Académique Documents
Professionnel Documents
Culture Documents
Hult
oth corporations and researchers are becoming customers view logistics services similarly across segments,
Journal of Marketing
82 / Journal of Marketing, October 2001 Vol. 65 (October 2001), 82104
Williams 2001; Morash, Droge, and Vickery 1996), in part An approach to investigate LSQ further is to build on the
because of its visible service impact on customers (Bien- service quality literature prevalent in marketing. The service
stock, Mentzer, and Bird 1997; Pisharodi and Langley 1990; quality approach, in general, is an attempt to understand cus-
Sharma, Grewal, and Levy 1995). To successfully leverage tomer satisfaction from the perspective of the differences
logistics excellence as a competitive advantage to cus- between customer perceptions and actual customer service on
tomers, logisticians must coordinate with marketing depart- various attributes (Parasuraman, Zeithaml, and Berry 1985).
ments (Kahn 1996; Kahn and Mentzer 1996; Mentzer and Researchers have begun to examine whether the service qual-
Williams 2001; Murphy and Poist 1996; Williams et al. ity model can be used to measure logistics service (Brensinger
1997). The quality of logistics service performance is a key and Lambert 1990). They have modified the original service
marketing component that helps create customer satisfaction quality model by developing logistics attributes that fit into the
(Bienstock, Mentzer, and Bird 1997; Mentzer, Gomes, and previously customer-defined dimensions and identifying addi-
Krapfel 1989) and has been recognized as such for some tional gaps that could be applied to the logistics service context
time (Perrault and Russ 1974). (Lambert, Stock, and Sterling 1990). These views of logistics
There are many definitions and descriptions of how service provide the building blocks to create a customer-based
logistics creates customer satisfaction. The most traditional foundation for better definitions and measures of LSQ.
are based on the creation of time and place utility (Perreault The use of customer-based definitions of LSQ brings
and Russ 1974). The so-called seven Rs describe the attrib- physical distribution research, which traditionally has focused
utes of the companys product/service offering that lead to on physically observable operational attributes, more in line
utility creation through logistics service; that is, part of a with marketing, which has devoted attention to understanding
products marketing offering is the companys ability to such unobservables as customers perceived value. By recog-
deliver the right amount of the right product at the right nizing, tapping into, and measuring customer perceptions of
place at the right time in the right condition at the right price LSQ, logistics practitioners and researchers can add to the tra-
with the right information (Coyle, Bardi, and Langley 1992; ditionally measured set of operational service attributes.
Shapiro and Heskett 1985; Stock and Lambert 1987). This
conceptualization implies that part of the value of a product
is created by logistics service. Service Quality
As the business environment has changed, the operations- Many researchers have tried to replicate empirically the five-
based definitions of logistics service have evolved. As such, dimensional structure (tangibles, responsiveness, empathy,
the idea of value has been broadened to include several value- reliability, and assurance) of Parasuraman, Zeithaml, and
added operational logistics tasks, such as packaging, third- Berrys (1985) original service quality instrument,
party inventory management, bar coding, and information sys- SERVQUAL. In developing SERVQUAL, Parasuraman,
tems (Ackerman 1989; Mentzer 1993; Mentzer and Firman Zeithaml, and Berry followed a general procedure of quali-
1994; Witt 1991). The value-added concept expanded the tra- tative research (interviews and focus groups) to develop the
ditional time and place utilities to include form utility (Acker- initial scale and then performed quantitative surveys to refine
man 1991) but was still an operations-based concept. LaLonde and empirically test the scale. These interviews and surveys
and Zinszer (1976) describe customer service as possessing included retail consumers of appliance repair or mainte-
three components: (1) an activity to satisfy customers needs, nance, retail banking, long-distance telephone service, secu-
(2) performance measures to ensure customer satisfaction, and rities brokers, and credit card services. Additional research
(3) a philosophy of firmwide commitment. However, these has expanded the use of SERVQUAL to include retail con-
components all focus on the provider firm, not on the cus- sumers of health care, residential utilities, job placement,
tomer. Similarly, other research has developed a framework pest control, dry cleaning, financial services, and fast-food
for quantifying the value created by logistics operations that is services, and the resultant dimensions have ranged from one
heavily focused on the service provider (Novack, Langley, and to eight (e.g., Babakus and Boller 1992; Babakus and Inhofe
Rinehart 1995). Although this research incorporates internal 1993; Babakus and Mangold 1992; Babakus, Pedrick, and
and external customers, it predominantly involves provider Inhofe 1993; Brown, Churchill, and Peter 1993; Carmen
firmsthat is, how logistics executives can quantify the value 1990; Cronin and Taylor 1992; Finn and Lamb 1990; Mishra,
they create for customers. A process is needed to measure cus- Singh, and Wood 1991; Parasuraman, Zeithaml, and Berry
tomers perceptions of the value created for them by logistics 1985, 1988, 1991, 1993, 1994; Spreng and Singh 1993).
