = .2826) is not surprising. Likewise,
a talented manager would be expected to have a powerful direct impact on office
revenue. In this study, this theory also tested true as reflected by a path coefficient
of .4763.
Overall, the relationship talents of the manager are the most powerful measures
in the study impacting business performance. A number o f implications should be
considered with this finding. Other strategic measurement models do not include
talent as a part o f the equation, yet manager talent may be one o f the most influential
variables driving business performance. The ability o f a model to accurately depict a
simplified reality is heavily dependent upon the inclusion o f the key variables that best
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
capture the majority o f the variation in business performance. Perhaps Kaplan and
Norton’s balanced scorecard (1997) should be extended to include manager talent.
Intuitively, we recognize the importance o f a strong manager to the success o f
a business. The results o f this study clearly illustrate that some individuals have talents
that are better suited to manage an office than others. A careful inspection o f manager
talent can help us to interpret the components driving these results in order to draw
inferences around their meaning and relevance. Talent has been narrowly defined in
this study to include four o f the dimensions - mission, developer, stimulator and
arranger - found by The Gallup Organization to best explain the success o f effective
dental managers (see Appendix 1 for technical report). These dimensions were
selected a priori because cumulative data from research studies conducted by The
Gallup Organization, as well as the original dental manager studies, show that the
ability of managers to build relationships and communicate a mission contributes to
their effectiveness and overall performance (Appendix 1: The Dental Practitioner
Perceiver, 1989). Gallup’s proprietary research is supported by published studies
measuring similar aspects o f management (Crazier, 1964; Czepiel, Solomon and
Surprenant, 1985 and Desatnick, 1987).
Research Q uestion #5: Is custom er sa tisfa ctio n b etter explained when
considering the e ffe c ts o f m anager talent?
The composite o f these talent dimensions would naturally drive the quality and
efficiency of the service provided, and consequently, customers will tend to be more
satisfied when service quality is higher. This assumption is supported by the path
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
between manager and customer satisfaction (beta = .1600), although the path is not as
strong as this researcher had anticipated.
Talented dental managers will tend to invest in their employee teams and
stimulate higher levels o f performance in servicing customers. Employees will likely
be more satisfied with their work environments when they are given opportunities to
learn, grow and develop. This hypothesis is also supported by this study with a path
coefficient o f . 1200 as talent helps to explain the variation in employee engagement.
Once again, the magnitude o f this relationship falls short o f what this researcher had
expected, although the insignificant relationship may have been in part due to the
limited sensitivity o f the study due to the small sample o f offices (n = 50).
Dental managers with mission want to make a direct and significant difference in
the lives of employees (staff) and customers (patients). This clear sense o f purpose
provides motivation, direction and stability for employees and customers. These dentists
will likely thrive most on helping and serving others. This dimension taps into an aspect
of manager talent discussed in early research studies involving the workplace (Crozier,
1964; Desatnick, 1987). The altruistic motives behind the service encounter will
naturally elicit the trust and loyalty of customers as demonstrated by the path coefficient
between manager talent and customer retention (beta= .2826).
Finally, managers high in the Arranger dimension exhibit the ability to position
people and processes for maximum efficiency. This in effect will drive per-person
productivity and thus overall office revenue. The leading path coefficient of .4763
between manager talent and overall office revenue suggests that talented managers
find ways to arrange and develop their people for higher rates of productivity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
R esearch Q uestion =5: H aw do em ployee engagem ent and citstom er satisfaction
relate to custom er retention a ndrevem te w ithin an office?
The indirect impact o f manager talent on customer retention and office revenue
is not quite as easy to interpret. When we include the mediating effects o f customer
satisfaction and employee engagement, the indirect paths to customer retention and
office revenue show strong inverse relationships o f -.2825 and -3419 respectively.
O f these, perhaps the most counter-intuitive result in the study is the direct link
between customer satisfaction and customer retention (beta = -2825). This inverse
relationship elicits careful inspection. If customers are satisfied, why would they
choose to leave? Empirical evidence about customer satisfaction indicates that a
threshold exists at which level customer satisfaction and customer behavior are
positively related. But before that customer reaches a certain level of satisfaction, the
relationship to behavior is unclear (Jones and Sasser, 1995). This may in part explain
why a satisfied customer may not be a loyal customer.
From another angle, we might try to understand why a dissatisfied customer
might choose to stay. In this case with dental offices, customers (patients) may be
restricted from exiting because there is not another dental practice in close proximity
to the patient’s residence. This exiting restriction would be most likely in rural areas
without heavily concentrated populations. Thus, market size in this study may be
impacting the nature of this relationship. Perhaps segmenting the sample to control for
market size would improve the appropriate interpretation of this data. For this study,
the sample was not robust enough to support conducting two separate path analyses to
control for rural verses urban offices.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
A close inspection o f the measurement tool may also provide further insight to
the implications of the inverse relationship between customer satisfaction and
customer retention. For this study, the total score of a customer (patient) satisfaction
survey was applied as the measurement instrument. In addition to overall satisfaction,
this survey includes dimensions like accessibility, relationships, trustworthiness, and
impact. In analyzing survey results, certain dimensions of service quality will better
explain the loyalty and retention o f responding customers. In short, breaking down the
customer survey instrument to isolate those dimensions with the strongest effect on
patient behavior would likely improve the utility o f the customer satisfaction measure
