LOGISTICS SERVICE PERFORMANCE:

ESTIMATING ITS INFLUENCE ON MARKET SHARE

by

Theodore P. Stank, Ph.D.

Associate Professor of Logistics and Supply Chain Management
Michigan State University

Thomas J. Goldsby, Ph.D.

Assistant Professor of Marketing and Logistics
The Ohio State University

Shawnee K. Vickery, Ph.D.

Professor of Operations Management
Michigan State University

Katrina Savitskie
Doctoral Candidate
Michigan State University

With correspondence, please contact:

Theodore P. Stank
Associate Professor of Logistics and Supply Chain Management
Dept. of Marketing and Supply Chain Management
Eli Broad College of Business
Michigan State University
N370 North Business Complex
East Lansing, MI 48824-1122
Ph. (517) 353-6381
Fax: (517) 432-1112

1
 ABOUT THE AUTHORS

Theodore P. Stank (Ph.D. in Marketing and Distribution from The University of Georgia) is
Associate Professor of Logistics at Michigan State University. He is co-author of 21st Century
Logistics: Making Supply Chain Integration a Reality, and has published articles in the areas of
logistics strategy, customer relevance, and integration in various journals including the Business
Horizons, Journal of Business Logistics, Journal of Operations Management, Supply Chain
Management Review, and Transportation Journal.

Thomas J. Goldsby (Ph.D. in Marketing and Logistics from Michigan State University) is
Assistant Professor of Logistics at Ohio State University. His research interests focus on
logistics customer service and supply chain integration. He also has interest in the development
and implementation of environmental, or “green,” business practices. He has published articles
in academic and professional journals, such as the Journal of Business Logistics, International
Journal of Logistics Management, Supply Chain Management Review, and Journal of
Operations Management.

Shawnee K. Vickery (Ph.D. in Business Administration from the University of South Carolina)
is Professor of Operations Management at Michigan State University. Her research has been
published in Decision Sciences, Journal of Operations Management, The European Journal of
Operational Research, The International Journal of Production Research, Journal of Product
Innovation Management, Production and Inventory Management, The Journal of Supply Chain
Management, International Journal of Physical Distribution and Logistics Management,
International Journal of Operations and Production Management, and Journal of Business
Logistics.

Katrina Savitskie is a fourth year doctoral candidate at Michigan State University where she is
studying Marketing, Logistics, and International Business. Her research interests include
information technology and its role in the supply chain. She has been published in the Journal
of International Marketing and numerous conference proceedings.

2
 LOGISTICS SERVICE PERFORMANCE:
ESTIMATING ITS INFLUENCE ON MARKET SHARE

3
 Logistics creates value by accommodating customers’ delivery requirements in a cost

effective manner. Logistics service performance, therefore, assesses a provider’s ability to

consistently deliver requested products within the requested delivery timeframe at an acceptable

cost (Bowesox, Closs, and Cooper 2002). Logistical services, a unique subset of industrial

services that span the boundaries between suppliers and customers, have become increasingly

important to successful supply chain operations. Logisticians understand that these activities

constitute the very essence of their business. Communicating the importance of logistical

activities to other functional activities, as well as to corporate officers, has been a difficult feat.

Professor Donald Bowersox, speaking at the Council of Logistics Management Annual

Conference in Toronto in 1999, described establishing the link between functional logistics

performance and overall firm performance as our discipline’s equivalent to finding a cure for

cancer.

Definitive empirical results that link improvements in logistics performance to overall

firm performance have been difficult to achieve. Some progress, however, has been made.

Recently, Daugherty, Stank, and Ellinger (1998) conducted an in-depth assessment of the

relationships among logistics service performance and customer satisfaction, loyalty and market

share in an industrial setting. Their findings indicated that high levels of logistics service are

indirectly related to market share through satisfaction and loyalty.

4
 The current research seeks to expand the knowledge of logistics service performance and

its link to overall business performance in four significant ways. First, a more complex

operationalization of logistics service performance consisting of three distinct constructs is used

to capture logistical service in a comprehensive, yet parsimonious manner. Second, the model is

applied in a unique industrial setting, the third party logistics (3PL) industry. Third, a secondary

source of objective market share data (from Armstrong and Associates 1999) is used to anchor

respondents’ assessments of market share. Fourth, relationships among core dimensions of

logistics service performance, satisfaction, loyalty, and market share are tested using structural

equations modeling (SEM) in contrast to regression in the Daugherty et al. research to

simultaneously assess the structural paths among the constructs

The paper is organized as follows. First, logistics operational, cost, and relational

performance are defined. A synthesis of the logistics, operations, and marketing literature forms

the basis of a conceptual model relating the operational, cost, and relational determinants of

logistics service performance to customer satisfaction, loyalty, and market share. The methods

used to collect data and test the model are described next, followed by a discussion of results.

Finally, the managerial implications of the findings and directions for future research are

examined.

THEORETICAL BACKGROUND

Logistics service performance research can be broadly segmented into work oriented

toward 1) service dimensions and 2) service outcomes. This section reviews relevant work in

these two areas, leading to a discussion of the research model and hypotheses.

Logistics Service Performance Dimensions

5
 Logistics operational and relational performance construct definitions were fostered by

the dimensions of overall service quality originally defined by Parasuraman, Zeithaml, and Berry

(1985; 1988). Parasuraman et al.’s research identified five broad dimensions of service quality:

(1) reliability (the ability to perform the promised service dependably and accurately); (2)

responsiveness (the willingness to help customers and to provide prompt service); (3) assurance

(the knowledge and courtesy of employees and the ability to convey trust and confidence), (4)

empathy (the provision of caring, individualized attention to customers), and (5) tangibles (the

appearance of physical facilities, equipment, personnel, and communications materials). The

dimensions were operationalized in a measurement scale called SERVQUAL that assessed

service quality as the gap between pre-transaction customer expectations of quality and their

perceptions of service quality after consumption.

A substantial amount of research has documented the relative importance of service quality

dimensions to end-use consumers, i.e., customers of retail services (see for example Babakus and

Mangold 1989; Berry and Parasuraman 1991; Berry 1995; Bojanic 1991; Carman 1990; Crompton and

Mackay 1989; Johnson, Dotson, and Dunlop 1988; Parasuraman, Berry, and Zeithaml 1991; Zeithaml,

Parasuraman, and Berry 1990; and Zeithaml 2000).

Empirical evidence suggests that the proposed delineation of five components is not consistent

when compared across different types of service industries (Babakus and Boller 1992; Carman 1990;

Cronin and Taylor 1992; and Finn and Lamb 1991). In particular, researchers have had difficulty

replicating the SERVQUAL dimensions in industrial service contexts (Bienstock, Mentzer, and Bird

1997). One possible explanation for the disparate results is that the dimensions of service quality vary

from one industry to the next. This is especially true for industrial services like logistics that focus on

tangible things directed toward physical objects versus intangible actions directed toward thoughts and

6
attitudes (Lovelock, 1983). Accordingly, Brown, Churchill, and Peter (1993) advise researchers to

carefully assess which issues are important to service quality in their particular situations and to modify

the SERVQUAL scale accordingly. Another possible explanation is that a more generic conceptual

scheme has yet to be identified.

Recently, Stank, Goldsby, and Vickery (1999) used the SERVQUAL dimensions as a

starting point for producing a more generic conceptualization of logistics service performance, a

unique example of industrial service. Their research identified two core dimensions of logistical

service: (1) operational performance, and (2) relational performance. Operational performance

consisted of two key elements -- reliability (which captured the dependability and accuracy of a

service a la Parasuraman et al. (1985) and related to the consistent quality aspect of operational

performance) and price. Parasuraman et al.’s responsiveness, assurance, and empathy attributes

were encompassed in relational performance, the second dimension of service performance in

their study.

