Bou-Llusar Et Al 2009 EFQM Excellence Model

8/10/2019 Bou-Llusar Et Al 2009 EFQM Excellence Model

http://slidepdf.com/reader/full/bou-llusar-et-al-2009-efqm-excellence-model 1/22

An empirical assessment of the EFQM Excellence Model:Evaluation as a TQM framework relative to the MBNQA Model

J. Carlos Bou-Llusar *, Ana B. Escrig-Tena, Vicente Roca-Puig, Inmaculada Beltra ń-Martı ńDepartamento de Administracio ´ n de Empresas y Marketing, Universitat Jaume I, Campus Riu Sec - Avda. Sos Baynat s/n, 12071 Castello ´ n, Spain

1. Introduction

Since the 1990s, most rms have used the modelsunderpinning quality awards, such as the Deming Prize(DP Model) in Japan, the Malcolm Baldrige NationalQuality Award (MBNQA) Model in the USA, and theEuropean Quality Award (EFQM Excellence Model) inEurope, as a framework for implementing TQM initia-tives. Many researchers have considered quality modelsas operational frameworks for TQM (e.g., Bohoris, 1995;Ghobadian and Woo, 1996; Curkovic et al., 2000; Van derWiele et al., 2000; Yong and Wilkinson, 2001; Lee et al.,2003 ). These authors consider that quality awardmodels reproduce TQM by capturing its main constitu-

ent parts and by replicating its core ideas in clear andaccessible language. Nevertheless, the empirical valida-tion of the extent to which these models reproduce TQMis scarce, partial, and limited to some empirical studiessuch as Curkovic et al. (2000) , who conclude thatMBNQA and its criteria do capture TQM core concepts.In the context of the EFQM Excellence Model, thisquestion remains unanswered and, therefore, moreresearch is needed.

The purpose of this paper is to understand the EFQMmodel as a framework for TQM, that is, to analyze whetherthe internal structure of the EFQM Excellence Model takesinto account the basic TQM assumptions. As McAdam andLeonard (2005) point out, there is a paucity of studies onthe effectiveness of quality award models for developingTQM in organizations. By improving the understanding of the internal structure of quality award models, there is anopportunity to assess the application of TQM.

Journal of Operations Management 27 (2009) 1–22

A R T I C L E I N F O

Article history:Received 28 September 2006Received in revised form 4 April 2008Accepted 9 April 2008Available online 20 April 2008

Keywords:Total quality managementEFQM Excellence ModelMBNQAStructural equation modelsEmpirical research

A B S T R A C T

Total quality management (TQM) is an approach to management embracing both socialand technical dimensions aimed at achieving excellent results, which needs to be put intopractice through a specic framework. Nowadays, quality award models, such as theMalcolm Baldrige National Quality Award (MBNQA) and the European Foundation forQuality Management (EFQM) Excellence Model, are used as a guide to TQMimplementation by a large number of organizations. Nevertheless, there is a paucity of empirical research conrming whether these models clearly reect the main premises of TQM. The purpose of this paper is to analyze the extent to which the EFQM ExcellenceModel captures themain assumptions involved in theTQM concept, that is, the distinctionbetween technical and social TQM issues, the holistic interpretation of TQM in the rm,and the causal linkage between TQM procedures and organizational performance.

Based on responses collected from managers of 446 Spanish companies by means of astructured questionnaire, we nd that: (a) social and technical dimensions are embedded

in the model; (b) both dimensions are intercorrelated; (c) they jointly enhance results.These ndings support the EFQM Excellence Model as an operational framework for TQM,and also reinforce the results obtained in previous studies for the MBNQA, suggesting thatquality award models really are TQM frameworks.

2008 Elsevier B.V. All rights reserved.

* Corresponding author. Tel.: +34 964 72 85 34; fax: +34 964 72 86 29.E-mail address: [email protected] (J.C. Bou-Llusar).

Contents lists available at ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

0272-6963/$ – see front matter 2008 Elsevier B.V. All rights reserved.doi: 10.1016/j.jom.2008.04.001

mailto:[email protected]

http://www.sciencedirect.com/science/journal/02726963

http://dx.doi.org/10.1016/j.jom.2008.04.001

http://dx.doi.org/10.1016/j.jom.2008.04.001

http://www.sciencedirect.com/science/journal/02726963

mailto:[email protected]



Table 1Different views of the core concepts which constitute TQM and their embedding in the TQM framework

TQM core concepts TQM frameworks b

Anderson et al. (1994) Powell (1995) Tummala y Tang (1996) Sila and Ebrahimpour (2002) a EFQM Criteria (2

Customer satisfaction Closer customerrelationship

Customer focus Customer focus and satisfaction 5. Processes

Visionary leadership Committed leadership Leadership Leadership and topmanagement commitment

1. Leadership

Adoption andcommunication of TQM

Strategic quality planning 2. Policy and strategy

Continuous improvement Continuous improvement Continuous improvementand innovation

Process management Process improvement Design quality, speed

and prevention

Process management 5. Processes

Zero-defects mentalityFlexible manufacturing

Internal cooperation Increased training People participationand partnership

Employee training 3. People

Learning Employee Empowerment TeamworkEmployee fulllment Open organization Employee involvement

Everybody’s participation

External cooperation Benchmarking 4. Partnership andresources

Closer supplierrelationshipMeasurement Fact-based management Quality information and

performance measurement6. Customer resul7. People Results8. Society results9. Key performan

a Summary of core concepts after analyzing 347 survey-based articles published between 1989 and 2000.







The paper is structured as follows. In the next section,we present a review of previous literature in order topresent quality award models as TQM implementationframeworks. Based on the review of the literature, aresearch model to assess the capability of the EFQMExcellence Model to reproduce the TQM concept is offeredin Section 3. Section 4 describes the methodology, and

attention is paid to the survey procedure and theconstruction of measures. Finally, Sections 5 and 6 presentthe results of the empirical study carried out and the mainconclusions and implications stemming from this research.

2. Review of the literature

2.1. Denition of TQM

A variety of denitions of total quality management(TQM) have been offered over the years. Reviewingprevious contributions (e.g. Dean and Bowen, 1994; Sitkinet al., 1994; Hackman and Wageman, 1995; Wilkinson

et al., 1998; Oakland, 2000; Dale, 2003; Eriksson andGarvare, 2005 ) a dominant insight among experts seems todene TQM as an approach to management characterizedby some guiding principles or core concepts that embodythe way the organization is expected to operate, which,when effectively linked together, will lead to highperformance. Although with some differences, there is ageneral agreement regarding the assumptions included inthe TQM concept, which can be summarized in three mainpoints.

Firstly, the core concepts of TQM can be classied intotwo broad categories or dimensions: social or soft TQM,and technical or hard TQM ( Dotchin and Oakland, 1992;

Yong and Wilkinson, 2001; Prajogo and Sohal, 2004;Rahman, 2004; Rahman and Bullock, 2005; Lewis et al.,2006 ). The social issues are centered on human resourcemanagement and emphasize leadership, teamwork, train-ing, and employee involvement. The technical issuesreect an orientation toward improving productionmethods and operations and seek to establish a workingmethod through the establishment of well-dened pro-cesses and procedures to make possible the constantimprovement of goods and services to customers.

Secondly, the management of social or technical TQMissues cannot be performed in isolation. Social andtechnical dimensions (and the core concepts that form

them) should be interrelated and mutually support oneother ( Flynn et al., 1994; Wruck and Jensen, 1994;Hackman and Wageman, 1995; Sun, 1999 ) reecting theholistic character of TQM initiatives. This holistic characteris also extended to the expected results of a TQM initiative,as a balance of the stakeholders’ interests should beconsidered when the rm denes TQM practices ( StainerandStainer, 1995; Oakland andOakland, 1998; Fissher andNijhof, 2005 ).