in predicting customer (patient) retention for this study.
R esearch Q uestion #4: W hat is the relationship betw een em ployee
engagem ent a n d custom er satisfaction?
The magnitude and direction o f the relationship between employee engagement
and customer satisfaction (beta = .4100) was consistent with what has been
documented in the literature (Albert and Zemke, 1985; Johnson, 1996; Schneider,
Parkington & Buxton, 1980; Schneider, 1990; and Schneider and Bowen, 1992).
Overall, this link was one of the paths in the model with the most conclusive evidence
from the greatest number of previously conducted research studies. One example
from a medical setting found a similar strong relationship (r = .535, p < .005) between
nurse satisfaction and patients’ intent to return in the future or to recommend the
hospital to others (Atkins, Marshall and Javalgi (1996). A longitudinal study
conducted by Schneider, Ashworth, Higgs and Carr (1996) found a significant causal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
relationship between employee perceptions o f the service climate and customer
perceptions o f service quality.
While the literature provides consistent, conclusive evidence to support the
path between employee engagement and customer satisfaction, the empirical data to
support the path between employee engagement and revenue is inconclusive.
Logically, this researcher has proposed that engaged employees will make a greater
contribution the business, be more productive, and thus increase revenues. Contrary
to this theory, this study found a direct inverse path between employee engagement
and office revenue (beta = -.3419). Likewise, a previous study conducted by Tomow
and Wiley (1996) also found an inverse relationship between employee
engagement/satisfaction and revenue.
Interpretation of Results
Two considerations may apply in interpreting the inverse relationship between
engaged employees and overall revenue. There may be times when a company is
doing a high volume o f business, but employees are feeling overworked and
underappreciated. This may explain the relationship in some dental offices where a
high volume of patients are treated yet employees feel rushed. Employees may even
feel dissatisfied with the quality o f service they provide.
Another plausible explanation is that employee engagement is a leading
indicator of office revenue. Perhaps the concurrent relationship that was measured in
this study is not positive, but the predictive relationship o f employee engagement
today has a positive predictive relationship to subsequent revenue measured at a later
date. If this is true, employees may be engaged at a particular point in time and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
revenues may be low, but if you would track the trend over a period o f time, revenues
would rise. Trend analysis can reveal if such a lag effect, or time lapse, exists between
a change in employee engagement and a change in overall revenue. A study
conducted by the Gallup Organization in 1994 with a national retail chain found this to
be the case. While employee engagement and revenue were inversely related
concurrently, over time high morale led to increased revenues and sustained
performance.
Theory across the literature as well as a handful o f reach studies have
supported the logical link between patient satisfaction and the financial performance of
a business. Researchers have found that employee and customer perceptions of
service quality are conditionally related to the profitability o f an organization (Gale,
1992; Phillips, Chang & Buzzell, 1983; Schneider, 1991; and Harter & Creglow,
1998).
Gale (1992) found that businesses in the top quintile o f relative service quality
on average realized an 8 percent higher price than their competitors (Gale, 1992). The
Hospital Corporation o f America found a strong link between perceived quality of
patient care and profitability across its many hospitals (Koska, 1990).
Schneider (1991) reports that for the most part, employee and customer
satisfaction are positively related to the profits of a business; however, Schneider
points out that there are a number o f contingencies that complicate and reduce the
reliability of this relationship. Among these are the age o f the service facility, the
tenure o f its employees and whether profits are dependent upon facility volume or
repeat business.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
In this study o f 50 dental offices, the negligible path coefficient (beta = .0050)
between patient satisfaction and office revenue suggests that no meaningful
relationship exists between these variables. This result could be caused by the
contingencies that Schneider (1991) references above or perhaps it is the difference in
the financial measure applied in this case. Customer (patient) satisfaction may
demonstrate a different relationship to profit, than the relationship presented here to
revenue volume. Therefore the path in this model may not be comparable to patterns
reported in the literature.
Once again, a predictive analysis of the trend for this relationship may also be
helpful in drawing clear conclusions about this result. If customer satisfaction is a
leading indicator o f the financial performance of the office, one would anticipate that
when customer satisfaction is increasing, repeat measures of revenue would be
necessary to determine if a relative increase occurs in revenue after a particular period
of time.
This study found a positive direct path of .1042 from employee engagement to
customer retention, which replicates the results of a study by Schneider and Bowen
(1985). Logically, one would expect satisfied employees to provide a better quality of
service and thus retain their customers over time. Yet we should again consider how
employee satisfaction in this study is defined. When the questions asked are strictly
measures of overall satisfaction o f the employee, as opposed to measures o f employee
engagement (i.e. [p la n to be w orking h erfive years fro m now ), the impact on
customer retention may be significantly reduced. While this conclusion is in some
aspects conjecture, it does make intuitive sense.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
If satisfaction represents only a state o f satiated complacence, this may be very
different from an engagement measure that asks questions about one's psychological
commitment to the organization. Employees may be satisfied overall, but still not
recommend the products and services of their company to others. The study by
Schneider and Bowen (1985) did not technically measure employee satisfaction, but
instead measured the perceptions o f employees about service quality. The employee
measure in this study about dental offices combines questions about satisfaction,
commitment and service quality into one composite dimension - employee
engagement.
The Gallup Organization has conducted numerous studies to understand the
impact of customer and employee attitudes on performance outcomes in business
environments. Many studies show that it is not the overall satisfaction question that