While Parasuraman et al.’s original research considered the price or cost of service to be

part of communication, Stank et al. included cost as a key aspect of logistics operational

performance in their fast food industry research. In the current research, cost is conceptualized

as a unique, third dimension of logistics service performance, separate and distinct from the

operational and relational components of service. Literature in manufacturing and service

operations provides substantial support for treating price (or cost) as a separate dimension of

service performance (Hayes and Wheelwright 1984; Krajewski and Ritzman 1987; Hill 1989;

Cleveland, Schroeder, and Anderson 1989; Roth and Miller 1990; Wood, Ritzman and Sharma

1990; Ferdows and De Meyer 1990; Roth and Van der Velde 1991). Additional support for this

approach is provided by Porter’s generic strategies scheme in which cost leadership (in contrast

7
with quality differentiation, for example) appears as a distinct, yet viable, path for attaining

competitive advantage (Porter 1980; 1985).

Aside from theoretical considerations, there are two practical advantages of modeling

cost (or price) separately from logistics operational performance or consistent quality. First, it

allows us to determine the individual effects of these entities on the endogenous variables of the

research model and, second, it enables us to examine the relationship of antecedent variables to

cost.

Outcomes of Logistics Service Performance

Research frequently has sought to determine the relationship between service

performance and perceptual factors such as customer satisfaction and repurchase intentions

(Zeithaml 2000). Customer satisfaction may be considered a cumulative evaluation based on the

total purchase and consumption experience with a good or service over time (Fornell 1992;

Fournier and Mick 1999). The evaluation is based upon post purchase confirmation or

disconfirmation of the buyer’s preconceived expectations of product or service standards. High

customer satisfaction has been linked to improvements in a firm’s economic returns, including

market share and profitability (Anderson, Fornell, and Lehmann 1994; Crosby, Evans, and

Cowles 1990; Leuthesser and Kohli 1995; and Reichheld and Sasser 1990).

A significant number of findings strongly support the notion that logistics service quality

improvements can increase customer satisfaction (Daugherty, Stank, and Ellinger 1998; Innis

and LaLonde 1994; and Leuthesser and Kohli 1995). Operational elements of logistics service

related to product availability, product condition, delivery reliability, and delivery speed, as well

as relational elements such as communications and responsiveness have been shown to have a

8
positive relationship with customer satisfaction (Daugherty, Stank, and Ellinger 1998; Innis and

LaLonde 1994; and Stank, Goldsby, and Vickery 1999).

The literature suggests a strong link between customer satisfaction and loyalty (Anderson

and Sullivan 1993; Bearden and Teel 1983; Fornell 1992; Innis and La Londe 1994; and Jones

and Sasser 1995). Customer loyalty is conceptualized as having both behavioral and

perceptual/psychological components (Anderson and Sullivan 1993; Jacoby and Kyner 1973).

Repeat purchase behavior stemming from positive assessments of product and/or service

offerings has often been used as a measure of customer loyalty (Leuthesser and Kohli 1995;

Sharma and Lambert 1990). Repeat purchase behavior, however, may result from a number of

underlying factors that are not related to a customer’s favorable impression of a provider’s

product or service offering. Jacoby and Kyner (1973) assert that loyalty, in contrast to simple

repurchase behavior, is the result of a psychological decision-making process that results in a

nonrandom, behavioral purchase response with respect to one alternative out of a set of such

alternatives expressed over time by a decision-making unit. Repeat industrial purchases

representing increased resource expenditure between parties in industrial buyer-seller

relationships are likely to result from an enduring desire on the part of the customer to maintain

an important, valued relationship with a supplier (Cronin and Morris 1989; Morgan and Hunt

1994). Consistent with Morgan and Hunt (1994), we define loyalty as a long-term commitment

to repurchase involving both a favorable cognitive attitude toward the selling firm and repeated

patronage.

The literature hints at a strong link between customer loyalty and market share.

Reichheld and Sasser (1990) claim that loyal customers improve market share by purchasing a

greater volume and variety of product. In addition, loyal customers demonstrate greater

9
resistance to counter persuasion and negative word-of-mouth (Rust and Zahorik 1993). Innis and

LaLonde (1994) confirmed the relationship in logistics, linking customer service, attitudes,

satisfaction, and repurchase intentions. In their work, they refer to “repurchase intentions” as a

“proxy for market share” (p. 19).

Dick and Basu (1994) introduced a model that illustrated a linkage between consumer

loyalty and specific consequences that may contribute to market share. Similarly, Daugherty,

Stank, and Ellinger (1998) concluded that logistics operational elements related to product

availability, product condition, delivery reliability, and delivery speed, as well as relational

elements such as communications and responsiveness were found to positively influence

customer satisfaction and purchasing patterns that contribute to market share growth. Their

research in the personal products industry, however, concludes that loyalty predicts market share

variance only when operating with satisfaction (p. 44). The authors note that future research

should attempt to further capture the interrelationships that exist among logistics service,

customer satisfaction, loyalty, and market share given the lack of support for the hypothesis

solely linking loyalty to market share.

Conceptual Model and Research Hypotheses

This research seeks to expand upon past research to further understand the relationships

among logistics service performance, satisfaction, loyalty, and firm market share. Importantly, it

does so in a model that provides a comprehensive picture of the relationships among key

logistics performance elements. The research also explicitly examines the linkage between

loyalty and a measure of market share that is anchored by actual outcomes.

The model is tested in the context of logistical services provided by 3PL providers. The

3PL industry was considered an excellent setting for examining the interrelationships of interest

10
since the evaluation of the service provider (3PL) is independent of the manufacturer/shipper of

the materials or goods. In essence, the provision and quality of the logistics service is distinct

from the materials or goods themselves, reducing the potential for a confounding or “halo”

effect. In other words, the nature of the relationship between the service provider and the

customer is based solely on the nature of the service provision and not on the quality of the

materials or goods. In addition, there are usually multiple logistics service alternatives available

to shippers in the U.S. market. Therefore, customer loyalty is typically based upon the quality of

logistical service rather than a lack of alternatives (i.e., “captive commitment”). For these

reasons, the customers of third-party logistics service providers served as the target population

for the research.

The conceptual model is presented in Figure 1. The theoretical foundations for the

relationships depicted in this figure are summarized below based upon the prior review of

relevant literature.

------------------------------------------------------------------------------------------------------------------
***take in Figure 1 approximately here***
------------------------------------------------------------------------------------------------------------------

In Figure 1, customers’ perceptions of logistics relational performance are portrayed as

the antecedent of logistics operational and cost performance. Stank, Goldsby, and Vickery

(1999) present empirical evidence to suggest that relational performance is an antecedent of

operational performance (which included a cost component). Their research indicates that

creating strong relationships with important customers allows select service firms to achieve

sustainable advantage by tailoring logistical operational offerings to the needs of each customer.

Such firms go beyond logistics basics by developing a meaningful understanding of customers’

needs supported by flexible processes that enable customized solutions. The ability to customize

11
logistics operational services with a high level of relational performance represents an “order

winning” combination. It becomes significantly more difficult for rivals to intercede in supplier-

customer relations once they reach this level of intimacy. In the current research, relational

performance encompasses attributes such as responsiveness, assurance, and empathy for

customer needs. Measures include assessments by customers’ of whether the service supplier

“knows your needs well,” “cooperates with you to help do the job well,” and “makes

recommendations for continuous improvement on an on-going basis.” Conclusions regarding

intimacy level, therefore, are based upon such assessments rather than on operational elements

such as information system linkages.

In the Stank et al. study, logistics cost performance was considered to be an element of

operational performance. In the current research, however, cost performance was decoupled

from operational performance since it is considered to be a third, unique dimension of service

performance. While the literature provides little guidance on the relationship between relational

performance and cost performance, it is reasonable to extrapolate a relationship similar to that

found between relational performance and operational performance in Stank et al. It also seems

reasonable to surmise that a firm’s relationships with customers might yield benefits similar to

those resulting from a firm’s relationships with suppliers. Scannell, Vickery, and Droge (2000)

found a positive and highly significant relationship between a firm’s use of supplier partnerships

(a relational item) and cost performance. Thus:

H1: Logistics relational performance has a positive effect on operational performance.

H2: Logistics relational performance has a positive effect on cost performance.