Thirdly, the literature suggests that the optimalmanagement of TQM core concepts will lead to betterorganizational performance, as studies such as Powell(1995) , Terziovski and Samson (1999) , Zhang (2000) ,

Hendricks and Singhal (2001) , or Kaynak (2003) haveveried. The basic theoretical foundation for this relation- T a b

l e 2

( C o n t i n u e

d )

A u t h o r s

A w a r d

A p p r o a c h

M a i n p u r p o s e o f t h e p a p e r

M a i n r e l a t i o n s

f o u n

d

B a d r i e t a l . ( 2 0 0 6 )

B a l d r i g e E d u c a t i o n

C r i t e r i a

f o r P e r f o r m a n c e E x c e l l e n c e

M o d e l

2 0 0 4

C a u s a l

T o e m p

i r i c a l l y t e s t t h e c a u s a l r e l a t i o n s h

i p s

i n t h e M

B N Q A E d u c a t i o n P e r f o r m a n c e

E x c e l l e n c e

C r i t e r i a a n

d t o d e v e l o p a

c o m p r e

h e n s i v e m e a s u r e m e n t m o d e l

L e a d e r s h i p i s i d e n t i e d a s a

d r i v e r f o r a l l c o m p o n e n t s i n t h e B a l d r i g e

S y s t e m

, i n c l u d

i n g m e a s u r e m e n t , a n a l y s i s a n

d k n o w

l e d g e m a n a g e m e n t ,

s t r a t e g i c p l a n n

i n g ,

f a c u l t y a n

d s t a f f

f o c u s a n

d p r o c e s s m a n a g e m e n t . A l l

B a l d r i g e c o m p o n e n t s a r e s i g n

i c a n t l y l i n k e d w

i t h o r g a n

i z a t i o n a l

o u t c o m e s a s r e p r e s e n t e d b y t w o c a t e g o r i e s : o r g a n

i z a t i o n a l p e r f o r m a n c e

r e s u

l t s , a n

d s t u

d e n t , s t a k e h o l d e r a n

d m a r k e t f o c u s

J.C. Bou-Llusar et al. / Journal of Operations Management 27 (2009) 1–22 5





3.2. The holistic interpretation of TQM in the EFQM ExcellenceModel

The internal structure of the EFQM model should alsoreect the holistic character of the TQM initiatives andconsider the interrelationships in both the enabler and theresult criteria.

3.2.1. Interrelationships within the enabler domainThe implementation of quality practices in concert with

one another is necessary in order to realizethe full benetsof TQM ( Tamimi, 1998 ). Some empirical work supports theexistence of interrelationships between the enabler side of

the EFQM Excellence Model ( Dijkstra, 1997; Eskildsen,1998; Eskildsen and Dahlgaard, 2000; Prabhu et al., 2000;Reiner, 2002; Bou-Llusar et al., 2005 ), based on theassumption that these criteria are components of theunique TQM philosophy. As Eskildsen et al. (2000) suggest,previous research on the causal structure of the EFQMExcellence Model has shown that the enabler criteria arelinked together in a very complex structure, making it very

difcult to discern between them. Enabler excellence is thusinterpreted in this study as the overall approach that rmsshould adopt when they implement the EFQM, and this isreected in the level achieved by the rm in both the socialand technical dimensions, together with the policy andstrategy criteria. According to this interpretation of theenabler side of the EFQM Excellence Model, changes in onedimension are related to changes in other dimensions, andthere is thereforea reciprocal interdependencebetween allenabler components. This interdependence is representedby the common latent factor enabler excellence (see the leftside of Fig. 2).

This conceptualization of the enabler side of the model

maintains clear parallelisms with the study by Dijkstra(1997) , who maintains the existence of a common latentgeneral factor that causes the associations between theenablers. Likewise, Dow et al. (1999) show that the mainquality management dimensions are usually implementedin combination with one other, and that they show a highlevel of correlation with the other dimensions. Our secondresearch question addresses this issue.

Table 3The EFQM Excellence Model criteria

Criterion Denition

Leadership Excellent leaders develop and facilitate the achievement of the mission and vision. They develop

organisational values and systems required for sustainable success and implement these via theiractions and behaviours

Policy and strategy Excellent organisations implement their mission and vision by developing a stakeholder focused strategythat takes account of the market and sector in which it operates. Policies, plans, objectives and processesare developed and deployed to deliver strategy

People Excellent organisations manage, develop and release the full potential of their people at an individual,team-based and organisational level. They promote fairness and equality and involve and empower their people

Partnerships and resources Excellent organisations plan to manage external partnerships, suppliers and internal resources in orderto support policy and strategy and the effective operation of processes

Processes Excellent organisations design, manage and improve processes in order to fully satisfy, and generateincreasing value for, customers and other stakeholders

Customer results Excellent organisations comprehensively measure and achieve outstanding results with respect to their customersPeople results Excellent organisations comprehensively measure and achieve outstanding results with respect to their peopleSociety results Excellent organisations comprehensively measure and achieve outstanding results with respect to societyKey performance results Excellent organisations comprehensively measure and achieve outstanding results with respect to the key

element of their policy and strategy

Source : adapted from EFQM (2003) .

Fig. 1. The EFQM Excellence Model. Source : EFQM (2003)






TQM framework. In order to avoid this limitation, it isnecessary to test whether all enablers explain theexcellence in the result domain. In our model, this matteris addressed in following research question and isimplemented considering the inuence of enabler excel-lence on result excellence.

RQ4: Will enablers have a positive inuence on resultsin the EFQM Excellence Model?

In summary, our analysis of the EFQM Excellence Modelhas led to the structural model illustrated in Fig. 2. Theempirical validation of the proposed structural model willenable us to examine the extent to which the TQMdimensions (social vs. technical) are included in the EFQMExcellence Model. Moreover, the model will provide someinsights in how excellence in the enablers explains theachievement of excellent results by simultaneously con-sidering the intercorrelations between all the elements of the model caused by the latent factors enabler excellenceand results excellence. The validation of this model will

allow us to determine whether EFQM effectively reectsthe main TQM assumptions and could be considered anoperational framework for TQM.

4. Methodology

4.1. Sample

The multidimensional structural model proposed toanalyze the EFQM framework was tested using relevantdata from a survey on quality practices, competences,and performance compiled by the authors. The sample of companies was obtained from the Spanish national

‘‘ARDAN’’ information service which provides informa-tion on more than 100,000 rms. In ARDAN rms can beselected according to various classication criteria suchas sector, name, activity, size or location. In order toavoid possible bias in the selection of rms in thesample, we used the stratied sampling method. Thedistribution of the sample throughout the differentgroups was performed by dividing it into proportionalparts of the population of each stratum according tosector and size. The division by sectors was madeaccording to their SIC code, including industrial andservice sectors. As in other TQM studies, we includedifferent service sectors with different degrees of

interaction and adjustment to the customer, anddifferent intensity of labor ( Dotchin and Oakland,1994; Silvestro, 1999 ). We include different industrialsectors with different degree of complexity of thetransaction ( De Vasconcellos and Hambrick, 1989 ), arelevant variable to evaluate the importance of quality inthe sector. Within each of these sub-samples, small,medium, and large companies were analyzed. Accordingto European Union criteria ( Recommendation of Eur-opean Commission 96/280/CE ) three size segments weredened: small (10–49 workers), medium-sized (50–249workers) and large companies (250 or more workers).

The eldwork was carried out during October and

December 2000 by means of a structured questionnaire tothe CEO or the quality manager in the companies, and 446

valid responses were included in our analysis. All itemswere measured in a 7-point Likert scale, where 1represented a very low score and 7 a very high one.According to ARDAN the population was 2695 rms, whichconsidering a condence level of 95% implies a sampleerror of 3.28% ( p = q = 50%) for the overall sample. Thesample represented a good cross-section in terms of sectors

and size. Of the 446 companies 52% belonged to manufactur-ing sectors, and 48% to service sectors. With regard to size,50% were small rms, 42% medium-sized rms, and 8% largerms. The average number of employees per rm for thewhole sample was 93.17 (standard deviation = 219.48). Thepredominance of small and medium-sized rms is repre-sentative of Spanish industry.

4.2. Statistical procedure

The empirical validation of the model was carried outusing structural equations modeling (SEM). SEM allows forthe introduction of latent variables that can only be

measured through observable indicators. In this research,enabler and result excellence , as well as the social andtechnical dimensions are variables that cannot be directlyobserved. Moreover, SEM takes into account the existenceof measurement error, and offers the possibility of simultaneously estimating all the relationships proposedin the theoretical model, thus attaining a completerepresentation of the model. SEM is thus a suitablemethodology to test in a single model the relationshipsbetween enabler excellence and TQM dimensions (factorialapproach) and the causal relationship between the enabler excellence and result excellence (causal approach). We usethe statistical software EQS 6.1 ( Bentler, 1995 ), using the

maximum likelihood estimation method. To protect ourresults for possible deviations of normality assumption, allthe chi-square values (as well standard errors) reportedcorrespond to Satorra and Bentler (1994) scaled goodness-of-t test statistics. For the chi-square scaled differencetest statistics, we used the Satorra and Bentler (2001)procedure to compute the appropriate value.