best predicts performance, but instead core sets o f employee engagement questions
that predict performance. Through a meta-analytical study across organizations and
cultures (Harter & Creglow, 1998), the Gallup Organization has found a core set o f 12
behavioral questions measuring the quality o f the work environment that have been
found to consistently build employee engagement and predict business performance.
Likewise on customer surveys, questions measuring intent to return, repurchase or
recommend products and services would be expected to have a stronger relationship
to customer retention than questions measuring customer’s satisfaction with service
quality.
The employee survey in this study did measure employee engagement and not
employee satisfaction; however the customer survey in this study did not measure
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
customer loyalty, but instead measured customer’s satisfaction with service quality.
This may have had a direct impact on the effectiveness o f the customer survey
instrument in predicting customer retention and office revenue. Isolating and
understanding the predictive patterns o f these questions to actual business
performance is increasingly valuable as organizations apply measurement as a strategic
performance management tool.
Conclusions
Multiple implications can be drawn from the results of this study. First,
manager talent merits careful consideration and investment in creating a productive,
sustainable workplace as defined by customer retention and overall revenue. While a
growing number o f organizations depend upon strategic measurement models to focus
and manage their resource investments, the models reported in the literature have
failed to include manager talent as a leading indicator o f future business unit
performance. This represents a significant gap that has been addressed by this study.
Second, customer satisfaction and employee engagement measures did not
demonstrate a very powerful relationship to financial performance or sustainability of
the workplace. Subsequent research is needed to better define these constructs and to
explore the measures of employee engagement and customer loyalty as they explain
the variance in customer behaviors and financial performance across business units. In
addition, understanding the salience o f customer satisfaction and employee
engagement on office performance may require a trend analysis to examine whether
lag effects exist.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
This study has extended preliminary research around employee and customer
attitudes to include a rich and valuable element o f the workplace — manager talent. By
applying a path analysis to link these measures to performance outcomes, new
dynamics among variables measuring the workplace have been uncovered for
exploration. A replication o f this study with a larger sample may be an initial next step
in understanding the complexities o f the components that create a productive,
sustainable work environment. Finally, analyzing the components o f the service
climate by dimension, instead o f as a complete construct, would provide considerable
applied value to business leaders. With more specific information, leaders within
organizations would be able to make wiser business decisions around how best to
focus limited resources for optimal return to their businesses, as well as to the people
they employ and serve.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
REFERENCES
Ajzen, I. (1991). The theory of planned behavior: Some unresolved issues.
Organizational Behavior and Human Decision Processes. 50. 179-211.
Ajzen, I. & Fishbein, M. (1980). Understanding attitudes and predicting social
behavior. Englewood Cliffy NJ: Prentice-Hall.
Albrecht, K. & Zemke, R. (1985). Service America: Doing business in the new
economy. Homewood, EL: Dow Jones-Irwin.
Araki, C.T. (1982). Leadership study in Hawaii — How characteristics o f principals
affect the schools. NASSP Bulletin 66 (21 88-96
Atkins, P.M., Marshall, B.S., & Javalgi, R.G. (1996). Happy employees lead to loyal
patients. Journal of Health Care Marketing. 16. 15-23.
Atkinson, Waterhouse, J.H., & Wells, R.B. (1997). A stakeholder approach to
strategic performance measurement. Sloan Management Review 25-37.
Bagozzi, R. P. (1992). The self-regulation o f attitudes, intentions, and behavior.
Social Psychology Quarterly. 55. 178-204
Bass, B.M. (1985). StogdilPs Handbook o f Leadership. New York: The Free Press.
Bowen, D. E., & Schneider, B. (1988). Service marketing and management:
Implications for organizational behavior. In B.M. Staw & L.L. Cummings
(Eds.L Research in organizational behavior (Vol. 19, pp. 43-80). Greenwich,
CT; JA1 Press.
Burke, M. J., Borucki, C.C., & Hurley, A.E. (1992). Reconceptualizing
psychological climate in a retail service environment: A multiple-stakeholder
perspective. Journal of Applied Psychology. 77. 717-729.
Buzzell, R.D. & Gale, B.T. (1987). The PIMS Principles. New York: The Free
Press.
Cox, T. (1978). Stress. MD: University Park Press.
Dempsey, S.J., Gatti, J.R., Grinnell, D.J. & Cats-Baril, W.L. (1997). The use of
strategic performance variables as leading indicators in financial analysts’
forecasts. The Journal of Financial Statement Analysis. 61-79.
Gallup Organization. (1998). Building a stronger workplace. Unpublished.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
Gallup Organization. (1989). Dental practitioner perceiven Technical report.
Unpublished.
Gale, B. (1992). Monitoring customer satisfaction and market-perceived quality.
American Marketing Association Worth Repeating Series. 922CSO.
Chicago, American Marketing Association.
Gotlieb, J.B., Grewal, D., & Brown, S.W. (1994). Consumer satisfaction and
perceived quality: Complimentary or divergent constructs? Journal of
Applied Psychologv.79. 875-885.
Glisson, C. & Durick, M (1988). Predictors o f job satisfaction and organizational
commitment in human service organizations. Administrative Quarterly. 33.
0161-81.
Hallowell, R., Schlesinger, L. & Zomitsky, J. (1995). Internal service quality,
customer and job satisfaction; Linkages and implications for management.
Human Resource Planning 21-31.
Harter, J.K., & Creglow, A. (1998). A meta-analysis and utility analysis of the
relationship between core GWA employee perceptions and business
outcomes. In press.
Harris, J.H. & Arendt, L.A. (1998). Stress reduction and the small business:
Increasing employee and customer satisfaction. Sam Advanced Management
Journal 27-34.
Hartline, M.D. & Ferrell, O.C. (1996). The management o f customer-contact
service employees: An empirical investigation. The Journal o f Marketing. 60
(4) 52-70.
Helmer, F T. & Suver, J.D. (1988). Pictures o f performance: The key to improved
nursing productivity. Health Care Management Review. 13 (4) 65-70.
James, L.A., & James, L.R. (1989). Integrating work environment perceptions;
Explorations into the measurement o f meaning. Journal o f Applied
Psvchologv.74. 739-751.
Johnson, J. (1996). Linking employee perceptions o f service climate to customer