Empirical studies in operations and marketing provide considerable support for links

between service performance and customer satisfaction (Youngdahl and Kellogg 1997; Crosby,

12
Evans, and Cowles 1990; Cronin and Taylor 1992; and Leuthesser and Kohli 1995). Relational

performance provides service suppliers with enhanced insight regarding customer needs and

wants. Upon learning of these needs and wants, the service provider can focus on operational

means of meeting them at the lowest cost to the customer as possible. Previous research on

logistics service performance provides some indication that these relationships hold in a

logistical service context (Innis and La Londe 1994; Daugherty, Stank, and Ellinger 1998; Stank,

Goldsby, and Vickery 1999). The following hypotheses are based upon the findings from this

stream of logistics research:

H3: Logistics relational performance has a positive effect on customer satisfaction.

H4: Logistics operational performance has a positive effect on customer satisfaction.

H5: Logistics cost performance has a positive effect on customer satisfaction.

Customer satisfaction with service capabilities has been shown to have significant and

positive impact on cognitive attitudes and repurchase intentions in multiple business settings

(Anderson, Fornell, and Lehmann 1994; Cronin and Morris 1989; and Youngdahl and Kellogg

1997). Oliva, Oliver, and MacMillan (1992), for example, found that when satisfaction

increased above a critical level, repeat purchases increased rapidly. An equal decline in repeat

purchases occurred when satisfaction fell below a critical threshold. In logistics, the relationship

between customer satisfaction and loyalty has also been strongly supported (Innis and La Londe

1994; Daugherty, Stank, and Ellinger 1998; Stank, Goldsby, and Vickery 1999). These findings

provide the theoretical basis for our next hypothesis:

H6: Customer satisfaction has a positive effect on customer loyalty.

13
 The literature supports a strong link between customer loyalty and market share.

Loyal customers improve market share by purchasing a greater volume and variety of

products (Reichheld and Sasser 1990). Loyalty increases repurchase behavior because

loyal customers demonstrate greater resistance to counter persuasion and negative word-

of-mouth (Dick and Basu 1994; Rust and Zahorik 1993). They often are less price

sensitive, decreasing risk of defection due to competitive price undercutting. It follows

that firms that successfully prevent defection enjoy a lasting, stronger market presence

than those that incur frequent defections. Hence, our last hypothesis:

H7: Customer loyalty has a positive effect on market share.

Note that the model postulates that logistics service performance is linked with

market share through its direct relationship with customer satisfaction and its indirect

relationship (via customer satisfaction) with loyalty.

RESEARCH METHOD

This section reviews the research design. Discussion of the sample design is followed by

descriptions of the measurement scales. Results of the measurement model assessment then

precede a presentation of hypothesis tests.

Sample Design

The research model calls for data requirements from both the service provider and

customer in the supplier-buyer relationship. Data from the service provider were necessary for

the market share assessment. The providers’ customers then provided responses to measures for

the remaining five constructs. The sampling method proceeded with a two-step “snowball” or

hierarchical sampling approach, where the 3PL respondents were first contacted and asked to

identify prospective respondents for the customer sample. Snowball sampling uses a procedure

14
in which initial respondents are selected and additional respondents are then obtained from

referrals or by other information provided by the initial respondents (Green and Tull 1978). The

snowball approach was deemed most appropriate given the inherent difficulty of identifying

extensive users of outsourced logistics services. This basic sampling approach has been used to

yield sample frames that may be difficult to otherwise garner participation (see for example,

Daugherty, Stank, and Ellinger 1998). Statistics referenced in Tables 2b and 2c illustrate the

diverse composition of the customer sample.

The population for the 3PL dataset included all providers listed in Armstrong's Guide To

Third Party Logistics Services Providers (1999) that offered comprehensive logistics services

across the domestic U.S. Firms that provided comprehensive services (either on an asset or non-

asset basis) were preferred because they serve as “one-stop shops” for logistical services and are

ultimately responsible for the quality of customer service rendered. This requirement narrowed

the population to 68 firms. Of these 68, two firms were acquired by other 3PLs during the data

collection timeframe and were dropped from consideration, reducing the number of 3PLs to 66

firms.

The top executive (president or CEO) of each 3PL was contacted by telephone to

determine his or her willingness to respond to the questionnaire and to ascertain the appropriate

contact for the information being sought. Letters containing the Internet address of a web-based

questionnaire were mailed or e-mailed (depending upon respondent preference) to the 3PL

executive to further encourage participation in the research. The survey instrument requested

information regarding the 3PL’s basic service characteristics, the executive’s perceptions of the

firm’s own service performance, and assessments of the firm’s market performance for the

previous fiscal year. In communications by telephone and letter, the executive was asked to

15
direct the survey to individuals best suited to complete the survey’s various sections. The CEOs

indicated in many instances that their marketing or finance executives would be best suited to

answer the financial performance aspects of the survey while sales or operations executives

would address the remainder. The web-based survey instrument afforded respondent firms the

opportunity to gather the input of multiple executives by simply accessing the survey on-line to

complete designated sections. The survey instrument was developed with the assistance of a

panel of industry experts, consisting of two 3PL CEOs, two senior logistics researchers (external

to the research team), and one president of a major consulting firm. Of the 66 3PL firms

contacted, 35 (53 percent) fully completed the survey instrument. Descriptive statistics for these

35 firms appear in Table 1.

------------------------------------------------------------------------------------------------------------
***take in Table 1 approximately here***
------------------------------------------------------------------------------------------------------------

The 3PL respondents were also asked to identify up to eight primary customers. A

second questionnaire tailored to the customer perspective of 3PL service performance was then

e-mailed to these customers. Customers were asked to evaluate the performance of the named

3PL and to determine their levels of satisfaction and loyalty. When 3PLs failed to identify

customers, the research team referenced the Armstrong and Associates (1999) directory once

again to select at least two customers among those listed for that 3PL. In total 129 customers

were identified: 99 by 3PLs themselves and 30 from the Armstrong and Associates directory. A

total of 113 responses were obtained (89 from the 3PL provided sample and 24 from those

identified from the directory) for an overall customer response rate of 89 percent. Of the 113

surveys returned 111 were deemed complete for the measures of interest in this research. This

represents an effective response rate of 86 percent.

16
 The target respondent within each customer firm was the individual that works most

closely with the 3PL and is believed to be in the best position to evaluate service performance as

well as the firm’s satisfaction and loyalty to the 3PL. In the vast majority of cases, this resulted

in responses from those in functional management, director, or senior management positions.

Tables 2a, 2b and 2c provide demographic information for the 111 complete responses. The data

were segmented by industry, firm size, and magnitude of business relationship, respectively, in

order to assess the possibility of firm biases in the sample. Analysis of variance results indicated

no presence of biases by industry, firm size or annual transaction size.

------------------------------------------------------------------------------------------------------------
***take in Tables 2a, 2b, and 2c approximately here***
------------------------------------------------------------------------------------------------------------

The mean and standard deviation figures reported for the measurement items (Appendix)

provide insight regarding differences between customers identified by providers and customers

identified through the Armstrong and Associates listing. While customers generally responded

very favorably to their 3PL provider (as illustrated by mean values ranging between 4.76 and

5.16 for satisfaction and 4.78 and 5.68 for loyalty), the standard deviation values suggest that the

degree of favorable attitudes was not universal across the sample. The same may be said of the

measures of service performance. Statistical analyses were conducted to determine if differences

existed across the customer groups. Results from these analyses indicated little difference

between respondent groups. Specifically, t-tests of differences in means of responses to all

customer perception variables between the provider identified customer group and customers

identified through the Armstrong listing revealed only 5 significant differences out of 15

variables at the p < .05 level. Additionally, Levene’s Test for Equality of Variances showed that

the variances in the responses between the groups was different for only 2 of the 15 variables.

17
This indicates that the distribution of responses across groups was not significantly affected by

the identification source.

Measurement Scales

A review of literature in logistics, marketing and operations provided measures for the

customer survey. Service performance measurements across the operational, relational and cost

dimensions are derived primarily from the work of Parasuraman et al. (1985), Bienstock et al.