4.3. Measures

Following the self-assessment philosophy for the EFQMExcellence Model, each EFQM criterion was operationa-lized by selecting a set of most relevant items from an

original questionnaire about quality practices elaboratedby the authors. The initial set of items used to operatio-nalize the EFQM criteria is shown in Appendix A . Onehundred and six items were initially selected, 74 tooperationalize the 5 enabler criteria and 32 for the resultcriteria. The items measured quality practicesassociatedtothe EFQM criteria and were assigned to each EFQM sub-criterion based on its content domain. An attempt wasmade to include several items for every sub-criterion tocover the entire set of practices associated to them.However, in some sub-criteria there were not enoughitems in the questionnaire to reect the constructaccurately, or no possibility of getting a meaningful

distinction between sub-criteria. This was particularlyimportant in the case of the ‘‘Partnerships and Resources’’




and ‘‘Processes’’ criteria, where some sub-criteria werecombined (see Appendix A ).

Next, to measure thesub-criteria in theenabler domain,items sharing the same sub-criterion were averaged toform composite measures ( Landis et al., 2000 ), alsoreferred to as testlets (Wainer and Kiely, 1987 ) or item parcels (Bandalos and Finney, 2001 ). Composite measures

are combination of items to create score aggregates thatare then subjected to conrmatory factor analyses (CFA) asindicator variables in the scale validation process. In CFA,the use of composite measures is useful by two reasons.Firstly, it enables to better meet the normal-distributionassumption of maximum likelihood estimation. Secondly,it results in more parsimonious models because it reducesthe number of variances and covariances to estimate, thusincreasing the stability of the parameter estimates,improving the variable-to-sample-size ratio and reducingthe impact for sampling error on the estimation process(Bagozzi and Edwards, 1998; Bandalos and Finney, 2001;McCallum et al., 1999; Little et al., 2002 ). Thus, a composite

measure for each sub-criterion was introduced as anindicator variable in the analyses conducted to assess thedimensionality, reliability and validity of the enablercriteria.

By contrast with the enabler criteria, the EFQMExcellence Model comprises only two sub-criteria for eachresult criterion, named ‘‘perception measures’’ and ‘‘per-formance indicators’’. The former refers to informationgathered from the organization’s stakeholders (customers,people, society and shareholders), whereas the ‘‘perfor-mance indicators’’ are based on self-assessment by thecompany. As our research design was based on a surveyadministered to the CEO or quality manager of the

company, data from customers, employees or otherstakeholders was not available. So, in the result domain,all criteria were considered unidimensional, and the itemsselected were directly assigned to measuring eachcriterion and introduced in the CFA as indicator variables.Fig. 2, which represents the structural model proposed,also reects the way each enabler and result criterion hasbeen measured.

5. Results

5.1. Scale validation

To assure that all criteria had the desirable character-istics of dimensionality, reliability and construct (con-vergent and discriminant) validity, we conduct severalanalyses following the procedure recommended by Ahireet al. (1996) and Ahire and Ravichandran (2001) .

5.1.1. DimensionalityScale dimensionality was assessed by executing a CFA

for each criterion. The goodness-of-t values for the CFA(see Table 4 ) were all above the recommended values,indicating that all criteria possessed adequate unidimen-sionality, except for ‘‘People Result’’ ( x 2

ð14 Þ ¼ 132 : 13; CFI =0.840; RMSEA = 0.137; BBNFI = 0.760; x 2 /d.f. = 9.4) and‘‘Key Performance Results’’ ( x 2

ð27 Þ ¼ 553 : 55; CFI = 0.706;RMSEA = 0.219; BBNFI = 0.608; x 2 /d.f. = 20.5) criteria thatshowed a poor t. The Lagrange multiplier test (LMTEST)was used to introduce successive modications in thescales until the t indices reached values within therecommended limits. As Joreskog and Sorbom (1996)suggest, only one parameter was altered in each iterationto avoid over-modifying the model. As a result of thisprocess, both the ‘‘People Results’’ and ‘‘Key PerformanceResults’’ were considered three-dimensional constructs.The three dimensions for the ‘‘People Results’’ criteria werenamed ‘‘People Motivation’’, ‘‘People Achievement’’ and‘‘People Satisfaction’’; and ‘‘Financial Results’’, ‘‘Results onExternal Resources’’, and ‘‘Process Results’’ in the case of the ‘‘Key Performance Results’’ criterion. Those dimensionsfollow the proposed structure of the EFQM model when itdenes the questions to address in each of these criteria(see EFQM, 2003). In addition, 18 items were deletedbecause they showed low reliability and/or presentedsignicant cross-loadings (the eliminated items aremarked with an asterisk in Appendix A ). In summary,the dimensionality analysis reects that 7 of the 9 EFQMExcellence Model’s criteria were unidimensional, and 2were three-dimensional, and that 88 items (65 for enablers

Table 4Unidimensionality and reliability of the scales

Construct Unidimensionality Reliability

S-B x 2 a d.f. p-Value CFI RMSEA Cronbach’s alpha Composite reliability

Leadership 14.196 5 0.01441 0.985 0.063 0.839 0.851Policy and strategy 1.198 2 0.549 1.000 0.000 0.801 0.818People 19.528 5 0.015 0.982 0.080 0.882 0.899Resources 21.941 b 17 0.187 0.994 0.026 0.714 0.722Process 21.941 b 17 0.187 0.994 0.026 0.733 0.779Customer results 3.254 2 0.197 0.998 0.037 0.873 0.826People motivation 12.823 11 0.305 0.998 0.019 0.642 0.663People achievement 12.823 11 0.305 0.998 0.019 0.764 0.764People satisfaction 12.823 11 0.305 0.998 0.019 0.840 0.842Results on society 21.941 b 17 0.187 0.994 0.026 0.909 0.910Financial results 17.763 24 0.814 1.000 0.000 0.903 0.905Results on external resources 17.763 24 0.814 1.000 0.000 0.841 0.884Process results 17.763 24 0.814 1.000 0.000 0.878 0.876

a Satorra–Bentler scaled chi-square ( Satorra and Bentler, 1994 ).b

To avoid negative degrees of freedom, a pooled measurement model was executed, with indicators loading on the corresponding criteria ( Ahire andRavichandran, 2001 ), for those criteria with less than four indicators (i.e., ‘‘Partnership and Resources’’, ‘‘Processes’’ and ‘‘Results on Society’’).

J.C. Bou-Llusar et al. / Journal of Operations Management 27 (2009) 1–2210



and 23 for results) from the 106 initially proposed were

retained in the measurement scales.

5.1.2. ReliabilityCronbach’s alpha ( Cronbach, 1951 ) and composite

reliability ( Fornell and Larcker, 1981 ) were used to assessthe reliability of the scales. Table 4 shows the values of both indices. These are all above the 0.7, except for ‘‘Peoplemotivation’’, which showed reliability coefcients slightlybelow from the threshold. However, as leaving them outwould affect the content validity of the model, we decidedto retain it.

5.1.3. Convergent validity

Convergent validity was assessed using the Bentler–Bonett normed t index (BBNFI) ( Bentler and Bonett,1980 ). BBNFI is the ratio of the difference between themodel chi-square for the given model minus the modelchi-square for the null model (i.e., the ‘‘independence’’model in which all of the correlations are zero), divided bymodel chi-square for the null model. A BBNFI greater than0.90 indicates strong convergent validity ( Ahire et al.,1996 ). Considering this cut-off value, a high level of

convergent validity was found in all criteria, as Table 5

shown.

5.1.4. Discriminant validityThree approaches were used to assess discriminant

validity ( Ghiselli et al., 1981; Bagozzi and Phillips, 1982 ).First, for all scales Cronbach’s alpha was higher than theaverage interscale correlation (AVISC) (see 4th column inTable 5 ). Second, the average correlation between the scaleand non-scale items(6th column in Table 5 ) was lower thanbetween the scale and scale items (5th column in Table 5 ).Finally, for each pair of criteria, we conducted a ‘‘pair-wisetest’’ ( Bagozzi and Phillips, 1982 ) to testwhether a CFA withtwo factors ts the data signicantly better than a single-

factor model; that is, whether each criterion really doesrepresent a different concept. A statistically signicantdifference between the chi-squared values for the twomodels (degree of freedom = 1) will enable us tosupport theexistence of discriminant validity. The scaled chi-squaredifference values ( Satorra and Bentler, 2001 ) for all pairswere found to be statistically signicant at 5% level (andhence they are not presented), providing evidence of theexistence of discriminant validity.