satisfaction. Personnel Psychology. 49. 831-847.
Joreskog, K.G., & Sorbom, D. (1993). LISREL8: Structural equation modeling
with the SIMLIS command language. Chicago, IL: Scientific Software
International, Inc.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
Judge, T.A. (1992). The dispositional perspective in human resources research.
Research in Personnel and Human Resources Management. 10 127-172.
Judge T.A. & Locke E.A. (1993). Effect o f dysfunctional thought processes on
subjective well-being and job satisfaction. Journal o f Applied Psychology.
73 f3) 475-490.
Klimoski, R.J. & Hayes, N.J. (1980). Leader behavior and subordinate motivation.
Personnel Psychology. 33 131 543-555.
Kaplan, R.S. & Norton, D.P. (1996). Linking the balanced scorecard to strategy.
California Management Review. 39. 53-79.
Koska, M.T. (1990). High-quality care and hospital profits: Is there a link?
Hospitals. 64 62-63.
Kouzes, J.W. & Posner, B.Z. (1990). Leadership Practices Inventory. San Diego,
C A University Associates, Inc.
Lazarus, R. S. (1991). Emotion and Adaptation. New York: Oxford University
Press.
McNeese-Smith, D. (1996). Increasing employee productivity, job satisfaction, and
organizational commitment. Hospital and Health Services Administration. 41
1-16.
Nahavandi, A. & Malekzadeh, A.R. (1993). Leader style in strategy and
organizational performance: An integrative framework. Journal of
Management Studies. 30 (31405-425.
Phillips, L.D., Chang, D.R., & Buzzell, R. (1983). Product quality, cost position and
business performance: A test o f some key hypotheses. Journal of Marketing.
£L 26-43.
Rucci, A.J., Kim, S.P., & Quinn, R.T. (1998). The employee-customer-profit chain
at Sears. Harvard Business Review 83-97.
Schlesinger, L.A., & Zomitsky, J. (1991). Job satisfaction, service capability, and
customer satisfaction: An examination o f linkages and management
implications. Human Resource Planning. 14. 141-150.
Schmit, M.J. & Allscheid, S.P. (1995). Employee attitudes and customer satisfaction:
Making a theoretical and empirical connection. Paper presented at the 1995
Society for Industrial Organizational Psychology Meeting, Chicago, IL.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
Schneider, B. (1987). The people make the place. Personnel Psychology. 40.
437-453.
Schneider, B. (1991). Service quality and profits: Can you have your cake and eat it,
too? Human Resource Planning. 14. 151-157.
Schneider, B., Ashworth, S.D., Higgs, A.C., Carr, L. (1996). Design, validity, and
use o f strategically focused employee attitude surveys. Personnel
Psychology. 49. 695-706.
Schneider, B., & Bowen, D.E. (1992). Personnel/human resources management in
the service sector. Research in Personnel and Human Resources
Management. 10. 1-30.
Schneider, B., Wheeler J.K., & Cox, J.F. (1992). A passion for service: Using
content analysis to explicate service climate themes. Journal o f Applied
Psvchologv.77. 705-716.
Taylor, J.K. (1978). Hospitals: Organizational variables, productivity and
efficiency- Ph.D. diss, Georgia State University - College of Business
Administration.
Tomow, W.W. & Biley, J.W. (1991). Service quality and management practices:
A look at employee attitudes, customer satisfaction, and bottom-line
consequences. Human Resource Planning. 14. 104-115.
Ulrich, D., Halbrook, R., Meder, D., Stuchlik, M. & Thorpe, S. (1991). Employee
and customer attachment: Synergies for competitive advantage. Human
Resource Planning. 14. 89-103.
Wiley, J.W. (1991). Customer satisfaction: A supportive work environment and its
financial cost. Human Resource Planning. 14.2. 117-127.
Zeithaml, V.A., Berry, L.L., & Parasuraman, A. (1996) The behavioral consequences
of service quality. Journal o f Marketing. 60. 2. 31-45.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
APPENDIX
DKHTAL PRACTITIONER PERCEXVER
TtShPlOctober 1989
301 SOUTH-68TH STREET. LINCOLN. NEBRASKA 6 8 5 1 0 • P.O. BOX 5 7 0 0 . LINCOLN. NEBRASKA 6 8 5 0 5 • [AOS) 4 8 9 -9 0 0 0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
introduction
The purpose of this document is to present the results cf a
research study conducted by SRI Gallup in conjunction with
numerous dental practitioners across the United States and Canada,
in an effort to continue to learn about attitudinal patterns
related to successful dentistry.
Underlying this study is a belief in talent--that consistently
recurring patterns of thought, feeling, and behavior are identi
fiable and measurable within individuals. These patterns (themes)
remain fairly consistent over the course of one's life and, when
accompanied with the correct stimulus, provide a higher potential
to achieve excellence. The underlying philosophy, then, involves
focusing on and studying patterns of excellence within talented
people.
It has been clearly documented at SRI Gallup that, when objec
tively interviewed, successful individuals within specific
positions (when asked the appropriate questions) respond differ
ently than do less successful individuals; thus, there is a belief
in finding and selecting the successful (talented) people and
placing them into positions where they can utilize their talents
maximally, resulting in greater efficiency and productivity within
an organization.
Before any selection instrument or process is adopted for use by
an organization, that instrument or process must go through a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
series of well-defined, psychometric, developmental steps. Each
phase of this research process must be guided by the standards of
the American Psychological Association (APA).
Many dental offices throughout the United States and Canada have
traditionally administered the Dentist Manager Perceiver. Dentist
Sales Perceiver, and Dentist Perceiver as part of the information
utilized in the selection of dental practitioners. The purpose of
this research study was to create an interview that would, in
abbreviated form, include the most essential elements from each of
the three dental perceivers.
Validity is an issue concerned with whether an instrument or
process measures what it is intended to measure. In the case of a
selection instrument, that instrument is presumed to measure the
relationship between talent in an individual and the subsequent
performance of that individual upon being hired. Mo interview or
selection process can ever be 100 percent accurate, yet accuracy
in selecting the right person to fit into a given environment can
increase significantly by the use of an objective process.
While validity cannot be proved absolutely, evidence of validity
does exist, and can be inferred, through a rigorous, scientific
process, such as the one described in this study. The purpose of
this analysis was to determine the degree to which the SRI Dental
Practitioner Perceiver exhibited content, construct, and
concurrent validity within the parameters of the sample studied.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
Content validity is concerned with, whether components within the
interview make sense to the respondents and appear to be
conceptually related to performance on the job? construct