(1997) and Stank et al. (1999). Consistent with Stank et al. (1999), these measures were

submitted to protocol analysis within the specific industry of interest for assessment of

measurement item relevance and importance, per Brown et al. (1993). The aforementioned panel

of industry experts provided critical input toward the final selection of appropriate service

performance measures.

Measurements of customer satisfaction and loyalty were originally drawn from the

marketing literature and subsequently validated in Stank et al. (1999). Satisfaction measures are

derived from the work of Leuthesser and Kohli (1995). These measures are global in nature,

providing an overall judgment of the extent to which service matches expectations. They do not

assess satisfaction with any single element of the service received or specific interaction

associated with the service relationship (Anderson and Sullivan 1993). Likewise, measures of

loyalty determine the general relationship between the customer’s relative attitude toward the

service provider and the customer’s subsequent patronage behavior (Dick and Basu 1994; and

Morgan and Hunt 1994). Cronin and Morris (1989) and Morgan and Hunt (1994) are the

primary sources for these measures.

The service provider’s market share measurement is a relative measure reflecting the 3PL

managers’ assessments of their share of the North American 3PL market on a 7-point scale

18
where 1 = Worst in Industry, 4 = Average, 7 = Best in Industry. In addition to this relative

assessment of share, each service provider was asked to supply a good-faith estimate of the

firm’s actual market share (percentage of total North American 3PL market). The “good-faith

estimate” is a statement of the actual percentage share enjoyed by the firm (open-ended).

Estimated market share values ranged from a low of .0125% to 10.0%.

Thirty-three firms provided a subjective assessment, fifteen provided an estimate of the

percentage of share enjoyed by the firm, and 33 reported gross revenues in the Armstrong guide.

The subjective scale was used in this research since it provided a larger available sample size

than that available with the reported objective measure of market share. Past research has also

found that managerial assessments are consistent with objective internal performance (Dess and

Robinson 1984) and with external secondary data (Venkatraman and Ramanujam 1986). It

should be noted that while the subjective assessment is used in this research, there was a

significant, positive correlation between the subjective measure and actual measure of market

share (r = 0.634). The correlation between the subjective measure and gross revenue figure

reported by Armstrong (r = 0.679) is based on 32 common observations.

Upon concluding data collection, the measurement items were chosen for this research

based on a review of all items collected in the survey instrument. The review of measurement

items sought to identify measures that provided parsimonious representation of the model

constructs and complete data for the sake of analysis. The final selection of appropriate items

was verified in a confirmatory factor analysis of the measurement model (presented next). The

Appendix provides a list of all measurement items used in the research. Basic descriptive

statistics (mean and standard deviation) are provided as well as a correlation matrix.

Measurement Model Test

19
 The measurement and structural models were tested by performing latent variable

structural equation modeling using Bentler’s (1998) EQS for Windows (version 5.7) software.

Structural equation modeling (SEM) is a statistical approach that has the capacity to

comprehensively and simultaneously test hypotheses among observed and latent variables

(Hoyle 1995). Prior to testing the structural model, confirmatory factor analysis (CFA) was

performed to further assess the construct validity of the latent variables. CFA provides a more

rigorous method for assessing unidimensionality than Cronbach alpha, exploratory factor

analysis, and item-total correlations (Anderson and Gerbing 1988).

Table 3 reports the results of the CFA. The primary outputs of the CFA are the

assessments of measurement model fit. The traditional chi-square fit test indicates how well the

model-implied covariance matrix matches the covariances among the measured variables in the

sample data (Bollen 1989; Hayduk 1987; Marsh and Balla 1994). In a reversal of the typical

testing assumption, support for the null hypothesis of equal covariances is sought to demonstrate

sound model fit. In our case, the chi-square statistic of 185.324 (90 d.f.) results in a p-value

below 0.001 -- indicating rejection of the null hypothesis and poor model fit.

------------------------------------------------------------------------------------------------------------
***take in Table 3 approximately here***
------------------------------------------------------------------------------------------------------------

Chi-square, however, is not the sole measure of fit. Other fit statistics have been

developed to provide further indication of goodness-of-fit. These include the Bentler-Bonett

Normed Fit Index (BBNFI), Bentler-Bonett Nonnormed Fit Index (BBNNFI), and Comparative

Fit Index (CFI). Two of the three statistics reported for the current model have values greater

than the 0.90 cutoff suggested by the literature to indicate reasonable fit (Bagozzi and Yi 1988;

Fornell and Larcker 1981; Hu and Bentler 1995). Bentler (1990) and Byrne (1994) claim that

20
the CFI is the single most important index since it accounts for sample size – a common bias in

index calculations. The CFI for the current model is 0.931, indicating sound fit. Likewise, the

root mean square residual (RMSR) is acceptable at 0.040. A measure of parsimonious fit, the

normed chi-square, is the ratio of the chi-square value to degrees of freedom. The normed chi-

square for our CFA is approximately 2.06:1. While no consensus regarding a specific value for

this normed chi-square statistic has been reported, ratios varying from 2:1 to 5:1 have been

offered as upper thresholds for acceptable fit (Arbuckle 1997).

According to Anderson (1987), convergent validity is supported when factor loadings

(lambdas) demonstrate that measurement items load significantly on their designated latent

variables. The standardized lambda estimates in Table 3 present ample evidence for this form of

construct validity. The lowest value among the estimates is 0.585 (item OP2). Further support

of convergent validity is provided by the lack of significant, theory-driven modification indices

(e.g. the Wald and Lagrange Multiplier (LM) Tests in EQS). Significant modification indices

suggest that better model fit is possible by either dropping a “trouble” item (per the Wald Test)

or respecifying a measurement item to a latent variable other than that proposed in the CFA (per

the Lagrange Multiplier Test). While the Wald Test indicated that no parameters should be

dropped in the analysis, the LM Test suggested four model respecifications (i.e., measurement

items that cross-loaded with latent variables beyond those for which they serve as indicators). In

each case, however, the research team deemed that the current model specification provides for

stronger nomological validity than that suggested by the LM Test statistic. Therefore, the

measurement model remained unchanged.

Discriminant validity assesses whether two or more constructs are the result of a single

underlying construct (Dröge and Germain 2000). The most common method used to assess

21
discriminant validity is the nested model approach, where comparisons are made between the

original measurement model and successive models with correlations (phis) among latent

variables fixed equal to one. As long as the alternative measurement models fail to demonstrate

better fit (significantly lower chi-square goodness-of-fit values) than the original, support for

discriminant validity among constructs exists (Bagozzi and Yi 1988). This was found to be true

in the current research. Given the overall sound assessment of the measurement model, attention

will turn to the structural model and testing of hypothesized relationships.

RESULTS AND DISCUSSION

The seven hypotheses illustrated in Figure 1 were tested simultaneously in a single

structural equation model with EQS (Bollen 1989). Results of the full model analysis are

provided in Table 4. The fit statistics for the full model are comparable to those of the

measurement model (chi-square = 197.129; df = 98; p < 0.001). Two of the three fit indices

again demonstrate sound model fit, with the Comparative Fit Index sufficiently exceeding 0.90.

Also, the normed chi-square value of 2.01 is just above the stringent 2.0 standard for

parsimonious fit per Arbuckle (1997). Examination of the hypotheses can proceed given an

overall sound assessment of model fit.

------------------------------------------------------------------------------------------------------------
***take in Table 4 approximately here***
------------------------------------------------------------------------------------------------------------

The first two hypotheses examine the direct influence that relational performance exerts

on the other two dimensions of service performance (i.e., operational performance and cost

performance). Model results indicate strong support for hypothesis H1, demonstrating that

relational performance is positively related to operational performance. Likewise, hypothesis H2

is strongly supported, suggesting that relational performance also is positively related to cost

22
performance in a positive manner. Support for each hypothesis is illustrated in the significance

of the standardized parameter estimates (gamma values) and the associated t-values in Table 4.

Hypotheses H3 through H5 suggest that each of the three service performance

dimensions are positively related to customer satisfaction. Results of the analysis provide

support for only one of the three hypotheses. Relational performance demonstrates a positive,

significant relationship with satisfaction (at the 0.01 level of significance), but interestingly,

operational performance and cost performance are found to have no significant relationship with

satisfaction. Therefore, support is found for H3, but not H4 and H5.