Table 6Means, standard deviations, correlations

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)LeadershipPolicy and strategy 0.790People 0.804 0.766Partnership and resources 0.724 0.691 0.706Process 0.687 0.693 0.613 0.723Customer results 0.544 0.495 0.541 0.533 0.478People motivation 0.525 0.421 0.451 0.436 0.442 0.362People achievement 0.682 0.534 0.638 0.567 0.513 0.442 0.664People satisfaction 0.469 0.424 0.505 0.452 0.342 0.589 0.409 0.537Society results 0.430 0.340 0.411 0.504 0.407 0.523 0.323 0.329 0.454Financial results 0.411 0.344 0.351 0.477 0.460 0.456 0.337 0.349 0.364 0.417External results 0.442 0.393 0.422 0.505 0.471 0.564 0.353 0.416 0.495 0.445 0.490Processes results 0.553 0.491 0.560 0.529 0.556 0.606 0.389 0.460 0.421 0.399 0.462 0.512

Mean 5.789 5.672 5.176 5.192 5.415 5.905 5.615 5.359 5.372 5.271 5.399 5.186 5.788

S.D. 0.688 0.755 1.080 0.960 0.953 0.876 1.166 1.040 1.271 1.279 1.151 1.149 0.963

Table 5Convergent and discriminant validity

Convergent validity Discriminant validity

BBNFI Average interscalecorrelation (AVISC)

Cronbach’salpha (AVISC) a

Average item to scale correlations

Scale items Non-scale items

Leadership 0.978 0.588 0.251 0.539 0.473Policy and strategy 0.998 0.532 0.269 0.530 0.434People 0.976 0.564 0.318 0.639 0.429Partnership and resources 0.985 0.571 0.143 0.647 0.431Process 0.985 0.532 0.201 0.614 0.413Customer results 0.993 0.511 0.362 0.857 0.335People motivation 0.983 0.426 0.216 0.859 0.307People achievement 0.983 0.511 0.253 0.825 0.378People satisfaction 0.983 0.455 0.385 0.929 0.303Results on society 0.985 0.415 0.494 0.920 0.273Financial results 0.991 0.410 0.493 0.916 0.261Results on external resources 0.991 0.459 0.382 0.844 0.300Process results 0.991 0.495 0.383 0.944 0.348

a Crombach’s alpha minus AVISC.




5.1.5. Descriptive statistics of the validated measurement scales

Table 6 shows the mean, standard deviation andcorrelations of the 13 unidimensional measurement scalesidentied in the scale validation process. All correlationswere positive and signicantly different from zero, a result

that is not surprising as the constructs were part of anintegrated approach to TQM ( Flynn and Saladin, 2001 ).Giving that the scales were developed using relevant itemsselected from a common survey, we conducted a Harman’ssingle-factor test ( Podsakoff and Organ, 1986; Podsakoff et al., 2003 ) to assess whether common method varianceexists and to deal with the potential social desirability of the responses. The results of the CFA with the 13 indicatorsloading into a single-factor ( x 2

ð65 Þ ¼ 459 : 09; CFI = 0.861;RMSEA = 0.119; BBNFI = 0.842; x 2 /d.f. = 7.06) showed apoor t, suggesting that the single-factor does not accountfor all of the variance in the data.

Finally, in order to reduce the number of parameters to

be estimatedandto reducethe complexity of thestructuralmodel, the 13 constructs were introduced into thestructural model as latent factors measured by a singleindicator resulting from computing the mean of all items(or composite measures in the case of the enabler domain)for each construct. Reliability coefcients of the seven-unidimensional criteria were introduced in the structuralmodel to correct for attenuation. Fig. 3 shows themultidimensional structural model (MDSM) resultingfrom the scale validation process.

5.2. Results for the multidimensional structural model

The 13 scales obtained from the scale validationprocess were used to estimate the multidimensionalstructural model. In this section, we present the good-ness-of-t indices of the model, followed by a descrip-

tion of the estimated parameters. In the next section, wecompare the model with alternative models in whichTQM dimensions, enabler excellence and results excel-lence respectively, are not considered. Finally, we applythe model to the sub-samples of services and manu-facturing sectors and compare the results obtained in

both samples to test whether there are differences in theEFQM Excellence Model between service and manufac-turing rms.

5.2.1. Model t First row of Table 7 shows the goodness-of-t indices of

the multidimensional structural model posed in Fig. 3.Several statistics were used to evaluate the goodness-of-tof the model ( Browne and Cudeck, 1993 ), and the values of all of them show the adequacy of the MDSM for the sampledata. These results suggest that the internal structure of the EFQM Excellence Model proposed by the MDSM is thussupported by our data set.

5.2.2. Item-factor loadingsThe estimated parameters, standard errors and t -tests

are shown in Table 8 . The rst part of Table 8 shows theloadings ( l coefcients) of the enabler and resultexcellence factors, respectively. In the enabler domain,all loadings are high and statistically signicant, rangingfrom 0.916 for ‘‘Leadership’’, to 0.825 for the ‘‘Process’’criterion. Social and technical dimensions also show highloadings (0.987 and 0.932, respectively) indicating that alldimensions and criteria represent the enabler excellenceconstruct to a large degree.

To assess the relationships between the enabler

excellence construct and its dimensions, Table 8 showsthe coefcient of determination ( R2 ) for every enablercriterion and TQM dimension. The high values indicatethat the enabler excellence construct explains a high

Fig. 3. The multidimensional structural model (MDSM) nally analyzed.

Table 7Goodness-of-t indices for the structural models

Model S-B x 2 a d.f. p-Value CFI RMSEA BBNFI BBNNFI S-B x 2 a

Multidimensional structural model (MDSM) 259.240 60 0.000 0.930 0.088 0.911 0.909 4.321Unidimensional structural model (UDSM) 277.907 62 0.000 0.924 0.090 0.904 0.904 4.482Multivariate structural model (MVSM) for enablers 213.359 53 0.000 0.943 0.084 0.927 0.919 4.026Multivariate structural model (MVSM) for results 220.941 55 0.000 0.941 0.084 0.924 0.917 4.017MDSM for multiple-group analysis (services and manufacturers) 329.334 122 0.000 0.928 0.063 0.891 0.907 2.700

a Satorra–Bentler scaled chi-square ( Satorra and Bentler, 1994 ).




degree of variance in every case. To assess the overallrelationship between the enabler excellence constructand the TQM dimensions, we also compute the total(multivariate) coefcient of determination ( R 2

m ), whichrepresents the amount of variance in the set of dependentvariables (TQM dimensions) explained by the indepen-dent variable (enabler excellence). An R 2

m value of 0.979indicates that the enabler excellence construct explainsalmost 98% of the variation of the social and technicaldimensions.

The same procedure was applied to evaluate the

adequacy of the result excellence construct. Table 8 showsthat all loadings are statistically signicant, indicating thatcriteria embody the result excellence construct in highdegree. The coefcients of determination for the resultcriteria ( Table 8 ) show that result excellence constructexplains a high percentage of variance in all result criteriaexcept for the ‘‘Society Result’’ (37%). By computing thetotal coefcient of determination ( R 2

m ¼ 0 : 710) we see thatthe result excellence explains more than 70% of the varianceof the result criteria.

5.2.3. Enabler–results structural relationshipTable 8 shows that the enabler excellence construct has

a statistically signicant effect on the result excellence .The high value of the regression parameter (0.841)implies that there is a strong causal relationship betweenthe two domains in the EFQM Excellence Model. Thecoefcient of determination is also high ( R2 = 0.707),reecting that the enabler domain explains 70% of thevariance of the result domain. These results support thepositive inuence of enabler excellence on result excel-lence , as the EFQM Excellence Model proposes. In Table 8we also show the indirect effect of enabler excellence oneach of the result criteria. The effect of the enablerexcellence is high and statistically signicant for allcriteria, although it varies between them, being higher

for ‘‘Key performance results’’ (0.797) and lower for‘‘Society results’’ (0.511).

5.3. Comparison with alternative models

To provide additional evidence of the suitability of theMDSM, it was compared with three alternative models(Shook et al., 2004 ). Firstly, the MDSM was compared witha model in which enabler excellence is modeled as aunidimensional construct, thus excluding the social andtechnical dimensions from the model. The comparisonbetween the MDSM and the unidimensional structuralmodel (UDSM) tests whether a model that distinguishbetween the social and technical TQM dimensions ts the

data better than a model in which these two dimensions donot exist. As UDSM is nested in MDSM, a scaled chi-squaredifference test was carried out, rejecting the null hypoth-esis of unidimensionalty of the enabler domain(D x 2 = 16.813; D d.f. = 2; p < 0.05) at 5% level.