validity, with the internal consistency-related issues and whether
components, identified as constructs, appear to be related to an
acceptable degree, yet also appear to be measuring rather unique
and independent psychological constructs; and concurrent validity,
with the degree to which the interview successfully distinguishes
excellent talent from mediocre talent within a group of already
"in-place" associates.
Research Questions
Research Question Wo. 1: Old the 8RX Dental Practitioner
Perceiver demonstrate an acceptable
degree of internal consistency?
Research Question Ho. 2: What was the relationship between
the SRX Dental Practitioner Per-
eeiver and the effectiveness and
productivity of dental practitioners
in this study?
T n«iJ-T -H ffl«»w *-»f--jo n /M S fc h O d O lO < T V
1. Interviews conducted between 1985 and 1989 with
46 dental practitioners from across the United. States
and Canada.
2. Hypotheses tested, utilizing objective and subjective
performance criteria.
3. Field interview configured and studied as to evidence
regarding its content, construct, and concurrent
validity properties.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
SRI used a success model for the development o f the SRI Dental
Practitioner Perceiver which involved studying many of the best
dentists.
The process was initiated by conducting a series of interviews
with dentists of varying talent levels. Forty-six (46) dentists
from across the United States and Canada were administered the
55-item Dentist Manager Perceiver, the 70-item Dentist Sales
Perceiver, and the 98-item Dentist Perceiver. Each interview was
conducted by certified interviewers from SRX Gallup. Interviews
were tape-recorded with permission of the participant for later
transcription and analysis. Each interview was conducted via
telephone in the same manner.
At this point, the content validity of the instrument was
assessed. Any item on which respondents were either confused or
did not understand the terminology was rewritten or deleted.
Through a careful, behavioral, and statistical analysis of the
responses given by the total sample of 46 dental practitioners, an
instrument was reconfigured to include 112 of the most predictive
and discriminatory items with respect to talent. An item was
defined as one interview question, along with its corresponding
predictive response. The 112 items were equally spread across
sixteen (16) life themes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
The f discovered ho characterize highly talented dental
practitioners were:
Focus Developer
Mission Delegator
Achiever Stimulator
Ego Drive Arranger
Courage Performance Orientation
Activation Sophistication
Relator Mastery
Empathy Ethics
Each item vas scored on a credit/no credit basis which quanti
tatively translates to a "1" or "0" scoring. Ninety-four (94) of
the 112 items in the interview were contained in the original
three perceiver instruments. The additional eighteen (18) items
were added, based on information in SRX's research data base,
indicating that, in past research, they had had predictive
capacity in a management context.
Results presented in this report represent the 94 items from the
original three instruments. Data was not available for the sample
with regard to the additional 18 items. Because of this, the
number of items representing each theme did vary from four to
seven for the research presented in this report. There are seven
items composing each theme for the perceiver that is to be used in
the field. As such, theme scores on the full perceiver, would be
the sum of the seven items composing each theme and could range
from 0 to 7 raw score points. The total interview score would be
the sum of all 16 themes and could range from 0 to 112 raw score
points. The interview developed was a structured process with,
objectively scored answers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
In addition to the various descriptive statistics utilized, the
Pearson Product Moment Coefficient of Correlation (PPM) was used
to determine the magnitude and direction of the linear relation
ship between variables within the interview. The PPM was a
statistic that could range from -1.00 to a +1.00, both extremes
indicating a perfect relationship between two variables. A
positive sign, or the absence of a sign, in front of the
coefficient would indicate that the relationship between the two
variables was direct; that is, high on one variable and high on
the second or low on one variable and low on the second. A
negative sign in front of the correlation would indicate that an
inverse relationship existed between the two variables; that is,
high on one variable and low on the second. Each correlation
coefficient generated was tested for statistical significance
which, consistent with international standards, was defined at the
0.05 level of probability. This means that the results, as
observed, if statistically significant, would have approximately a
95 percent chance of being due to a systematic observation of
variance within the sample, as opposed to a random occurrence.
Results
Research Question Wo. 1: Did the SRI Dental Practitioner
Pereeiver demonstrate an acceptable
degree of internal consistency?
The first analysis completed was an examination of the item
difficulties. Because of the scoring format— a dichotomous
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
scoring technique where "1" was assigned to a response predicrive
of success as a dentist and "0" was assigned to a response non-
predictive of success as a dentist— the item average or difficulty
translates directly to the percentage of respondents receiving
credit for that particular item. Within this sample, the average
item difficulty was .60, with a standard deviation of .17. The
median item difficulty was .61 (Q*.14). Seventy-nine (79) out of
the 94 items (84%) within the interview had item difficulties
ranging from .30 to .70. The range of item difficulties was from
.20 to .96. Complete item difficulty frequencies are presented in
Table 1.
Table 1
Frequency of Item Difficulties
(ns46)
Ranee Freauencv %
1.00 - .90 3 3.2%
.89 - .80 9 9.6
.79 - .70 16 17.0
.69 - .60 21 22.3
.59 - .50 16 17.0
.49 - .40 17 18.1
.39 - .30 9 9.6
.29 - .20 2 2.1
.19 - .10 1 1.1
.09 - .00 0 0.0
The next analysis completed was an examination of the theme inter-
correlational patterns. Each theme was correlated to each other
theme to gain some understanding of the relationships between
themes. Of the 120 correlation coefficients generated, 119 (99%)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
were in the positive direction; and 83 (69%) were of statistically
significant magnitude. Of the 16 themes, all were significantly
correlated with at least four other themes.
The most consistent underlying constructs within the interview
were Focus, Stimulator, and Relator. Focus and Stimulator were
significantly correlated with all 15 of the other themes. Relator
was significantly correlated with 14 of the other themes. The
magnitude and direction of theme intercorrelations, as summarized
in Table 2 , suggest an excellent pattern where themes were