Hypothesis H6 suggests that customer satisfaction is positively influenced by loyalty.

The analysis provides strong support for this contention. Hypothesis H7 subsequently identifies

market share as positively related to customer loyalty. The data support this hypothesis as well

(at the 0.05 level of significance). Figure 2 portrays the hypothesis test results.

------------------------------------------------------------------------------------------------------------
***take in Figure 2 approximately here***
------------------------------------------------------------------------------------------------------------

Similar to earlier work conducted by Daugherty, Stank and Ellinger (1998), the research

identifies service performance as a key antecedent of market share through its relationship with

customer satisfaction and loyalty. The operationalization utilized in the current research, based

upon the two-dimensional performance portrayal of Stank, et al. (1999) refined with a third

dimension representing cost to customer, re-explored the basic premise of the model introduced

by Daugherty, et al. The new operationalization fostered a deeper understanding of the

interrelationships among the performance dimensions as well as providing some indication as to

which elements are key to improvements in satisfaction, loyalty, and market share.

23
 The current operationalization gave rise to an interesting result. Previous findings

revealed that both relational and operational performance were significant predictors of customer

satisfaction. The current findings refuted this. Specifically, the current findings cast relational

performance as having the only significant relationship with customer satisfaction, while the

operational performance-satisfaction and cost performance-satisfaction relationships were not

significant. The results support the belief that basic operational service and cost performance are

recognized as order qualifiers and not differentiators in the eyes of customers. Relational

performance elements are the key to differentiating excellent 3PL service from the ordinary.

Service providers must not, however, look past the fundamental delivery of time and place

utilities that meet customer expectations at a rate commensurate with delivered value.

The findings support the critical importance of establishing strong relationships with

customers to gain the insights needed to tailor services to enhance both operational effectiveness

and cost efficiency. Strong relational performance allows firms to proactively seek information

on customer preferences and use it to be more responsive. In addition, service providers ensure

that resources are invested only in areas that customers perceive to be important, improving asset

and capital utilization and enabling cost efficiency. Ultimately, it becomes difficult for

competitors to intercede in supplier-buyer relations once they reach this level of intimacy.

Conversations with practitioners following data analysis provided anecdotal support for

the finding. A global supply chain manager for a large computer manufacturer, for example,

discussed his firm’s supplier scorecard with the authors. The manager noted that quality service

on key operational elements was the central element of the supplier scorecard. This element,

however, was not given the highest weighting in the supplier evaluation algorithm because it

represented a basic qualifier used to determine whether a service provider could even remain a

24
potential supplier. In other words, 3PLs were not even considered unless they could demonstrate

sustainable levels of high performance on basic operational service elements. Rather, suppliers

were rewarded based upon their ability to respond to the manufacturer’s specific requests, and do

so under increasingly stringent cost requirements.

Such findings are not new in logistics. Stank, Daugherty, and Ellinger (1998)

interviewed restaurant managers in support of research conducted in the food service industry

and noted that distributors that go beyond “core” operating capabilities to do “whatever it takes”

to serve customers on the customers’ terms earn the continuing business of fast food restaurants.

Such distributors provide frequent communications and easily accommodate special requests.

Poorly regarded distributors exhibit deficiencies in non-core distribution capabilities

characterized by communications breakdowns and inflexibility in responding to special requests

or service failures. They may have developed basic delivery capabilities but tend to miss the

point regarding customer focus. One manager told of a story that exemplifies the finding. “The

delivery truck blocked the restaurant parking lot and the manager could see potential diners

pulling away because they could not enter the lot. When told of the problem, the driver

expressed his need to proceed with his delivery so that he could he complete his route on time.

The manager noted that the driver was ensuring that the food distribution firm performed well on

logistical service, although it was the wrong service to satisfy the customer. The driver was

working toward an inappropriate and myopic performance goal that totally ignored the supply

chain bottom line (Stank, Daugherty, and Ellinger 1998, p. 78).

Partial explanation for the finding may be found in the industry context of the research.

In a situation where an external supplier assumes ownership of all or part of a key business

process, successfully forging a strong relationship between the customer and service provider

25
may be the single most critical determinant of customer satisfaction. The influence of relational

characteristics may dwarf all other considerations. Our results suggest that the development of

an effective relationship is prerequisite to success on all other performance dimensions. Strong

relationships can even ensure rapid service recovery when operational or cost activities falter.

Previous, more narrowly focused research has identified a direct link between customer

satisfaction and overall firm performance (Anderson, Fornell, and Lehman 1994; Ittner and

Larcker 1996). Although we did not hypothesize a direct link between customer satisfaction and

market share (the overall firm performance measure in our study), it should be noted that the

SEM model did not suggest one (per the LM Test). Similar to earlier logistics research, our

findings indicate that customer satisfaction does not directly affect overall firm performance, but

rather affects overall performance indirectly through customer loyalty. Additional research is

needed to examine relationships among customer satisfaction, loyalty, and other measures of

overall firm performance (e.g., ROI, ROA) in a holistic, service performance context.

An important contribution of the research was that the data represented both sides of the

supplier-buyer dyad. Customers of third-party logistics providers evaluated their perceptions of

service performance as well as their overall satisfaction with and loyalty to the provider. Our

research linked customer perceptions to a market share indicator provided by the 3PL suppliers

themselves that correlated significantly with objective data of market share drawn from a

secondary data source. Such efforts in data collection help ensure that provider biases regarding

relationships among model antecedents and outcomes are removed.

CONCLUSION

This research examined the relationships of service performance, customer satisfaction,

customer loyalty and market share in the context of a holistic model that allowed for the

26
simultaneous testing of these relationships. The model introduced a three-dimensional

conceptualization of service performance for the 3PL industry using data from service providers’

customers to assess key constructs. The results indicated that this conceptualization might

provide a more generic approach for capturing service performance in an industrial services

context than previous customizations of Parasuraman et al.’s five-dimensional measurement

scheme. Future research, however, might explore alternate operationalizations of the logistics

service performance constructs to ensure generalizability of results. The results also support

Stank et al.’s finding that relational performance is antecedent to operational performance. This

antecedent relationship extends to a discrete assessment of cost performance. Finally, the

research establishes an empirical link between customer loyalty and a perceptual measure of

market share that is anchored by objective data from secondary sources. Future research should

seek to assess the direct relationship between perceptions of service performance and objective

measures of firm performance. Future research should also investigate whether length of

relationship has an impact on the interrelationships identified in the model.

27
 REFERENCES

Anderson, James C. (1987), “An Approach for Confirmatory Measurement and Structural
Equation Modeling of Organizational Properties,” Management Science, Vol. 33, (April), pp.
525-541.

Anderson, Eugene W., Claes Fornell, and Donald R. Lehman (1994), “Customer Satisfaction,
Market Share, and Profitability: Findings from Sweden,” Journal of Marketing, Vol. 58, No. 3,
pp. 53-66.

Anderson, Eugene W. and Mary Sullivan (1993), “The Antecedents and Consequences of
Customer-Satisfaction for Firms,” Marketing Science, Vol. 12, No. 2, pp. 125-143.

Anderson, James C. and David W. Gerbing (1988), Structural Equation Modeling in Practice: A
Review of the Two-Step Approach. Psychological Bulletin, Vol. 103, No. 3, pp. 411-423.

Arbuckle, J.L. (1997), Amos Users’ Guide, Version 3.6, Chicago: SmallWaters Corporation.

Armstrong and Associates (1999), Armstrong's Guide To Third Party Logistics Services
Providers, Vol. 7, Stoughton, WI: Armstrong and Associates, Inc.

Babakus, Emin and W. Glynn Mangold (1989), “Adopting the ‘SERVQUAL’ Scale to Health
Care Environment: An Empirical Assessment,” in P. Bloom, et al. (Eds.), AMA Educators’
Proceedings. Chicago, IL: American Marketing Association, p. 195.

Babakus, Emin and Gregory W. Boller (1992), “An Empirical Assessment of the SERVQUAL
Scale,” Journal of Business Research, Vol. 24, No. 3, pp. 253-268.