services, because it is the customers perspective of service Several researchers have argued for the addition of items
quality that determines their satisfaction level. and/or dimensions to SERVQUAL. For example, from a less
Mentzer, Gomes, and Krapfel (1989) argue that two ele- sociological and more operational perspective, Crosby
ments exist in service delivery: marketing customer service (1979) defines quality as conformance to requirements and
and physical distribution service (PDS). They recognize the argues that those requirements should be specifically
complementary nature of the two elements to satisfy the defined to measure quality. From Crosbys (1979) view and
customer and propose an integrative framework of customer the general total quality management perspective, certain
service. This view is shared by others (Rinehart, Cooper, aspects of quality (of services or otherwise) intuitively ought
and Wagenheim 1989) and is regarded as an intellectual base to be incorporated. Along these lines, in applying
for integrating marketing and logistics activities. Here, PDS SERVQUAL to measure perceived quality of retail financial
is composed of three crucial components: availability, time- services, Brown, Churchill, and Peter (1993, p. 138) note the
liness, and quality. We view PDS as a component of LSQ. omission of items we a priori thought would be critical to
FIGURE 2
Hypothesized Model of LSQ as a Process
H1c H1b
Timeliness H10
H1h H3
Order Release H1e
Quantities Order H4
H1f Accuracy H7
H1g H1i H5
H6 Satisfaction
H1l Order
Condition
Information
Quality H1j H8
H2a
H1k H 2b
H1m Order H9
Order Discrepancy
Quality H2c
Handling
H1n
H1p
H1o
Ordering
Procedures
Scale Item
Information Quality
IQ1 Catalog information is available.
IQ2 Catalog information is adequate.
Ordering Procedures
OP1 Requisitioning procedures are effective.
OP2 Requisitioning procedures are easy to use.
Order Accuracy
OA1 Shipments rarely contain the wrong items.
OA2 Shipments rarely contain an incorrect quantity.
OA3 Shipments rarely contain substituted items.
Order Condition
OC1 Material received from DLA depots is undamaged.
OC2 Material received direct from vendors is undamaged.
OC3 Damage rarely occurs as a result of the transport mode or carrier.
Order Quality
OQ1 Substituted items sent by DLA work fine.
OQ2 Products ordered from DLA meet technical requirements.
OQ3 Equipment and/or parts are rarely nonconforming.
Timeliness
TI1 Time between placing requisition and receiving delivery is short.
TI2 Deliveries arrive on the date promised.
TI3 The amount of time a requisition is on back-order is short.
Satisfaction
SA1 (1 = terrible, 5 = excellent) What is your general impression of the service DLA provides?
SA2 (1 = very dissatisfied,
5 = very satisfied) Which word best describes your feelings toward DLA?
SA3 (1 = very dissatisfied,
5 = very satisfied) How satisfied are you with DLA service?
Notes: All nine LSQ construct items were measured on a five-point Likert-like scale (1 = strongly disagree, 5 = strongly agree).
the scales (Anderson and Gerbing 1988). As such, our goal shown to be the most stable fit indices by Gerbing and
was to achieve a high level of scale reliability and validity Anderson (1992). The 2 statistics with corresponding
and ensure that we had measured each theoretical facet of degrees of freedom are included for comparison purposes
the intended construct. We evaluated the scales using CFA (Jreskog and Srbom 1996).