generally related to an acceptable degree, yet did also appear to
be measuring rather unique and independent constructs.
Each theme was correlated to total score to gain some under
standing of its contribution to the overall measure of the
dentists' talent. This correlation was adjusted downward, due to
the natural degree of inflation inherent whenever a subscore
(theme score) is correlated to a score (total score).
After adjustment for this inflation, all 16 of the themes were
positively correlated to total score and 15 of the 16 themes were
significantly correlated to total score. Themes most highly
predictive of total score were Focus, Relator, Stimulator, and Ego
Drive. The average total score (94 items) for the entire sample
was 56.94 (s*14.61) raw score points. Table 3 summarizes the
theme and total score statistics for the entire sample.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
« «
i 9 i n m
aeirt
I «0O^O
I AKIM
O
m
* *
m
« i « « « « « «
I «M)AA e*9 9 9
I « a a ia « ♦« a
•I M
y A
• §;
« s s
« « « « « € « « «
t tf) A A ^dkNM
I «MlA«
I Ifllf O " * 9 e *
I ANA AAN fl N W«
s
H
I
I
I AO(I n«A«lAnA« IA N IA A
I r» « • A * «
I nmaaOOaaa
I «9iA9
««««««««««« « « « «
i o»A)O(0M«Nafi« ia • m a
I H i C * 9 9 9 e * 9 9 9 9 9 9 9 9 m
c o
o -*
e
o
c
o
9 -*
S 9
O «
o-*
k * X
> ^
• Q
£ 0
o 9
< u
y
y y o
9 0**
o * * y > » a*» m
99 0 £ 0 9m
9 > * * * * < — 9 9
W < ~ 9 9 9 9 9
3 * * m a > r n m
0 0 9 9 0 9**
o < x w a O 9
c
9 S
9 y
e o
« % •
y y
y •
< 0i
e 4 *
• •
y x
o a
o
X
> •
y a
• o
9 * *
X u
o
V
a
9 m 9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Table 3
Theme and Total score
(n=46)
Statistics
Original Adjusted
No. of Standard Correlation Correlation
Theme Items Averaoe Deviation To Total Score T o T o t a l S c o r e
Focus 4 2.39 1.16 .83* . 8 0 *
Mission 7 4.76 1.39 .70* . 6 4 *
Achiever 4 1.89 1.10 .35* .23
Ego Drive 7 3.94 1.54 .75* . 7 0 *
Courage 7 4.13 1.49 .47* . 3 9 *
Activation 6 2.98 1.39 .53* . 4 5 *
Relator 7 4.17 1.97 .84* -79*
Empathy 7 4.46 1.57 .63* .56*
Developer 6 3.24 1.52 .64* .57*
Delegator 6 3.96 1.45 .69* .63*
Stimulator 5 3.59 1.19 .79* .75*
Arranger 6 3.89 1.40 .68* .63*
Performance
Orientation 6 3.04 1.55 .70* .64*
Sophistication 7 5.04 1.30 .61* .55*
Mastery 5 3.35 1.34 .61* .54*
Ethics 4 2.11 1.10 .46* .40*
TOTAL SCORE 94 56.94 14.61 — —
*p<.05
The overall reliability index, using the KR-20 Formula was . 9 1
which was judged to be excellent. The KR-20 Formula results in a
lower-bound estimate of internal consistency-related reliability.
Additionally, the standard error of measurement for the interview
was 4.38; in other words, given the total score evaluation of the
individual, the total score assessment would vary by no more than,
plus or minus, 4.38 raw score points approximately two-thirds of
the time; and the total score assessment would vary by no more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
than, plus or minus, 8.76 raw score points approximately
95 percent of the time.
From preceding analysis, it does appear to be justified to
conclude that the SRI Dental Practitioner Perceiver possesses an
acceptable degree of internal consistency. This provides a very
positive portion of the evidence necessary to determine the degree
of construct validity of the interview.
Research Question Wo. 2 : What was the relationship between
the SRI Dental Practitioner Per-
ceiver and the effectiveness and
productivity of dental practitioners
in this study?
Recall that two performance measures were available and considered
as estimates of the talent and/or productivity of dental practi
tioners in the study. One criterion variable was a 1988 office
production figure. The other variable was a dichotomous
effectiveness grouping. The 29 dental practitioners that were
judged to be very effective were placed into a Study group with
the remaining 17 placed into a Contrast group. The purpose of the
Contrast group was to provide a comparison whereby excellent
talent could be compared to mediocre talent.
In order to study the ability of the interview in discriminating
between talent levels of dental practitioners, themes and total
score were correlated to effectiveness group membership and 1988
productivity. For correlational purposes, a "2" was assigned to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
each individual in the Study group, and a "1" was assigned to
those in the Contrast group. Theme scores and total score were
then correlated to this variable and the raw 1988 office
production figures.
A positive and significant correlation of any component to
effectiveness group membership or 1988 productivity would indicate
that those perceived as more effective and/or productive tended to
have higher scores on that interview variable than did their
colleagues who were perceived as less effective and/or productive.
It is encouraging that each of the 16 themes was positively
correlated to effectiveness group membership and 1988 produc
tivity. - Fourteen of the 16 themes were positively and signifi
cantly predictive of effectiveness group membership. Thirteen of
the 16 themes were positively and significantly predictive of 1988
productivity.
Themes most highly predictive of effectiveness group membership
were Relator, Stimulator, and Focus. The themes most highly
predictive of 1988 productivity were Ego Drive, Performance
Orientation, Mission, and Stimulator.
Clearly, the most important statistic in this analysis was the
relationship between the effectiveness grouping variable and 1988
productivity with total score on the interview. The correlation
between the effectiveness group membership and total score on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
interview was .59, statistically significant well beyond the 0.05
level of probability- Similarly, the correlation between 19SS
productivity and total score on the interview was .59, again
highly significant. In fact, the relationship between total score
on the interview and the criterion variables in the study would
have approximately one chance out of ten thousand of being due to
a random occurrence within the sample.
The Study group of dentists had an average of 63.45 total raw
score points on the interview compared to the Contrast group with
an average total score of 45.82. From a psychometric perspective,
this difference was very large and highly significant. Table 4
summarizes the theme and total score statistics by effectiveness
grouping and the resulting correlations to the criterion
variables.
In a further attempt to understand the differences with regard to
the total score variable and the criterion variables, a quartile
grouping was derived. This quartile grouping was based on total
score; for example, the top 25 percent dental practitioners, with
regard to total score, had to have a total score between 67 and
85 raw score points. To be in the bottom 25 percent, the scores
should have been between 27 and 45 raw score points.
When looking at the composition of the top quartile group, it is
apparent that this top quartile group was composed of 100 percent
effective, or Study Group, members and 0 percent Contrast group