Bearden, W.O. and J.E. Teel (1983), “Selected Determinants of Consumer Satisfaction and
Complaint Reports,” Journal of Marketing Research, Vol. 20, No. 1 (February), pp. 21-28.

Berry, Leonard L. (1995), On Great Service, A Framework for Action. New York: The Free
Press.

Berry, Leonard L. and A. Parasuraman (1991), Marketing Services, Competing Through Quality.
New York: The Free Press.

Bagozzi, Richard P. and Youjae Yi (1988), “On the Evaluation of Structural Equation Models,”
Journal of the Academy of Marketing Science, Vol. 16, No. 1, pp. 74-94.

Bentler, Peter M. (1990), “Fit Indexes, Lagrange Multipliers, Constraint Changes, and
Incomplete Data in Structural Models,” Multivariate Behavioral Research, Vol. 25, pp. 163-172.

Bentler, Peter M. (1998), EQS for Windows, Version 5.7. Encino, CA: Multivariate Software.

28
Bienstock, Carol C., John T. Mentzer, Monroe Murphy Bird (1997), “Measuring Physical
Distribution Service Quality,” Journal of the Academy of Marketing Science, Vol. 25, No. 1, pp.
31-44.

Bojanic, David C. (1991), “Quality Measurement in Professional Services Firms,” Journal of

Professional Services Marketing, Vol. 7, No. 2, pp. 27-36.

Bollen, Kenneth A. (1989), Structural Equations with Latent Variables. New York: Wiley.

Bowersox, Donald J. (1999), “21st Century Logistics: Making Supply Chain Integration a
Reality” Keynote address to the Council of Logistics Management Annual Conference, Toronto,
Canada, October 17-20.

Bowersox, Donald J., David J. Closs, and M. Bixby Cooper (2002), Supply Chain Logistics
Management. New York: McGraw-Hill/Irwin, p. 34.

Brown, Tom J., Gilbert A. Churchill, Jr., and J. Paul Peter (1993), “Improving the Measurement
of Service Quality,” Journal of Retailing, Vol. 69, No. 1, pp. 127-139.

Byrne, Barbara M. (1994), Structural Equation Modeling with EQS and EQS/Windows.
Thousand Oaks, CA: Sage.

Carman, James M. (1990), “Consumer Perceptions of Service Quality: An Assessment of the

SERVQUAL Dimensions,” Journal of Retailing, Vol. 66 (Spring), pp. 33-55.

Cleveland, Gary, Roger G. Schroeder, and John C. Anderson (1989), “A Theory of Production
Competence,” Decision Sciences, Vol. 20, No. 4, pp. 655-668.

Crompton, John L. and Kelly J. Mackay (1989), “Users’ Perceptions of the Relative Importance
of Service Quality Dimensions in Selected Public Recreation Programs,” Leisure Sciences, Vol.
11, pp. 367-375.

Cronin, J. Joseph, Jr. and Steven A. Taylor (1992), “Measuring Service Quality: A
Reexamination and Extension,” Journal of Marketing, Vol. 56, No. 3, pp. 55-68.

Cronin, J. Joseph, Jr. and Michael H. Morris (1989), “Satisfying Customer Expectations: The
Effect on Conflict and Repurchase Intentions in Industrial Marketing Channels,” Journal of the
Academy of Marketing Science, Vol. 17, No. 1, pp. 41-49.

Crosby, Lawrence A., Kenneth R. Evans, and Deborah Cowles (1990), “Relationship Quality in
Services Selling: An Interpersonal Influence Perspective,” Journal of Marketing, Vol. 54, No. 3,
pp. 68-81.

Daugherty, Patricia J., Theodore P. Stank, and Alexander E. Ellinger (1998), “Leveraging
Logistics/Distribution Capabilities: The Impact of Logistics Service on Market Share,” Journal
of Business Logistics, Vol. 19, No. 2, pp. 35-51.

29
Dess, G.S. and R.B. Robinson (1984), “Measuring Organizational Performance in the Absence
of Objective Measures,” Strategic Management Research, Vol. 5, No. 3, pp. 265-273.

Delaney, Robert (1999), 10th Annual “State of Logistics” Report, St. Louis: Cass Information
Systems.

Dick, Alan S. and Kunal Basu (1994), “Customer Loyalty: Toward an Integrated Conceptual
Framework,” Journal of the Academy of Marketing Science, Vol. 22, No. 2, pp. 99-113.

Dröge, Cornelia and Richard Germain (2000), “The Relationship of Electronic Data Interchange
with Inventory and Financial Performance,” Journal of Business Logistics, Vol. 21, No. 2, pp.
209-230.

Ferdows, Kasra. and Arnoud De Meyer (1990), “Lasting Improvements in Manufacturing

Performance: In Search of New Theory,” Journal of Operations Management, Vol. 9, No. 2, pp.
168-184.

Finn, David W. and Charles W. Lamb Jr. (1991), “An Evaluation of the SERVQUAL Scales in a
Retail Setting,” in Holman, R.H., Solomon, M.R. (Eds.), Advances in Consumer Research, Vol.
18. Provo, Utah: Association for Consumer Research.

Fornell, Claes (1992), “A National Customer Satisfaction Barometer: The Swedish Experience,”
Journal of Marketing, Vol. 55, No. 1, pp. 1-21.

Fornell, Claes and David F. Larcker (1981), “Evaluating Structural Equation Models with
Unobservable Variables and Measurement Error,” Journal of Marketing Research, Vol. 18, No.
1, pp. 39-50.

Fournier, Susan T. and David G. Mick (1999), “Rediscovering Satisfaction,” Journal of

Marketing, Vol. 63, No. 4, pp. 5-23.

Green, Paul E. and Donald S. Tull (1978), Research For Marketing Decisions, 4th ed. Englewood
Cliffs, NJ: Prentice-Hall, Inc., pp. 210-211.

Hayes, Robert H. and Steven C. Wheelwright (1984), Restoring Our Competitive Edge:
Competing Through Manufacturing. New York: Wiley.

Hayduck, Leslie A. (1987), Structural Equation Modeling with LISREL: Essentials and
Advances. Baltimore, MD: Johns Hopkins University Press.

Hill, Terry (1989), Manufacturing Strategy Text and Cases. Homewood, IL: Irwin.

Hoyle, Rick H. (1995), “The Structural Equation Modeling Approach: Basic Concepts and
Fundamental Issues”, in R.H. Hoyle (ed.), Structural Equation Modeling: Concepts, Issues, and
Applications, Sage: Thousand Oaks, CA, pp. 1-15.

30
Hu, Li-Tze and Bentler, Peter M. (1995), “Evaluating Model Fit,” in R.H. Hoyle (ed.),
Structural Equation Modeling: Concepts, Issues, and Applications, Sage: Thousand Oaks, CA,
pp. 76-99.

Innis, David E. and Bernard J. LaLonde (1994), “Customer Service: The Key to Customer
Satisfaction, Customer Loyalty, and Market Share,” Journal of Business Logistics, Vol. 15, No.
1, pp. 1-27.

Ittner, Christopher D. and David F. Larcker (1996), Measuring the Impact of Quality Initiatives
on Firm Financial Performance,” in Ghosh, S. and D. Fedor, (Eds.), Advances in the
Management of Organizational Quality, Vol. 1. Greenwich, CT: JAI, pp. 1-37.

Jacoby, Jacob and David B. Kyner (1973), “Brand Loyalty vs. Repeat Purchasing Behavior,”
Journal of Marketing Research, Vol. 10 (February), pp. 1-9.

Johnson, L.L., M.J. Dotson, and B.J. Dunlop (1988), “Service Quality Determinants and
Effectiveness in the Real Estate Brokerage Industry,” The Journal of Real Estate Research, Vol.
3, pp. 21-36.

Jones, T.O, and W.E. Sasser, Jr. (1995), “Why Satisfied Customers Defect,” Harvard Business
Review, Vol. 73, No. 6 (November-December), pp. 88-99.

Krajewski, Lee J. and Larry P. Ritzman (1987), Operations Management, Strategy and Analysis,
1st Ed. Reading, Massachusetts: Addison Wesley.

Leuthesser, Lance and Ajay K. Kohli (1995), “Relational Behavior in Business Markets,”
Journal of Business Research, Vol. 34, No. 1, pp. 221-233.

Lovelock, Christopher H. (1983), “Classifying Services to Gain Strategic Marketing Insights,”

Journal of Marketing, Vol. 47, No. 3, pp. 9-20.

Marsh, H.W. and J.R. Balla (1994), “Goodness-of-Fit in Confirmatory Factor Analysis: The
Effects of Sample Size and Model Parsimony,” Quality and Quantity, Vol. 28, pp. 185-127.

Mentzer, John T., Daniel J. Flint, and John L. Kent (1999), “Developing a Logistics Service
Quality Scale,” Journal of Business Logistics, Vol. 20, No. 1, pp. 9-32.

Morgan, Robert M. and Shelby D. Hunt (1994), “The Commitment-Trust Theory of Relationship
Marketing,” Journal of Marketing, Vol. 58, No. 3, pp. 20-38.

Oliva, Terence A., Richard L. Oliver, and Ian C. MacMillan (1992), “A Catastrophe Model for
Developing Service Satisfaction Strategies,” Journal of Marketing, Vol. 56, No. 3, pp. 83-95.

31
Parasuraman, A., Valerie A. Zeithaml, and Leonard L. Berry (1985), “A Conceptual Model of
Service Quality and Its Implications for Future Research,” Journal of Marketing, Vol. 49, No. 4,
pp. 41-50.

Parasuraman, A., Valerie A. Zeithaml, and Leonard L. Berry (1988), “SERVQUAL: A Multiple-
Item Scale for Measuring Consumer Perceptions of Service Quality,” Journal of Retailing, Vol.
64, No. 1 pp. 12-40.

Parasuraman, A., Leonard L. Berry, and Valerie A. Zeithaml (1991), “Refinement and
Reassessment of the SERVQUAL Scale,” Journal of Retailing, Vol. 67, No. 4, pp. 420-450.

Porter, Michael E. (1980), Competitive Strategy. New York: The Free Press.

Porter, Michael E. (1985), Competitive Advantage. New York: The Free Press.

Reichheld, Frederick F. and W. Earl Sasser, Jr. (1990), “Zero Defections: Quality Comes to
Services,” Harvard Business Review, Vol. 68, No. 5, pp. 105-111.

Roth Aleda V. and J.G. Miller (1990), “Manufacturing Strategy, Manufacturing Strength,
Managerial Success, and Economic Outcomes” in Ettlie, J.E., M.C. Burstein, and A. Fiegenbaum
(Eds.), Manufacturing Strategy, the Research Agenda for the Next Decade, Proceedings for the
Joint Industry University Conference On Manufacturing Strategy, Ann Arbor, Michigan, pp. 85-
96.

Roth, Aleda V. and Marjolijn van der Velde (1991), “Operations as Marketing: A Competitive
Service Strategy,” Journal of Operations Management, Vol. 10, No. 3, pp. 303-328.

Rust, Roland T. and Anthony J. Zahorik (1993), “Customer Satisfaction, Customer Retention,
and Market Share,” Journal of Retailing, Vol. 69, No. 2, pp. 193-215.

Scannell, Thomas V., Shawnee K. Vickery, and Cornelia L. Dröge (2000), “Upstream Supply
Chain Management and Competitive Performance in the Automotive Supply Industry,” Journal
of Business Logistics, Vol. 21, No. 1, pp. 23-48.

Sharma, Arun and Douglas M. Lambert (1990), “Segmentation of Markets Based on Customer
Service,” International Journal of Physical Distribution and Logistics Management, Vol. 20, No.
7, pp. 19-27.

Stank, Theodore P., Thomas J. Goldsby, and Shawnee K. Vickery (1999), “Effect of Service
Supplier Performance on Satisfaction and Loyalty of Store Managers in the Fast Food Industry,”
Journal of Operations Management, Vol. 17, No. 2, pp. 429-447.

Stank, Theodore P., Patricia J. Daugherty, and Alexander E. Ellinger (1998), “Pulling Customers
Closer Through Logistics Service,” Business Horizons, Vol. 41, No. 5 (September-October), pp.
74-80.

32
Venkatraman, N. and V. Ramanujam (1986), “Measurement of Business Performance in the
Absence of Objective Measures,” Strategic Management Review, Vol. 11, No. 4, pp. 801-814.

Wood, C. H., L.P. Ritzman, and D. Sharma (1990), “Intended and Achieved Competitive
Priorities: Measures, Frequencies, and Financial Impact” in Ettlie, J.E., M.C. Burstein, and A.
Fiegenbaum (Eds.), Manufacturing Strategy, the Research Agenda for the Next Decade,
Proceedings for the Joint Industry University Conference On Manufacturing Strategy, Ann
Arbor, Michigan, pp. 225-232.

Youngdahl, William E. and Deborah L. Kellogg (1997), “The Relationship between Service
Customers’ Quality Assurance Behaviors, Satisfaction, and Effort: A Cost of Quality
Perspective,” Journal of Operations Management, Vol. 15, No. 1, pp. 19-32.

Zeithaml, Valerie A., A. Parasuraman, and Leonard L. Berry (1990), Delivering Service Quality,
Balancing Customer Perceptions and Expectations, The Free Press, New York.

Zeithaml, Valerie A. (2000), “Service Quality, Profitability, and the Economic Worth of
Customers: What We Know and What We Need to Learn,” Journal of the Academy of
Marketing Science, Vol. 28 (Winter), pp. 67-85.

33
 APPENDIX
MEASUREMENT ITEMS

ASSESSMENTS OF SERVICE PERFORMANCE

How do you perceive [name of 3PL]’s service performance in 1998 (the past year)? Please show
your agreement with service feature described. If you strongly agree with the statement
regarding the service feature provided by [name of 3PL], select 7. If you strongly disagree with
the statement, select number 1. If your feelings are not strong, select one of the numbers in the
middle that best shows your level of agreement with the statement (where 2 = moderately
disagree, 3 = slightly disagree, 4 = neutral, 5 = slightly agree, and 6 = moderately agree).*

Latent Variable Item Measured variable Mean Std.

Reference Deviation
OP1 Meets promised deadlines 4.98 1.55
OP2 Delivers undamaged orders 5.41 1.19
Operational Performance OP3 Delivers accurate orders 5.39 1.01
(i.e., items ordered arrive,
Cronbach  = 0.7303 not unordered items)
RP1 Knows your needs well 5.57 1.45
RP2 Cooperates with you to help 5.83 1.30
Relational Performance do the job well
RP3 Makes recommendations for 4.41 1.73
continuous improvement on
Cronbach  = 0.8439 an on-going basis
CP1 Provides services that result 4.86 1.47
in the lowest total cost
Cost Performance logistics solution
CP2 Offers competitive prices 5.16 1.43
2
R = 0.8496 for services

* Note that the name of the 3PL to be evaluated by the customer respondent was positioned in
the [bracketed] space by the survey administrator.

34
Assessments of Satisfaction and Loyalty

Next, please indicate how strongly you agree or disagree with the following statements regarding
[name of 3PL]. Evaluate each item using a 7-point scale with 1 = strongly disagree, 2 =
moderately disagree, 3 = slightly disagree, 4 = neutral, 5 = slightly agree, 6 = moderately agree,
and 7 = strongly agree.

Latent Variable Item Measured variable Mean Std.

Reference Deviation
CS1 I am delighted with my firm’s overall 5.16 1.51
relationship with this service
provider.
Customer Satisfaction CS2 I wish more of my firm’s service 4.76 1.80
providers were like this one.
CS3 There is always some problem or 4.84 1.91
another with this service provider
Cronbach  = 0.9136 (reverse-scaled)
CL1 The relationship that my firm has 5.68 1.41
with this service provider is
something we are very committed to.
CL2 The relationship that my firm has 4.78 1.66
with this service provider is
something we intend to maintain
Customer Loyalty indefinitely.
CL3 Maintaining a long-term relationship 5.38 1.58
with this service provider is very
important to my firm.
CL4 I would recommend that my 5.43 1.62
successor continue using this service
Cronbach  = 0.9079 provider.

Assessment of Service Provider’s Financial Performance (Market Share) – Data from 3PLs

This section assesses your company’s financial performance for the most recently completed
fiscal year (1998). Subjectively rate your firm’s performance for each measure in comparison to
its major industry competitors for 1998, where 1 = Worst in Industry, 4 = Average, 7 = Best in
Industry.

Variable Item Measured variable Mean Std.

Reference Deviation
MS1 Market share of total North American 4.22 1.91
Market Share Third- Party Logistics Industry
(based on Cass Logistics’ estimate of
$40 Billion) (%)

35
 CORRELATIONS AMONG INDICATOR AND LATENT VARIABLES

OP1 OP2 OP3 RP1 RP2 RP3 CP1 CP2 CS1 CS2 CS3 CL1 CL2 CL3 CL4 MS1 OP RP CP CS CL MS
OP1 
OP2 .502 
OP3 .481 .513 
RP1 .592 .305 .401 
RP2 .685 .293 .423 .718 
RP3 .646 .429 .484 .580 .693 
CP1 .518 .246 .429 .655 .567 .605 
CP2 .470 .204 .415 .665 .616 .555 .739 
CS1 .645 .394 .507 .657 .772 .727 .677 .622 
CS2 .655 .448 .529 .689 .754 .746 .689 .650 .879 
CS3 .634 .418 .430 .491 .654 .585 .594 .556 .758 .748 
CL1 .515 .267 .468 .472 .607 .513 .544 .424 .650 .636 .414 
CL2 .440 .183 .430 .496 .568 .500 .463 .550 .650 .660 .499 .649 
CL3 .412 .065 .310 .545 .571 .462 .458 .493 .605 .548 .383 .668 .751 
CL4 .561 .224 .474 .699 .711 .613 .678 .670 .807 .803 .597 .699 .750 .755 
MS1 .043 -.034 .116 .177 .185 .207 .138 .169 .149 .129 .095 .097 .099 .233 .230 
OP 
RP .879 
CP .662 .856 
CS .826 .940 .821 
CL .660 .848 .786 .866 
MS .069 .219 .171 .139 .222 

36
 FIGURE 1
CONCEPTUAL MODEL AND HYPOTHESES

Operational
Performance
(OP)
+ +
H1 H4

Relational + Customer + Customer +

Performance Satisfaction Loyalty Market Share
(RP) H3 (CS) H6 (CL) H7 (MS)

+ +
H2 H5
Cost
Performance
(CP)

37
 FIGURE 2
HYPOTHESIS TEST RESULTS

Operational
Performance
(OP)
+ 0.862** + 0.079
H1 H4 (n.s.)
+ 0.872** + 0.213*
Relational +0.723** Customer Customer
Performance Satisfaction Loyalty Market Share
(RP) H3 (CS) H6 (CL) H7 (MS)

+ 0.833** + 0.175
(n.s.)
H2 H5
Cost
Performance
(CP)
Fit statisitics:
Chi-square: 197.129 (df = 98), p < 0.001
Bentler-Bonett Normed Fit Index: 0.869
Bentler-Bonett Nonnormed Fit Index: 0.912
* Indicates path is significant to p < .05
Comparative Fit Index: 0.928
** Indicates path is significant to p < .01
Root Mean Square Residual (RMSR): 0.042

38
 TABLE 1

3PL DEMOGRAPHICS

Category Mean Std. Minimum Maximum

Deviation
Full-time personnel employed by the firm 3,974.37 4,593.93 27 19,500
Annual sales volume ($millions) 332.03 420.32 12.00 2,225.00

TABLE 2a

TITLES OF CUSTOMER RESPONDENTS

Title Count
Manager 49
Director 30
Senior manager 11
Vice president 8
Supervisor 3
Analyst 3
President 1
Other/non-response 6

39
 TABLE 2b

INDUSTRY REPRESENTATION AMONG CUSTOMERS

Industry Count Industry Count Industry Count

Appliances 3 Electronics and related 13 Office equipment 3
instruments and supplies
Automotive 17 Electrical machinery 2 Paper and related 7
equipment products
Building 4 Food and beverage 14 Petroleum and 2
materials/lumber distribution/wholesaling petrochemicals
products
Industrial chemicals 9 Food and beverage 21 Health and 11
and plastics processing personal care
Consumer chemicals 8 Food and beverage 6 Pharmaceuticals 6
and plastics retailing and drugs
Clothing and textiles 7 Furniture 4 Primary metals 2
Computer hardware 15 Hardware 7 Rubber products 1
and peripheral and related goods
equipment
Construction, farm 6 Machine tools and 4 General 6
and garden machinery merchandise
equipment
Department 5 Mining and materials 2 Other 23
store/mass
merchandiser
Note that many respondents indicated activity in multiple industries

TABLE 2c

CUSTOMER SIZE AND VOLUME OF BUSINESS WITH 3PL

Category Mean Std. Minimum Maximum

Deviation
Full-time personnel employed by the firm 37,019 69,063 22 350,000
Annual sales volume ($millions) 11,263.36 24,656.86 5.00 160,000.00
Approximate ($millions) volume of
business conducted with 3PL service 12.152 16.027 0.050 80.000
provider in 1998

40
 TABLE 3

CONFIRMATORY FACTOR ANALYSIS OF MEASUREMENT MODEL

Item Standardized solution

Operational Relational Cost Customer Customer Market t-
Performance Performance Performance Satisfactio Loyalty Share valuea
n
OP1 0.859 --b
OP2 0.585 6.182
OP3 0.609 6.470
RP1 0.788 --b
RP2 0.858 10.235
RP3 0.809 9.474
CP1 0.866 --b
CP2 0.853 10.489
CS1 0.938 --b
CS2 0.942 19.300
CS3 0.790 11.860
CL1 0.758 --b
CL2 0.811 9.026
CL3 0.806 8.951
CL4 0.948 10.775
MS1 1.000 --b
a
t-values are from the unstandardized solution
b
t-values for these parameters were not available because they were fixed for scaling purposes

Fit statisitics:
Chi-square: 185.324 (df = 90), p < 0.001
Bentler-Bonett Normed Fit Index: 0.877
Bentler-Bonett Nonnormed Fit Index: 0.908
Comparative Fit Index: 0.931
Root Mean Square Residual (RMSR) 0.040
Maximum Likelihood (ML) solution
 TABLE 4

PARAMETER ESTIMATES FOR STRUCTURAL PATHS AND MEASUREMENT

RELATIONS

Structural path estimates

Parameter Standardized t-valuea
estimate
OP,RP 0.862 8.188
CP,RP 0.833 8.117
CS,RP 0.723 3.471
CS,OP 0.079 0.518
CS,CP 0.175 1.435
CL,CS 0.872 9.286
MS,CL 0.213 2.178
Measure model estimates
Parameter Standardized t-valuea
estimate
OP1,OP 0.852 --b
OP2,OP 0.573 5.949
OP3,OP 0.627 6.599
RP1,RP 0.788 --b
RP2,RP 0.870 10.301
RP3,RP 0.810 9.383
CP1,CP 0.875 --b
CP2,CP 0.844 10.327
CS1,CS 0.938 --b
CS2,CS 0.940 19.203
CS3,CS 0.787 11.762
CL1,CL 0.766 --b
CL2,CL 0.818 9.227
CL3,CL 0.809 9.095
CL4,CL 0.941 10.868
MS1,MS 1.000 --b
a
t-values are from the unstandardized solution
b
t-values for these parameters were not available because they were fixed for scaling purposes

Fit statisitics:
Chi-square: 197.129 (df = 98), p < 0.001
Bentler-Bonett Normed Fit Index: 0.869
Bentler-Bonett Nonnormed Fit Index: 0.912
Comparative Fit Index: 0.928
Root Mean Square Residual (RMSR) 0.042
Maximum Likelihood (ML) solution

42