Secondly, we also compare the MDSM with a Multi-variate Structural Model, MVSM ( Edwards, 2001 ) for theenabler constructs. In this model, all ve enabler criteriawere treated as a set of exogenous variables with directeffects on the result excellence construct, and the enablerexcellence construct and TQM dimensions were absent.Correlations between enabler criteria (the exogenousvariables) were also introduced into the model to account

for the relationship between the enabler criteria. Thecomparison between MDSM and MVSM for enablersallows us to test whether the model with the multi-dimensional construct enabler excellence ts the databetter and explains more variance of result excellence thanthe model in which enabler criteria directly affect theenabler result excellence construct. Results for the MVSMfor the enabler domain are reported in Table 7 . Althoughboth models have a good t, the MVSM seems to t thedata better than the multidimensional model, as thegoodness-of-t indices values show. This result is notunexpected, because multivariate models tend to have abetter t than multidimensional models. Multidimen-

sional models comprise dimensions that are necessarilydifferent from one another, and these distinctions are

Table 8Parameter estimates, standard errors and t -test for the MDSM

Effect Parameter estimates S.E. t -Test R2

Item-factor loadingsEnabler Exc. ! Social dimension 0.987 0.042 22.896 0.974Enabler Exc. ! Technical dimension 0.932 0.058 20.772 0.869Enabler Exc. ! Policy and strategy 0.862 – – 0.742Social dimension ! Leadership 0.916 – – 0.839

Social dimension ! People 0.878 0.054 27.602 0.770Technical dimension ! Partnerships and resources 0.877 – – 0.769Technical dimension ! Process 0.825 0.047 20.025 0.680Result Exc. ! Customer results 0.771 – – 0.595Result Exc. ! People results 0.831 0.097 10.635 0.691Result Exc. ! Society results 0.608 0.100 11.496 0.370Result Exc. ! Key performance results 0.948 0.090 11.240 0.898

Direct causal effectsEnabler Exc. ! Result excellence 0.841 0.066 13.247 0.707

Indirect causal effectsEnabler Exc. ! Customer results 0.648 0.066 13.247Enabler Exc. ! People results 0.699 0.130 6.928Enabler Exc. ! Society results 0.511 0.136 7.395Enabler Exc. ! Key performance results 0.797 0.120 7.339






TQM. To be a TQM framework, the internal structure of theEFQM Excellence Model should represent: the social andtechnical TQM dimensions; the interrelationships betweenthem; and their inuence on results.

Following previous studies ( Yong and Wilkinson,2001; Cua et al., 2001; Rahman, 2004 ) we have distin-guished between social (soft) and technical (hard)dimensions, represented in the enabler side of the EFQM

Excellence by the Leadership and People criteria for thesocial dimension, and by Resources and Partnership andProcesses criteria for the technical dimension. In studyingthe dimensionality of the EFQM Excellence Model it wasfound that, rstly, the variances of social and technicaldimensions are statistically signicant, indicating thatboth dimensions are present in the EFQM model. Inaddition, the comparison between the MDSM with theUDSM shows that social and technical dimensions arenecessary in order to adequately reproduce the inter-relationships within the enabler domain of the model.This nding conrms that social and technical elementsare embedded in the EFQM Excellence Model, providing

support to its consideration as a TQM framework. Theseresults converge with those obtained by Curkovic et al.(2000) for the MBNQA when they conclude that MBNQAcaptures the major dimensions of TQM and that thisquality award model is an appropriate framework for arm to base quality improvement efforts.

Secondly, to reect the holistic approach that TQMrepresents, social and technical dimensions should beintercorrelated. In our model, this interrelationship iscaptured by the enabler excellence construct. The highvalues of the loading of the enabler excellence on socialand technical dimensions (0.987 and 0.932, respectively)indicate that, for the sample of Spanish rms, TQM

dimensions are highly interrelated. Moreover, the multi-dimensional coefcientof determination indicates that the

enabler excellence construct explains to a high degree(almost 98%) the variation of the social and technicaldimension of TQM. These results conrm previous studiesin TQM literature such as Wilkinson et al. (1991) or Snapeet al. (1995) , when they consider that technical elementsrelated to production issues (such as systematic measure-ment of processes or performance standards) should beaccompanied by an emphasis on human resource activities

which shape employees’ attitudes relating to quality.Likewise, the interrelationship between the social andtechnical dimension supports Dale (1997) assertion thatthe integration between the social and technical issues of TQM characterizes organizations committed to TQM, andCua et al. (2001) empirical ndings about the comple-mentarities between the TQM technical-oriented practicesand the human- and strategic-oriented practices.

Thirdly, the holistic approach of TQM is also shown inthe results side of the EFQM Excellence Model, as theexcellence results construct explains more than 70% of thevariation in results. This nding conrms the resultsobtained by Reiner (2002) and Calvo-Mora et al. (2005)

regarding the existence of an internal logic between theresult elements.Finally, we also nd strong evidence of the causal

relationship between the enabler and result criteria,according to the TQM assumptions. The high value of the regression coefcient (0.841) and the high proportionof variance of the result excellence explained by theenabler excellence (70%) indicate that enabler criteriamust be implemented together to have an effectiveinuence on the result criteria. This assumption coincideswith a number of studies on TQM ( Belohlav, 1993; Flynnet al., 1994; Nabitz and Klazinga, 1999; Train and Williams,2000 ), and reinforces the importance of adopting a holistic

view in the EFQM Excellence Model by taking into accountall the elements of the model.

Table 9Estimated parameters, standard errors and t -test for the MDSM for multiple-group model

Effect Multidimensional structural model forservice companies

Multidimensional structural model formanufacturing companies

Parameter estimates S.E. t -Test Parameter estimates S.E. t -Test

Item-factor loadingsEnabler Exc. ! Social dimension 0.986 0.043 9.961 0.970 0.076 13.977

Enabler Exc. ! Technical dimension 0.952 0.091 12.890 0.942 0.090 14.846Enabler Exc. ! Policy and strategy 0.890 – – 0.825 – –Social dimension ! Leadership 0.911 – – 0.917 – –Social dimension ! People 0.860 0.094 14.900 0.887 0.065 23.137Technical dimension ! Partnerships and resources 0.868 – – 0.864 – –Technical dimension ! Process 0.847 0.064 14.697 0.835 0.067 14.922Result Exc. ! Customer results 0.751 – – 0.787 – –Result Exc. ! People results 0.847 0.159 7.857 0.807 0.120 7.484Result Exc. ! Society results 0.558 0.196 6.353 0.679 0.100 11.492Result Exc. ! Key performance results 0.911 0.146 7.427 0.990 0.986 9.300

Direct causal effectsEnabler Exc. ! Result excellence 0.858 0.079 9.742 0.816 0.110 8.953

Indirect causal effectsEnabler Exc. ! Customer results 0.644 0.079 9.742 0.642 0.110 8.953Enabler Exc. ! People results 0.726 0.192 5.017 0.659 0.191 5.962

Enabler Exc. ! Society results 0.479 0.210 4.556 0.554 0.178 4.977Enabler Exc. ! Key performance results 0.781 0.179 5.063 0.808 0.177 5.496




Our results extend the application of quality awardmodelsin several domains. Firstly, this study complementsprevious research analyzing quality award models, whichhave focused on both developing measurement models(and its associated constructs and scales) that accuratelycapture the content of quality award model criteria, andproviding insights into the directions of causation among

the award’s categories (e.g. Meyer and Collier, 2001; Leeet al., 2003; Badri et al., 2006 ). As TQM models, the causalrelationships between awards criteria should be inter-pretednot only as empirical regularities within thespecicmodel, but also as substantive relationships in the broadercontext of TQM research.

Secondly, our results give support to certain authors(e.g. Eskildsen, 1998; Van der Wiele et al., 2000; Westlund,2001 ) that have pointed out that the EFQM ExcellenceModel constitutes an appropriate framework to guide thesystematic implementation of TQM. Awards-based frame-works are seen by some authors only as a tool fororganizations seeking to be recognized as leaders in the

quality management eld or as a self-assessment tool.However, our results support the suggestion by Ghobadianand Woo (1996) that they can be used as guidelines forTQM implementation.

Finally, our research matches up with other empiricallybased TQM research as Flynn et al. (1994) , Powell (1995) ,Black and Porter (1995) , Ahire et al. (1996) or Dow et al.(1999) that used quality prescriptions in TQM literature todevelop measurement scales to analyze TQM implementa-tion through a set of critical factors. In this sense, the EFQMExcellence Model, as well as other quality award modelssuch MBMQA, provide detailed information through thedenition of the criteria, sub-criteria and guidance points

that can be useful in the measurement of TQM imple-mentation and in the comparison across countries basedon the international acceptance of the quality awardmodels.