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
Table 4
Thame and Total Seoro Statistics
By Effectiveness Group aad 1988 Productivity
Contrast Study Corre-
fn-17) fn-29) Corre- lation
Standard Standard lation To 19 SS
Theme Average Deviation Average Deviation To Group Produc.
Focus 1.65 .93 2.83 1.07 .50* .44*
Mission 4.24 1.44 5.07 1.28 .29* . 47*
Achiever 1.41 .71 2.17 1.20 .34* .32*
Ego Drive 3.12 1.32 4.41 1.48 .41* .43*
Courage 3.29 1.26 4.62 1.40 .44* .33*
Activation 2.35 1.37 3.35 1.29 .35* .32*
Relator 2.82 1.88 4.97 1.57 .53* .36*
Empathy 3.82 1.47 4.83 1.54 .31* .37*
Developer 2.29 1.31 3.79 1.37 .48* .32*
Delegator 3.35 1.62 4.31 1.23 .32* .33*
Stimulator 2.82 1.29 4.03 .87 .50* .47*
Arranger 3.12 1.45 4.35 1.17 .43* .38*
Performance
Orientation 2.24 1.44 3.52 1.43 .40* .48*
Sophistication 4.59 1.06 5.31 1.37 .27
.29
Mastery 3.06 1.39 3.52 1.30 .17 .27
Ethics 1.65 .93 2.38 1.12 .32* .26
TOTAL SCORE 45.82 11.19 63.45 12.35 .59* .59*
*
p<.05
members. For the bottom quartile group, the results were almost
reversed. This bottom quartile group was composed of 75 percent
Contrast (less effective) members and three (25%) study
(effective) group members. Table 5 summarizes the total score
quartile group by the two effectiveness groups.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Table S
Total Score Quartile Group
By EfftetiTMtu Group
Contrast Group Study Group
ouartile Grouo
— n-
ifisncflesstivBl
fEffective1
Top Quartile
(Total Score 67-85)
12 o
o
*
12 (100%)
Bottom Quartile
(Total Score 27-45)
12 9 (75%) 3 (25%)
with regard to 1988 productivity, the average productivity for the
top quartile group was $590,454.50, almost double that of the
bottom quartile average. The average 1988 productivity for the
bottom quartile group was $292,245.40. Again, the difference
between the top and bottom quartiles with respect to 1988
productivity was highly significant (p<.05) . Table 6 summarizes
1988 productivity by the top and bottom quartile groups.
Table 6
1988 Productivity For Top aad Bottom
Total Seore Quartile Groups
Ouartile Grpup
1988 Productivity
n Averaoe Deviation
Top Quartile
(Total Score 67-85)
11 $590,454.50 148,227.00
Bottom Quartile
(Total Score 27-45)
11 $292,245.40 91,185.64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
From 'the data presented in Tables 4, 5, and 6, one can see that
there was excellent discrimination with regard to the interviev
and effectiveness and/or productivity. As such, under the
parameters of this study, the interview and process was judged to
be acceptable with regard to concurrent validity.
Discussion
The central question under investigation in this research was to
determine whether an interview- could be developed that would
significantly distinguish highly effective and productive
(talented) dental practitioners from those who were less effective
and less productive. There was rather clear evidence to indicate
that the answer to this queistion is YZfl. The interview develop
ment process included a very thorough analysis of items
encompassing the Dentist Manager Perceiver, Dentist Sales
Perceiver, and Dentist Perceiver. The resulting correlation of
the reconfigured Dental Practitioner Perceiver with effectiveness
grouping and 1988 productivity was statistically significant
(p<.05).
The item difficulties were judged to be acceptable, as vere the
theme intercorrelational patterns. Themes did appear to be
measuring a similar underlying construct, yet at the same time did
appear to be measuring unique and independent constructs.
Moreover, each theme provided a positive contribution toward the
prediction of talent for individuals in this study. Additionally,
the KR-20 internal consistency-related reliability estimate was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
.91. Thus, the construct validity and internal consistency cf the
instrument and process for dentists was judged to be excellent.
Content validity was acceptable in that there were no iters
retained in the final version of the interview that respondents
did not understand or that they found to be confusing. This was
assessed at the time of the interviews.
The concurrent validity of the process was judged to be excellent,
given that the correlation of total score to 1988 office
productivity was .59 and highly significant. Additionally, the
correlation of total score to effectiveness group membership was
also .59 and highly significant. Furthermore, each theme corre*
lated in the positive direction to the criterion variables. The
vast majority of the themes were significantly correlated to the
criterion variables.
In a statistical sense, total score on the interview is the most
reliable and valid predictor of performance for the dental
practitioners within this study. Theme scores can be very helpful
in terms of gaining a more specific understanding of the
individual. As such, they can be implemented for management and
developmental purposes, although theme scores should not be used
exclusively for overall prediction of performance.
If should be mentioned at this point that no one study ever
definitively establishes the existence of validity in an absolute
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
sense. By definition, validity is a function of the time and
context of the research study conducted; and its establishment
must remain an ongoing and continuous process - The results cf
this study can be viewed as initial evidence of the validity of
the SRI Dental Practitioner Perceiver.
The next phase in this research continuum is to address the
predictive validity of the interview and process. This will be
studied by SRI Gallup in conjunction with dental practitioner
offices across the United States and Canada, after an appropriate
number of dental practitioners have been hired, using, in part,
the SRI Dental Practitioner Perceiver, and when objective
performance criteria become available On these hired individuals.
It is also important, after candidates have been hired, to allow
enough time to adequately judge the performance of these
individuals.
Generally, the sample size needed for such a study requires
approximately 100 dental practitioners that exhibit performance
across twelve or more months. SRI Gallup will begin immediately
to make plans for completing this predictive validity study and
for tracking performance of dental practitioners on an ongoing and
continuous basis.
It is important to note that the results of this study did offer
some preliminary indications of predictive validity, in that, many
of the dental practitioners were interviewed several years before
1988 productivity was achieved.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
Within the overall context of this study, there is justification
to conclude that the interview would be general izable in the
predictive sense with which it will be used by dental offices
across the United States and Canada as one part of the overall
assessment of the talent level of prospective dental practitioners.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An examination of U.S. Army Aviation Warrant Officers' personality and promotability