analyses for each of the four customer segment samples Using these criteria, a multisample test of the four seg-
general merchandise (n = 1765), textiles and clothing (n = ments, in which the parameter estimates were constrained to
446), electronics (n = 530), and construction equipment and be the same across the four segments (Model 1) (i.e., load-
supplies (n = 215). We evaluated the model fits using the ings, factor correlations, and error variances), resulted in
DELTA2 index, the RNI, and the CFI. These have been acceptable fits to the data (Table 3). Allowing the loadings
to be estimated independently from one another in the four Next we assessed the reliability of the measures. Within
segments resulted in similar fit statistics (Model 2). On the the CFA setting, composite reliability is calculated using the
basis of the 2 difference test suggested by Anderson and procedures outlined by Fornell and Larcker (1981) and
Gerbing (1988), the constrained and unconstrained mea- based on the work by Werts, Linn, and Jreskog (1974). The
surement models were found not to differ significantly. As a formula specifies that CR = (i)2/[(i)2+(i)], where
further examinination of the potential for differences, multi- CR = composite reliability for scale , i = standardized
sample tests were conducted on all possible pairs of the cus- loading for scale item i, and i = measurement error for
tomer segment samples. As with the four-sample test, fit scale item i. We also examined the parameter estimates and
indices were acceptable, and no significant differences were their associated t-values and assessed the average variance
found between Models 1 and 2 (Table 3). Similarly, no dif- extracted for each construct (Anderson and Gerbing 1988).
ferences were found between the models when the error As is shown in Table 4, the reliabilities for the ten constructs
variances were allowed to be estimated freely in addition to ranged between .76 (order quality for construction segment)
the loadings (Model 3) or when the loadings were allowed and .95 (personnel contact quality for general, textiles, and
to be invariant but the error variances were allowed to differ electronics segments), indicating acceptable levels of relia-
(Model 4). bility for the constructs (Fornell and Larcker 1981). The
Segment Pairings
aModel 1 and Model 2 comparison by means of difference in 2 and degrees of freedom indicates no significant difference between the mod-
els. Thus, constructs are valid in four customer segments.
bModel 1 and Model 2 comparison by means of difference in 2 and degrees of freedom indicates no significant difference between the mod-
els. Thus, constructs are valid in all customer segments.
order quality scale is the only scale below a composite reli- phi coefficient () to unity and once freeing the parameter.
ability of .79, suggesting that all other scale reliabilities are We then used a 2 test to test for differences between mod-
excellent (Gerbing and Anderson 1992). els. In all cases, the 2 results were higher in the constrained
We established discriminant validity by calculating the models, thereby indicating discriminant validity between the
shared variance between all possible pairs of constructs and constructs. These results, in combination with fit indices for
verifying that they were lower than the average variance each customer segment sample (i.e., in Table 5, DELTA2,
extracted for the individual constructs (Fornell and Larcker RNI, and CFI exceed .90 for all four segments), suggest that
1981; Jreskog et al. 1999). The shared variance was calcu- the measurement scales are reliable and valid in all four cus-
lated as 2 = 1 , where 2 = shared variance between con- tomer segments in this study.
structs and the diagonal element of indicates the amount Finally, we examined the validity of each of the 28 indi-
of unexplained variance. Because and are standardized, vidual items in the analysis. First, we maintained our prede-
2 is equal to the r2 between the two constructs. We calcu- termined criteria of modification indices (<10) and residuals
lated average variance extracted using the following for- (<4). Second, we tested the potential differences among
mula: V = i2/(i2 + i), where V = average variance each item (28 items) across the four samples relative to its
extracted for , i = standardized loading for scale item i, theoretical construct (10 constructs). This test involved con-
and i = measurement error for scale item i. The shared straining appropriate sets of estimates, one parameter esti-
variances between pairs of all possible scale combinations mate at a time, to be equal and different across the four sam-
ranged from a low of 8% to a high of 59% between the var- ples (general, textiles, electronics, and construction) and
ious scale combinations (Table 4). The average variances then evaluating whether the resulting change in the 2 value
extracted ranged between 52% and 85%, all having higher was significant with the appropriate difference in degrees of
average variances extracted than the shared variances among freedom (Bagozzi and Heatherton 1994). The results indi-
all applicable pairs of scales (Table 4). To assess discrimi- cated that all 28 items were robust across the four samples.
nant validity further, in line with suggestions by Anderson The 2s ranged from .21 to 6.47 with a d.f. = 3, which
(1987) and Bagozzi and Phillips (1982), we assessed pairs was lower than the 2 value of 11.34 to be significant at the
of scales in a series of two-factor confirmatory models using p < .01 level. As such, the ten scales and their 28 items were
LISREL. We ran each model twiceonce constraining the considered reliable and valid in the context of this study.
Item Loading
Reliability
Variance Extracted General Textiles Electronics Construction
Highest Shared Variance (n = 1765) (n = 446) (n = 530) (n = 215)
Information Quality
IQ1 .81 .81 .85 .78
IQ2 .91 .90 .92 .86
Composite reliability .85 .85 .86 .84
Variance extracted 75% 74% 76% 73%
Highest shared variance 8% 13% 13% 26%
Ordering Procedures
OP1 .91 .88 .91 .85
OP2 .84 .84 .85 .79
Composite reliability .86 .86 .86 .85
Variance extracted 76% 76% 76% 74%
Highest shared variance 15% 25% 18% 32%
Order Accuracy
OA1 .89 .83 .91 .81
OA2 .90 .85 .91 .79
OA3 .79 .77 .77 .72
Composite reliability .89 .88 .89 .87
Variance extracted 73% 70% 72% 68%
Highest shared variance 49% 59% 52% 58%
Order Condition
OC1 .92 .86 .93 .81
OC2 .86 .80 .90 .78
OC3 .66 .69 .66 .71
Composite reliability .85 .84 .86 .84
Variance extracted 66% 65% 67% 65%
Highest shared variance 49% 59% 52% 58%
Order Quality
OQ1 .65 .63 .70 .62
OQ2 .83 .77 .86 .73
OQ3 .78 .71 .81 .72
Composite reliability .79 .77 .81 .76
Variance extracted 56% 53% 59% 52%
Highest shared variance 26% 36% 29% 50%
Item Loading
Reliability
Variance Extracted General Textiles Electronics Construction
Highest Shared Variance (n = 1765) (n = 446) (n = 530) (n = 215)
Timeliness
TI1 .91 .93 .88 .88
TI2 .94 .93 .93 .90
TI3 .93 .93 .92 .90
Composite reliability .94 .95 .94 .94
Variance extracted 85% 85% 85% 84%
Highest shared variance 12% 23% 13% 24%
Satisfaction
SA1 .83 .84 .85 .83
SA2 .92 .92 .92 .92
SA3 .92 .92 .93 .92
H1d (PQ TI) + .20 Supported .15 Supported .13 Supported .23 Supported
t-Value 7.73 2.89 2.84 3.15
H1e (OR OA) + .25 Supported .26 Supported .23 Supported .34 Supported
t-Value 8.45 4.39 3.96 4.35
H1f (OR OC) + .25 Supported .38 Supported .21 Supported .37 Supported
t-Value 8.29 6.19 3.75 4.80
H1g (OR OQ) + .24 Supported .31 Supported .34 Supported .49 Supported
t-Value 7.73 4.71 5.71 5.40
H1h (OR TI) + .16 Supported .09 .21 Supported .34 Supported
t-Value 5.10 1.26 3.51 2.69
H1k (IQ OQ) + .11 Supported .17 Supported .17 Supported .08
t-Value 3.67 2.97 3.36 .98
H1m (OP OA) + .23 Supported .25 Supported .13 .47 Supported
t-Value 7.68 4.50 2.25 4.35
H1n (OP OC) + .19 Supported .20 Supported .13 .37 Supported
t-Value 6.26 3.62 2.30 3.73
H2a (OA OD) + .11 Supported .16 Supported .23 Supported .26 Supported
t-Value 4.51 3.06 4.84 3.14
H2b (OC OD) + .24 Supported .17 Supported .24 Supported .13
t-Value 9.53 3.27 5.12 1.63
H2c (OQ OD) + .42 Supported .42 Supported .25 Supported .47 Supported
t-Value 14.29 6.75 5.04 4.74
H9 (OP SA) + .41 Supported .35 Supported .46 Supported .57 Supported
t-Value 14.05 6.48 8.68 5.22
Fit Indices
DELTA2 .89 .88 .87 .89
RNI .89 .87 .87 .88
CFI .89 .87 .87 .88
2 4622.68 1798.98 1840.50 1064.89
d.f. 368 368 368 368
Notes: PQ = personnel contact quality, OR = order release quantities, IQ = information quality, OP = ordering procedures, OA = order accuracy,
OC = order condition, OQ = order quality, OD = order discrepancy handling, TI = timeliness, and SA = satisfaction. Significant hypothe-
ses are supported at p < .01.
PQ
PQ
TI
TI
OR
OR
OA
OA
SA
SA
IQ OC
IQ OC
OQ OD
OQ OD
OP
OP
Notes: PQ = personnel contact quality, OR = order release quanti- Notes: PQ = personnel contact quality, OR = order release quanti-
ties, IQ = information quality, OP = ordering procedures, OA ties, IQ = information quality, OP = ordering procedures, OA
= order accuracy, OC = order condition, OQ = order quality, = order accuracy, OC = order condition, OQ = order quality,
OD = order discrepancy handling, TI = timeliness, and SA = OD = order discrepancy handling, TI = timeliness, and SA =
satisfaction. satisfaction.
IQ1 IQ2 OP1 OP2 OR1 OR2 OR3 TI1 TI2 TI3 OA1 OA2 OA3 OQ1 OQ2 OQ3 OC1 OC2 OC3 OD1 OD2 OD3 PQ1 PQ2 PQ3 SA1 SA2 SA3
IQ1 IQ2 OP1 OP2 OR1 OR2 OR3 TI1 TI2 TI3 OA1 OA2 OA3 OQ1 OQ2 OQ3 OC1 OC2 OC3 OD1 OD2 OD3 PQ1 PQ2 PQ3 SA1 SA2 SA3
IQ1 1.0
IQ2 .74 1.0
OP1 .27 .34 1.0
OP2 .28 .35 .83 1.0
OR1 .19 .22 .39 .33 1.0
OR2 .19 .25 .41 .41 .56 1.0
OR3 .26 .32 .48 .45 .50 .82 1.0
TI1 .16 .21 .26 .24 .18 .30 .36 1.0
TI2 .16 .20 .21 .20 .21 .33 .38 .93 1.0
TI3 .16 .20 .24 .22 .23 .32 .37 .92 .91 1.0
OA1 .25 .28 .41 .37 .30 .36 .35 .24 .24 .24 1.0
OA2 .26 .32 .43 .37 .33 .39 .39 .26 .25 .26 .81 1.0
OA3 .31 .32 .38 .38 .33 .34 .35 .28 .28 .30 .69 .74 1.0
OQ1 .22 .23 .28 .28 .26 .32 .34 .41 .42 .41 .39 .39 .50 1.0
OQ2 .27 .29 .31 .32 .33 .33 .34 .31 .33 .31 .44 .46 .50 .50 1.0
OQ3 .16 .22 .23 .24 .27 .31 .32 .39 .40 .38 .41 .44 .45 .48 .60 1.0
OC1 .24 .27 .40 .38 .35 .45 .42 .31 .31 .30 .69 .68 .66 .41 .45 .42 1.0
OC2 .22 .24 .37 .37 .37 .39 .39 .27 .28 .26 .59 .61 .64 .42 .46 .44 .79 1.0
OC3 .26 .28 .29 .33 .32 .38 .40 .30 .31 .32 .59 .59 .57 .42 .45 .49 .66 .60 1.0
OD1 .20 .18 .36 .32 .36 .40 .42 .30 .33 .33 .38 .42 .44 .38 .41 .33 .42 .36 .41 1.0
OD2 .20 .24 .35 .31 .33 .35 .36 .30 .30 .32 .37 .42 .42 .38 .41 .32 .42 .36 .41 .78 1.0
OD3 .21 .22 .24 .25 .29 .32 .34 .32 .34 .33 .31 .36 .33 .35 .39 .38 .37 .32 .41 .60 .70 1.0
PQ1 .20 .24 .35 .28 .31 .38 .37 .38 .36 .37 .29 .31 .31 .37 .32 .25 .30 .24 .26 .30 .32 .29 1.0
PQ2 .20 .25 .30 .25 .31 .42 .39 .32 .31 .32 .29 .34 .31 .36 .26 .25 .32 .25 .27 .31 .34 .33 .87 1.0
PQ3 .20 .25 .33 .27 .32 .41 .42 .37 .36 .36 .31 .34 .33 .38 .31 .27 .36 .28 .28 .31 .33 .34 .82 .87 1.0
SA1 .18 .12 .35 .32 .01 .14 .21 .34 .30 .29 .16 .18 .17 .33 .24 .18 .18 .17 .17 .26 .23 .22 .26 .26 .27 1.0
SA2 .14 .15 .38 .35 .06 .18 .22 .29 .24 .23 .16 .18 .13 .28 .19 .11 .16 .10 .12 .30 .27 .24 .30 .27 .28 .80 1.0
SA3 .12 .16 .37 .35 .03 .16 .21 .29 .23 .23 .19 .20 .18 .27 .19 .13 .15 .10 .15 .27 .25 .25 .29 .27 .26 .79 .89 1.0
IQ1 IQ2 OP1 OP2 OR1 OR2 OR3 TI1 TI2 TI3 OA1 OA2 OA3 OQ1 OQ2 OQ3 OC1 OC2 OC3 OD1 OD2 OD3 PQ1 PQ2 PQ3 SA1 SA2 SA3
IQ1 IQ2 OP1 OP2 OR1 OR2 OR3 TI1 TI2 TI3 OA1 OA2 OA3 OQ1 OQ2 OQ3 OC1 OC2 OC3 OD1 OD2 OD3 PQ1 PQ2 PQ3 SA1 SA2 SA3
IQ1 1.0
IQ2 .73 1.0
OP1 .51 .48 1.0
OP2 .42 .38 .76 1.0
OR1 .20 .23 .36 .36 1.0
OR2 .20 .23 .39 .40 .63 1.0
OR3 .26 .22 .43 .47 .51 .77 1.0
TI1 .21 .14 .37 .39 .31 .44 .38 1.0
TI2 .24 .20 .37 .37 .35 .50 .43 .88 1.0
TI3 .22 .17 .33 .35 .33 .47 .37 .89 .90 1.0
OA1 .34 .36 .45 .52 .44 .44 .39 .42 .40 .38 1.0
OA2 .28 .30 .43 .48 .38 .42 .41 .39 .38 .35 .79 1.0
OA3 .32 .34 .35 .44 .44 .41 .38 .31 .34 .32 .71 .68 1.0
OQ1 .29 .37 .34 .42 .40 .38 .30 .31 .33 .29 .52 .45 .54 1.0
OQ2 .25 .30 .45 .44 .45 .48 .42 .34 .32 .35 .54 .52 .51 .56 1.0
OQ3 .33 .36 .45 .45 .48 .45 .43 .30 .30 .31 .57 .50 .47 .56 .67 1.0
OC1 .38 .32 .46 .53 .40 .39 .38 .34 .38 .32 .69 .60 .62 .50 .56 .63 1.0
OC2 .35 .40 .40 .48 .52 .49 .45 .30 .38 .34 .68 .58 .63 .53 .54 .63 .82 1.0
OC3 .31 .29 .45 .50 .46 .47 .36 .32 .29 .35 .68 .62 .52 .44 .57 .65 .74 .68 1.0
OD1 .30 .33 .38 .41 .47 .48 .42 .31 .34 .38 .51 .47 .50 .48 .50 .55 .49 .53 .47 1.0
OD2 .24 .19 .35 .41 .43 .46 .43 .34 .37 .40 .50 .54 .50 .48 .47 .50 .51 .46 .52 .78 1.0
OD3 .25 .19 .35 .38 .45 .45 .41 .43 .48 .47 .52 .49 .49 .55 .44 .49 .53 .52 .49 .67 .74 1.0
PQ1 .25 .25 .38 .36 .31 .35 .35 .37 .41 .41 .33 .34 .31 .23 .34 .36 .42 .39 .39 .40 .38 .39 1.0
PQ2 .21 .22 .36 .37 .34 .38 .34 .42 .45 .45 .36 .38 .31 .25 .29 .35 .39 .38 .41 .43 .42 .42 .90 1.0
PQ3 .23 .26 .32 .31 .39 .45 .39 .38 .41 .42 .36 .36 .33 .24 .30 .38 .35 .37 .39 .44 .39 .44 .84 .85 1.0
SA1 .25 .22 .33 .23 .04 .03 .02 .19 .13 .13 .22 .23 .18 .19 .14 .08 .12 .09 .20 .19 .18 .20 .16 .16 .13 1.0
SA2 .27 .23 .37 .27 .02 .02 .02 .17 .13 .12 .24 .24 .16 .16 .10 .08 .19 .11 .14 .22 .20 .24 .20 .18 .12 .76 1.0
SA3 .24 .20 .33 .27 .01 .03 .00 .21 .19 .16 .21 .21 .12 .19 .07 .06 .13 .09 .16 .18 .19 .23 .14 .15 .11 .75 .86 1.0
Notes: PQ = personnel contact quality, OR = order release quantities, IQ = information quality, OP = ordering procedures, OA = order accuracy, OC = order condition, OQ = order quality, OD =
order discrepancy handling, TI = timeliness, and SA = satisfaction.