6.2. Comparison with the MBNQA Model

Our results rely on the EFQM Excellence Model;however, it would be of interest to compare them withthose obtained in previous studies in the MBNQA. Inparticular, we are interested in comparing the extent towhich both models share the main assumptions involvedin the TQM concept, such as the following: the distinction

between technical and social dimensions, the holisticinterpretation of TQM, and the causal linkage betweenTQM procedures and performance.

Firstly, the holistic interpretation of the TQM issupported by Pannirselvam and Ferguson (2001) . Usingthe MBNQA categories, they found that the MBNQAframework reects the relationships between thevarious elements needed by managers to improveorganizational performance and proposed that organiza-tions need to broaden their focus from quality controltechniques and product design processes to a multi-faceted view of TQM. Similarly, for a sample of Koreanmanufacturing rms, Lee et al. (2003) found that

organizational success depends on adopting both pri-mary (i.e., strategic planning and customer and market

focus) and supportive (human resources and processmanagement) quality programs. Moreover, they foundthat enhanced links between leadership and qualitysystems and between quality information and analysisand quality systems are required. The capability of MBNQA to capture the holistic nature of TQM is alsoexplicitly recognized by Curkovic et al. (2000) , proposing

that TQM can be conceptualized as a latent factor thatcaptures covariation among the four basic factors thatdescribe the MBNQA framework.

Secondly, although previous studies using the MBNQAhave proposed alternative dimensions to capture theconstituent parts of TQM, we nd certain similarities withthe proposal of the existence of a ‘‘social’’ and a ‘‘technical’’dimension. For example, Pannirselvam and Ferguson(2001) found a strong relationship between leadership,human resource management, and customer focus, resultsthat are consistent with our results for the existence of a‘‘social’’ dimension. In the same vein, Terziovski andSamson (1999) , using the seven categories of the MBNQA

to explain operational performance, found a positiverelationship between leadership, human resources andcustomer focus and performance and a non-signicant ornegative effect of the other categories. Based on theseresults, they conclude that: ‘‘It is interesting that thestrongpredictors of performance were the so-called ‘soft’ factorsof leadership, human resource management and customerfocus’’ (p. 403).

Finally, the positive inuence of the enabler on resultscriteria is also supported by most of the studies that haveanalyzed the MBNQA (e.g. Wilson and Collier, 2000; Flynnand Saladin, 2001; Pannirselvam and Ferguson, 2001;Ghosh et al., 2003; Lee et al., 2003 ). All these authors report

the existence of signicant relationships among theBaldrige categories and emphasize the adequacy of theMBNQA to conrm the relationship between TQMpractices and performance.

6.3. A global approach to studying the internal structure of the EFQM Excellence Model

From a methodological point of view, we have adopteda global approach to analyzing the internal structure of theEFQM Excellence Model that combines the factorial andcausal approaches adopted in previous studies. In ourmodel, the interrelationships between the criteria in each

domain – enabler and result – are explained by the latentfactors enabler and result excellence , which measure thedegree of excellence reached by an organization in themanagement of theenabler andresult criteria respectively.The causal effect of enablers on results is summarized in asingle coefcient that relates enabler excellence and resultexcellence. Our approach is thus in accordance withDijkstra (1997) who asserts that there is a latent factorthat underlies the model’s criteria, as well as with otherstudies (e.g. Winn and Cameron, 1998; Wilson and Collier,2000; Calvo-Mora et al., 2005 ) which have found theexistence of causal relationships between awards criteria.We therefore complement the causal and factorial

approaches used in previous studies to analyze the internalstructure of the EFQM Excellence Model. Within this




approach, the assimilation and the commitment of a rmto the TQM core concepts will determine the behavior of the organization (i.e., the degree of excellence reached bytheorganization), that will be reected in thecriteria of theEFQM Excellence Model.

Our global approach provides some other interestingbenets for testing the EFQM Excellence Model. The

latent enabler and result excellence constructs helps toexplain the way in which TQM is implemented inorganizations, providing information about the degreeof deployment of each enabler criterion. For example, forthe sample of Spanish rms, our results indicate that theenabler excellence construct is reected to a greaterdegree by the criteria related to social dimension than forthe technical dimension. In the result domain, Societyresults is the criterion which shows a lesser relationshipwith the result excellence construct. This means that,although we advocate the analysis of enablers as a wholesystem and the consideration of the commonalitybetween all the dimensions in producing results, there

are differences between enablers and result criteria inthe degree in which enabler and result excellence ismanifested.

Finally, although previous studies have also adopted afactorial approach to analyzing the award-based frame-works ( Curkovic et al., 2000; Dijkstra, 1997; Bou-Llusaret al., 2005 ), our methodological approach extends thesestudies by considering both a factorial and causalapproach. In this sense, our study extends Dijkstra’sapproach, analyzing not only the enabler side of theEFQM Excellence Model but also the results domain.Curkovic et al. (2000) create a common factor namedTQM that capture the relationship between four con-

structs of the MBNQA (TQM strategy system, TQMoperations systems, TQM informational systems, TQMresults). While they already include results in their TQMfactor, we extend their approach by analyzing the causaleffect of enablers on results in the EFQM ExcellenceModel. In addition, our paper enhances the study by Bou-Llusar et al. (2005) by considering both technical andsocial TQM dimensions in the interpretation of theenabler latent factor.

6.4. Implications

Our results are also interesting for managers who are

applying the EFQM Excellence Model to improve quality intheir businesses and to guide the implementation of TQM,and for researchers interested in the study of theoperational frameworks for TQM. As Sila and Ebrahimpour(2002) state, in the absence of a universally accepted TQMmodel, many rms put their faith in quality award modelsand use them as guidelines. The understanding of theinternal structure of the EFQM Excellence Model in ourstudy facilitates the guiding role that award-based modelsplay in the implementation of TQM systems. This role willbe promoted with a greater comprehension of the linkagesbetween the elements that compose the models. Thus, asEskildsen and Dahlgaard (2000) and Reiner (2002) state,

knowledge about the structure of the model allowsorganizations to benet fully from the self-assessment

processes and to improve themanagement of thebusiness.In addition, according to Rusjan (2005) , the analysis of therelationships between the model criteria complements theguidelines included in the model in order to exploit thestrengths and to prioritize areas for improvement inorganizations.

Along these lines, managers can use our conclusions

to diagnose their TQM status and develop actions plans.The empirical validation of our model suggests practi-tioners should take on a systemic approach in the use of the EFQM Excellence Model to implement a TQMinitiative. Managers must balance the social and tech-nical practices in the context of a TQM initiative asefforts concentrated on one or a few issues would be lesseffective. As Curkovic et al. (2000) suggest, social andtechnical dimensions are needed to have a completeTQM system, and managers should focus on these bothcritical drivers of performance.

In addition, the systemic approach implies thatmanagers need to be committed to TQM core concepts

in order to take advantage of the practices considered inthe EFQM criteria. A general commitment to and accep-tance of TQM principles is essential in order to benet fromthe application of EFQM criteria. Otherwise managerswould have only a collection of unrelated and incoherentpractices.

Finally, our results suggest some opportunities fordeveloping the EFQM Excellence Model, for example bygrouping the social and technical criteria. Theapproach wehave used to operationalize EFQM criteria could also beapplied to other constructs in operations managementwhich, like TQM, could be considered as latent andmultidimensional in the existing literature.

6.5. Limitations and future research

As in other empirical studies, the ndings andimplications in this study should be interpreted withcaution, due to their limitations. Firstly, in our study themodel was developed and tested using the same data set.Although the model is suitable for both service andmanufacturing rms, and the results obtained remainstable across sub-samples, the application of the model toother data will help to validate our results and to assess itsgenerality in other contexts.

Secondly, we use perceptual data to measure the

EFQM criteria and it is worth recognizing the possibilitythat the perceptions of those surveyed do not provide acompletely accurate view of reality. So, a logicalextension would be to use multiple informants to verifyperceptions. Also, the study has focused on a genericSEM, while there may be cross-country differences inthis respect ( Prasad and Tata, 2003; Rungtusanathamet al., 2005; Flynn and Saladin, 2006 ). For example, Flynnand Saladin (2006) have reported evidence about thestrong role that national culture plays in the effective-ness of the MBNQA, and the need for countries to adaptquality initiatives to their national cultures. A naturalextension of this paper would be to analyze the existence

of differences between countries in the EFQM ExcellenceModel.




Appendix A. Original scales for measuring the EFQM criteria

1. Leadership1a. Leaders develop the mission, vision, values and ethics and are role models for a culture of Excellence

Managers encourage employee empowerment and autonomyManagers participate and give support to continuous improvement processesManagers collaborate in quality training by teaching people at lower hierarchical levelsManagers ensure that all members of the company have a clear idea of what the company’s position in the market should be

1b. Leaders are personally involved in ensuring the organization’s management system is developed, implemented and continuously improvedManagers become involved in running the company as a set of interrelated processes, all of them responsible for qualityManagers ensure that employees are capable of taking initiatives and assimilating better ways of doing their jobs

1c. Leaders interact with customers, partners and representatives of societyManagers take part in continuous improvement processes, even when these activities go beyond managerial responsibilitiesSatisfaction of current and future customers ensures the competitive success of the companyTo improve in a particular aspect, we collaborate with other companies to help us with the improvement

1d. Leaders reinforce a culture of excellence with the organization’s peopleThere is a strong communicative culture throughout all areas of the organizationThe involvement of workers can only be achieved if managers are the rst to show commitment, practicing what they preachManagers behave in a way that allows the integration and mobilization of members of a team

1e. Leaders identify and champion organizational changeContinuous improvement and change are necessary even when good results are being obtainedManagers stimulate the continuous improvement of products and processesManagers continuously acquire and update knowledge that is valuable for the organizationManagers act in a way that makes it easier for employees to accept proposed changes voluntarily

2. Policy and strategy2a. Policy and strategy are based on the present and future needs and expectations of stakeholders

The establishment of organizational objectives takes into account employee opinionsThe establishment of organizational objectives takes into account external opinionsEffective management is based on information about customersCustomers’ needs are taken into account when establishing objectives

2b. Policy and strategy are based on information from performance measurement, research, learning and external related activitiesContinuous improvement processes are based on a systematic assessment of organizational effectivenessBenchmarking techniques are used to establish improvement standards and objectives*Systematic measurement of quality and non-quality costs is carried outSelf-assessment processes take place on a regular basis*Information systems are in place to capture external information (about customers and markets)

2c. Policy and Strategy are developed, reviewed and updatedSystematic procedures are in place to plan, evaluate and control organizational goal achievementsQuality strategies affect all organizational areas and managerial activitiesQuality objectives stem from long-term strategic plansThe organization has formal strategic plans*Managers favor consensus about relevant objectives and future projects

2d. Policy and Strategy are communicated and deployed through a framework of key processesOrganizational processes and their interrelationships are identiedQuality policies are translated into a set of specic and measurable objectives*Managers inform employees about the quality strategyEvery member in the organization knows the organizational mission and objectives

3. People3a. People resources are planned, managed and improved

Formal processes are used (such as attitude surveys or employee brieng) to nd out employee opinionsEmphasis is placed on recruiting highly skilled employees

3b. People’s knowledge and competences are identied, developed and sustainedSpecic quality training is offered to employeesEmployees continuously update their skills in their specic area of knowledgeExtensive training means are provided for employees

3c. People are involved and empoweredEmployees are allowed to decide how the work is doneEmployee opinions are taken into account when dening organizational objectivesEmployees are given the opportunity to suggest and implement solutions to work problemsEmployee autonomy and participation is encouragedTeamwork is common practice

3d. People and the organization have a dialogueFormal communication channels are in place to provide organizational areas with information about customers’ needs

Formal communication procedures are established with staff, customers and suppliers




Appendix A ( Continued )

Employees have access to information about quality resultsEmployees maintain uid communication with one another, going beyond the formal structure of the organizationEmployees have worked together for a long time, which facilitates good co-ordination between them*Internal communication is totally open and transparentEmployees voluntarily pass on useful information between one another

3e. People are rewarded, recognized and cared forManagers explicitly recognize employees’ achievements at work

4. Partnerships and resources4a. External partnerships are managed

Quality agreements with suppliers are establishedCooperation with suppliers provides the organization with high quality raw materials and resourcesRelationships with customers and suppliers allow the organization to have rapid access to information about new products and technologyThe organization has a high capacity for external cooperation

4b, 4c, 4d, 4e. Internal resources4b. Finances are managed4c. Buildings, equipment and materials are managed4d. Technology is managed4e. Information and knowledge are managed

Policy and strategy guides the denition of operative and nancial objectivesInventory levels are highIntensive efforts are made to guarantee high quality raw materialsFrequent technological innovations are implementedEfforts are made to know what the workforce needs in terms of information and resources*

5. Processes5a. Processes are systematically designed and managed

Work methods and organizational process are explicitly denedThere is comprehensive documentation about work methods and organizational processesQuality manuals and organizational processes are periodically revisedSystems of indicators are in place to revise changes in processes*Work processes exist that promote efcient behavior patterns throughout the organization

5b. Processes are improved, as needed, using innovation in order to fully satisfy and generate increasing value for customers and other stakeholdersDevelopment and innovation of production processes is emphasized

5c, 5d, 5e. Products and Services are designed, developed, produced and delivered based on customers’ needs5c. Products and Services are designed and developed based on customer needs and expectations

5d. Products and Services are produced, delivered and serviced5e. Customer relationships are managed and enhanced

The organization knows which products and services customers needThe organization is oriented towards the fulllment of customers’ expectations and needsProduct design provides customers with high utilityThe organization is able to develop new products or services ahead of competitors*The products’ valuable features are superior to those of competitorsStandardized systems are in place to deal with customer complaintsMarketing techniques and methods are developed*

6. Customer resultsCustomer satisfaction has improvedCustomer consolidation has improvedCommunication with customers has improvedCustomer complaints have decreased

Services offered to customers are better than competitors*

7. People resultsV7a. Employee motivation and commitment

Employee willingness to work extra time has improvedHigh employee organizational commitment has improved

V7b. Employee achievementEmployees identify and provide solutions to work problemsEmployees share organizational valuesEmployees show high levels of initiative

V7c. Employee satisfactionEmployee absenteeism has decreasedEmployee turnover has decreasedEmployee opinions contribute to improving work performance*Employees have high levels of know-how*Communication with employees has improved*




References

Ahire, S.L.,Golhar,D.Y., Waller, M.A.,1996. Development and validationof TQM implementation constructs. Decision Sciences 27 (1),23–56.

Ahire, S.L., Ravichandran, T., 2001. An innovation diffusion model of TQMimplementation. IEEE Transactions on Engineering Management 48(4), 445–464.

Anderson, J.C., Rungtusanatham, M., Schroeder, R.G., 1994. A theory of quality management underlying the deming management method.Academy of Management Review 19 (3), 472–509.

Anderson, J.C., Rungtusanatham, M., Schroeder, R.G., Devaraj, S., 1995. Apath analytic model of a theory of quality management underlyingthe deming management method: preliminary empirical ndings.Decision Sciences 26 (5), 637–658.

Askey, J.M., Dale, B.G., 1994. From ISO series registration to total qualitymanagement: an examination. Quality Management Journal July, 67–76.

Badri, M.A., Selim, H., Alshare, K., Grandon, E.E., Younis, H., Adsulla, M.,2006. The Baldrige education criteria for performance excellenceframework. Empirical test and validation. International Journal of Quality and Reliability Management 23 (9), 1118–1157.

Bagozzi, R.P., Phillips, L.W., 1982. Representing and testing organizationaltheories: a holistic construal. Administrative Science Quarterly 27,459–489.

Bagozzi, R.P., Edwards, J.R., 1998. A general approach for representingconstructs in organizational research. Organizational Research Meth-ods 1, 45–87.

Belohlav, J.A., 1993. Quality, strategy, and competitiveness. CaliforniaManagement Review 35 (3), 55–67.

Bandalos, D.L., Finney, S.J., 2001. Item parceling issues in structuralequation modeling. In: Marcoulides, G.A., Schumacker, R.E. (Eds.),Advanced Structural Equation Modeling: New Developments andTechniques. Lawrence Erlbaum Associates, Mahwah, NJ.

Bentler, P.M., 1995. EQS Structural Equations Program Manual. Multi-variate Software Inc., Encino, CA.

Bentler, P.M., Bonett, D.G., 1980. Signicance tests and goodness of t inthe analysis of covariance structures. Psychological Bulletin 88, 588–606.

Black, J.A., Porter, L.J., 1995. An empirical model for total quality manage-ment. Total Quality Management 6 (2), 149–164.

Black, S.A.,Crumley, H.C.,1997. Self-assessment: what’s in it for us? TotalQuality Management 8 (2/3), S90–S94.

Bohoris, G.A., 1995. A comparative assessment of some major qualityawards. International Journal of Quality & Reliability Management 12(9), 30–43.

Bou-Llusar, J.C., Escrig, A.B., Roca, V., Beltra ń, I., 2005. To what extent doenablers explain results in the EFQM Excellence Model? An empiricalstudy. International Journal of Quality and Reliability Management22 (4), 337–353.

Brown, A., 2002. Using HR strategies to support business excellence. In:Proceedings of the 7th World Congressfor TotalQualityManagement,vol. 2, Verona, Italy, pp. 339–346.

Browne, M.W., Cudeck, R., 1993. Alternative ways of assessing model t.In: Bollen, K., Long, J.S. (Eds.), Testing Structural Equation Models.Sage Publications, Newbury Park, CA, pp. 136–162.

Calvo-Mora, A., Leal, A., Roldań, J.L., 2005. Relationships between theEFQM Model Criteria: a study in Spanish Universities. Total QualityManagement 16 (6), 741–770.

Castresana, J.I., Ferna ńdez-Ortiz, R., 2005. Theoretical foundation of theEFQM Model: the resource-basedview. Total Quality Management 16(1), 31–55.

Cronbach, L.J., 1951. Coefcient alpha and the internal structure of tests.Psychometrika 16, 297–334.

Crosby, P.B., 1980. Quality is Free. The Art of Making Quality Certain.McGraw-Hill, New York.

Cua, K.O., McKone, K.E., Schroeder, R.G., 2001. Relationships betweenimplementation of TQM, JIT, and TPM and manufacturing perfor-mance. Journal of Operations Management 19, 675–694.

Curkovic, S., Melnyk, S., Calantone, R., 2000. Validating the MalcolmBaldrigeNational Quality Award framework through structural equa-tion modelling. International Journal of Production Research 38 (4),765–791.

Dale, B.G., 1997. Characteristics of organizations not committed to totalquality management. Journal of Engineering Manufacture 21, 377–384.

Dale, B.G., 2003. Managing Quality, 4th ed. Blackwell Publishers, Oxford.De Vasconcellos, J.A., Hambrick, D.C., 1989. Key success factors: test of a

general theory in the mature industrial-product sector. StrategicManagement Journal 10 (4), 367–383.

Dean, J.W., Bowen, D.E., 1994. Management theory and total quality:

improving research and practice through theory development. Acad-emy of Management Review 19, 392–418.

Appendix A ( Continued )

Employee satisfaction has improved*Employee involvement at work has improved*

8. Society resultsProtection of environment has improvedNoise levels have decreasedPollution levels have decreasedThe organization has a positive impact in society*

9. Key performance resultsV9a. Financial results

Market share has improvedSales per employee have improvedProt levels have improvedThere has been a noticeable improvement in nancial results*

V9b. External resultsThe number of suppliers has decreasedQuality of raw materials has improvedRelationships with suppliers have improvedSupplier management has improved

V9c. Results on processesProcess efciency has improved

Knowledge about efcient operation management has improvedRecorded time has improved*

Note : The scale for measuring ‘‘Partnership and Resources’’ criteria was composed of two sub-criteria: ‘‘External partnerships’’ (sub-criterion 4a), and acombined sub-criterion (4b, 4c, 4d and 4e) called ‘‘Internal resources’’. This two-dimensional proposal is in accordance with the ‘‘Partnership andResources’’ EFQM criterion denition: ‘‘Excellent organizations plan and manage External Partnerships , suppliers and Internal Resources in order to supportpolicy and strategy and the effective operation of processes’’ (italics added).The ‘‘Process’’ criterion was formed by three sub-criteria: 5a, 5b, and the combined sub-criterion 5c, 5d, 5e called ‘‘ Products and Services are designed ,developed , produced and serviced based on customer needs’’ .






Reiner, G., 2002. Analysis of critical factors of company success based onthe EFQM Excellence model. In: Proceedings of the 7th WorldCongress for Total Quality Management, vol. 2, Verona, Italy, pp.361–366.

Rungtusanatham, M.,Forza, C.,Koka,B.R.,Salvador, F., Nie, W.,2005. TQMacross multiple countries: Convergence Hypothesis versus NationalSpecicity arguments. Journal of Operations Management 23, 43–63.

Rusjan, B., 2005. Usefulness of the EFQM Excellence model: theoreticalexplanation of some conceptual and methodological issues. TotalQuality Management 16 (6), 363–380.

Satorra, A., Bentler, P.M., 1994. Corrections to test statistics and standarderrors in covariance structure analysis. In: von Eye, A., Clogg, C.C.(Eds.), Latent Variables Analysis: Applications for DevelopmentalResearch. Sage, Thousand Oaks, CA, pp. 399–419.

Satorra,A., Bentler, P.M.,2001. A scaled difference chi-square test statisticfor moment structure analysis. Psychometrika 66, 507–514.

Saraph, J., Benson, P.G., Schroeder, R., 1989. An instrument for measuringthe criticalfactors of quality management. DecisionSciences20, 810–829.

Shook, C.L, Ketchen, D.J., Hult, G.T., Kacmar, K.M., 2004. An assessment of the use of structural equation modeling in strategic managementresearch. Strategic Management Journal 24 (4), 397–404.

Sila, I., Ebrahimpour, M., 2002. An investigation of the total qualitymanagement survey based research published between 1989 and2000. A literature review. International Journal of Quality & ReliabilityManagement 19 (7), 902–970.

Silvestro, R., 1999. Positioning services along the volume-variety diag-onal: the contingencies o service design, control and improvement.International Journal of Operations & Production Management 19 (4),399–421.

Sitkin, S.B.,Sutcliffe, K., Schroeder, R.G., 1994. Distinguishing control fromlearning in total quality management: a contingency perspective.Academy of Management Review 19 (3), 537–564.

Snape, E., Wilkinson, A., Marchington, M., Redman, T., 1995. Managinghuman resources for TQM: possibilities and pitfalls. Employee Rela-tions 17, 42–51.

Stainer, A., Stainer, L., 1995. Productivity, quality and ethics—a Europeanviewpoint. European Business Review 95 (6), 3–11.

Sun,H., 1999. Diffusion and contribution of total quality management: anempirical study in Norway. Total Quality Management 10 (6), 901–914.

Tamimi, N., 1998. A second-order factor analysis of critical TQM factors.International Journal of Quality Science 3 (1), 71–79.

Terziovski, M., Samson, D., 1999. The link between total quality manage-ment practice and organisational performance. International Journalof Quality & Reliability Management 16 (3), 226–237.

Train, L., Williams, C., 2000. Evolution of quality management: Britishbenet enquiry line. The International Journal of Public Sector Man-agement 13 (6), 526–539.

Tummala, V.M., Tang, C.L., 1996. Strategic Quality Management, MalcolmBaldrige and European Quality Awards and ISO 9000 Certication.International Journalof Quality& Reliability Management13 (4),8–38.

Van der Wiele, A., Williams, A.R.T., Dale, B.G., 2000. ISO 9000 seriesregistration to business excellence: the migratory path. BusinessProcess Management 6 (5), 417–427.

Vandenberg, R.J., Lance, C.E., 2000. A review and synthesis of the mea-surement invariance literature: suggestions, practices, and recom-mendations for organizational research. Organizational ResearchMethods 3 (1), 4–70.

Wainer, H., Kiely, C., 1987. Item clusters and computerized adaptivetesting: a case for testlets. Journal of Educational Measurement 24,185–202.

Westlund, A.H., 2001. Measuring environmental impact on society in theEFQM system. Total Quality Management 12 (1), 125–135.

Wilkinson, A., Allen, P., Snape, E., 1991. TQM and the management of labour. Employee Relations 13, 24–31.

Wilkinson, A., Redman, T., Snape, E., Marchington, M., 1998. Managingwith total quality management. Theory and practice. McMillan Busi-ness, Hong-Kong.

Wilson, D.D.,Collier, D.A.,2000. An empirical investigationof theMalcolmBaldrige National Quality award causal model. Decision Sciences 31,361–383.

Winn, B.A., Cameron, K.S., 1998. Organizational quality: an examinationof the Malcolm Baldrige National Quality Framework. Research inHigher Education 39 (5), 491–512.

Wruck, K.H., Jensen, M.C., 1994. Science, specic knowledge, and totalquality management. Journal of Accounting & Economics 18, 247–287.

Yong, J., Wilkinson, A., 2001. Rethinking total quality management. TotalQuality Management 12 (2), 247–258.

Yusof, S.M., Aspinwall, E., 2000. Total quality management implementa-tion frameworks: comparison and review. Total Quality Management11 (3), 281–294.

Zhang,Z., 2000. Developing a model of quality management methods andevaluating their effects on business performance. Total Quality Man-agement 11 (1), 129–137.


Documents

Bou-Llusar Et Al 2009 EFQM Excellence Model