PDF

Assessment of the quality of institutes of higher education: Correlational analysis of U.S. News and World Report's university and college rankings and institution size, age, selectivity, graduat...

PDF

Correlates of job satisfaction among California school principals

PDF

An evaluation of an Air Force formal mentoring program

PDF

Correlates of job satisfaction among school superintendents

PDF

Closing the leadership gap

PDF

A structural model of self-efficacy, goal orientation, worry, self-regulated learning, and high-stakes mathematics achievement

PDF

A comparison of analogical and quantitative methods for teaching mechanical functions in science education

PDF

Comparison of the predictive validity of a written test, an integrity test, a conscientiousness questionnaire, a structured behavioral interview and a personality inventory in the assessment of j...

PDF

Examining the relationship between writing self -efficacy, writing performance and general achievement for third graders

PDF

Collective efficacy, anxiety, creativity/innovation and work performance at the team level

PDF

Career advancement in managerial hierarchies in the United States firms: A multi-theoretic model and empirical tests of the determinants of managerial promotion

PDF

Effects of the STAR testing program on teachers and the curriculum

PDF

Current and future managerial competency requirements for manufacturing, assembly, and /or material processing functions

PDF

A structural model of problem-solving ability, self-efficacy, effort, worry, and achievement in calculus

PDF

Communications medium and product class: Their effect on negotiation outcome

PDF

Faculty perception of stress in a non-profit business and technology college

PDF

Approaches to inference in ordinal multiple regression

PDF

Advance base construction by civilian contractors in war zones

PDF

An evolution in listening: An analytical and critical study of structural, acoustic, and phenomenal aspects of selected works by Pauline Oliveros

Asset Metadata

Creator McCashland, Courtney R (author)

Core Title Core components of the service climate: Linkages to customer satisfaction and profitability

Degree Doctor of Education

Degree Program Education

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag business administration, management,education, business,OAI-PMH Harvest,psychology, industrial

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Hocevar, Dennis (committee chair), [illegible] (committee member), Baker, Robert (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-565603

Unique identifier UC11351282

Identifier 9955491.pdf (filename),usctheses-c17-565603 (legacy record id)

Legacy Identifier 9955491.pdf

Dmrecord 565603

Document Type Dissertation

Rights McCashland, Courtney R.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA