The mediating role of primary TQM factors and strategy in the relationship between supportive TQM factors and organisational results: An empirical assessment using the MBNQA model

Abstract The mediating role among total quality management and business excellence model factors, including the role of strategy, has rarely been addressed empirically. This research examines the mediating effects of primary total quality management factors (operations management and measurement, analysis, and knowledge management) and strategy on the relationship between supportive total quality management factors (leadership, workforce, and customers) and organisational results using the Malcolm Baldrige National Quality Award (MBNQA) model as the framework. Based on responses from managers of 217 Saudi firms, the questions and hypotheses suggested in this research are examined using the structural equation modelling technique. The results show that: (a) supportive factors enhance primary factors; (b) strategy mediates the relationship between supportive factors and primary factors; (c) primary factors mediate the relationship between supportive factors and results; and, more importantly, (d) collectively and sequentially, the combination of strategy and primary factors mediates the relationship between supportive factors and results more strongly than their respective individual mediation effects separately. The findings support the systems perspective provided by the MBNQA model.

Abstract: The mediating role among total quality management and business excellence model factors, including the role of strategy, has rarely been addressed empirically. This research examines the mediating effects of primary total quality management factors (operations management and measurement, analysis, and knowledge management) and strategy on the relationship between supportive total quality management factors (leadership, workforce, and customers) and organisational results using the Malcolm Baldrige National Quality Award (MBNQA) model as the framework. Based on responses from managers of 217 Saudi firms, the questions and hypotheses suggested in this research are examined using the structural equation modelling technique. The results show that: (a) supportive factors enhance primary factors; (b) strategy mediates the relationship between supportive factors and primary factors; (c) primary factors mediate the relationship between supportive factors and results; and, more importantly, (d) collectively and sequentially, the combination of strategy and primary factors mediates the relationship between ABOUT THE AUTHOR Mohammed Alanazi is an Assistant professor of operations management and strategic management at the Department of Administrative Sciences at King Fahd Security College, KSA where he teaches courses at both the graduate and undergraduate levels. His current research focuses on business excellence models and TQM practices (including empirical research on their causal structure), organizational self assessment, and business process management. He holds a PhD in management from University of Hull, UK.

PUBLIC INTEREST STATEMENT
Business conditions usually lead managers to ask the following questions. Where should we start the organizational improvement practices (e.g., total quality management [TQM] practices)? What are the relationships of such practices? What are their relative significance and sequence? This study provides useful answers for such enquiries. This study analyses the way quality management systems work based on relationships and integration among practices (e.g., leadership, operations management, and an organisation's results) using the Malcolm Baldrige National Quality Award (MBNQA) model as the framework.
This study reveals that managers should view their business as a connected, holistic system. Moreover, it shows that there should be a clear progression starting from leadership's commitment and direction, which engages a qualified workforce that understands and effectively participates in managing operations, quality tools, and information through an integrated strategy that leads to the desired results. Understanding this can improve the implementation process and resources management.
supportive factors and results more strongly than their respective individual mediation effects separately. The findings support the systems perspective provided by the MBNQA model.

Introduction
In today's competitive environment, customers have become increasingly aware of the quality of services and products, forcing organisations to adopt initiatives for continual quality improvement. Previous studies have considered many of these initiatives and approaches such as Balanced Scorecard (BSC), Total Preventive Management, Six-Sigma, and Total Quality Management (TQM). Among these, this research focuses on TQM.
Although there are many definitions of TQM, a commonly accepted one states that it is a comprehensive management approach for an organisation to operate through guiding principles, basic concepts, continuous improvement, and organisational change, which, when effectively linked, will lead to desirable organisational results (Kaynak & Hartley, 2005;Oakland, 2014;Spencer, 1994).
To achieve the expected TQM results, its key elements must be developed, implemented, and improved to form a quality management system (QMS) (NIST, 2015). The TQM approach is based on the proper integration and linkage of certain principles with the organisation's strategy, structure, and procedures (Bou-Llusar et al., 2005).
Previous studies have shown that BEMs provide the best reference model for implementing and improving TQM (Kim et al., 2010;Prajogo & Sohal, 2006) and that organisations reap benefits when implementing BEMs (Porter & Tanner, 2004). For example, these models can be used for a comprehensive, systematic, and regular assessment of the organisation's practices and results in comparison to BEM criteria (EFQM, 2013;NIST, 2015). Since BEMs provide a systematic review of an organisation's strengths and weaknesses, they enable development and implementation of strategic plans. Ritchie and Dale (2000) noted that the full benefit of self-assessment is achieved by linking the assessment results to strategic planning, thus confirming the fundamental TQM assumption of quality management (Dow et al., 1999).
TQM factors and practices can be categorised into three types (Flynn et al., 1995;Oakland, 2011;Rahman & Bullock, 2005): (1) Social, soft, or supportive factors (referred to as supportive TQM factors in this research) that focus on behaviour such as leadership, workforce practices, and customer focus.
(2) Technical, hard, or primary factors (referred to as primary TQM factors in this research) that focus on production methods and processes. Examples include operations management and measurement, analysis, and knowledge management.
(3) The strategy factor, reflecting the need to integrate TQM objectives, plans, and policies into the organisation's strategic plan (Reiner, 2002).
La Rotta and Rave (2016) stated that there is scant research focused on the causal relations between BEM factors. On the other hand, Gadenne andSharma (2009), Zeng et al. (2015), Para-González et al. (2015), Sweis et al. (2016), Danyen and Callychurn (2015), and Ho et al. (2001) all indicated there are few clear results on which critical TQM factors have greater impacts on performance and which ones affect results directly, indirectly, or through mediation roles of other factors. Ignoring quality as a strategic issue is the main reason for failure in the implementation of TQMs and BEMs (Gómez-López et al., 2016;Taylor & Wright, 2003). Boulter et al. (2013) and Kim et al. (2010) stated that there is a significant lack of research on the role of the strategic factor in BEMs. Existing research has not analysed the mediating role among BEM factors (e.g., Badri et al., 2006;Prybutok et al., 2011), nor has it been addressed with a focus on the strategy factor or the MBNQA model (e.g., Calvo-Mora et al., 2015).
Taking this into account, the purpose of this study is to analyse the way BEMs work as a QMS based on relationships and integration among its factors. Specifically, this study seeks to further researchers' work aiming at improving our understanding of the interrelationship between TQM factors as a QMS and determining which of these factors is more significant in enhancing results, as well as to help organisations to identify areas of focus for their QMS. A focus will be laid on leadership, workforce, and customers factors (supportive TQM factors); primary TQM factors (as a combined construct comprising operations management and measurement, analysis, and knowledge management); strategy; and results in the context of the MBNQA model. The following questions illustrate the objectives of this research: (1) What is the mediating role of strategy in the relationship between leadership, workforce, and customers and primary TQM factors and results?
(2) What is the mediating role of primary TQM factors in the relationship between leadership, workforce, and customers and results?
In answering these questions, this study aims to fill the gaps in the literature as follows: (1) Considering the limited findings on the integration mechanism and relationship between TQM and BEMs factors, and more specifically the role of strategy and primary TQM factors, this study improves the understanding of how to effectively manage and integrate such factors into a QMS; (2) Unlike studies on the direct and indirect effects of supportive TQM factors as a combined construct (e.g., Zeng et al., 2015), this study independently examines each factor (i.e. leadership, workforce, and customers) to analyse their direct and indirect roles; (3) While past research has focused on the EFQM model (e.g., Calvo-Mora et al., 2015;Suarez et al., 2016), this study focuses on the MBNQA model to analyse the mediation effects among TQM factors and organisational results.
The remainder of this paper is organised as follows. Section 2 provides a literature review and the research hypotheses. Section 3 presents the research methodology and Section 4 outlines the results. Section 5 concludes the paper with a discussion of results, conclusions, and limitations.

Literature review and research hypotheses
The MBNQA model presents a framework for analysing the relationships between organisational practices and results (NIST, 2015). The criteria proposed by the model are leadership; workforce; customers (customers focus); strategy; operations management; and measurement, analysis, and knowledge management. Additionally, there is a results criterion that reflects customer satisfaction, financial results, products, market, etc. Overall, these criteria represent the seven critical aspects of organisational management and performance.
The MBNQA model implies that all components are managed as one integrated system (i.e. QMS). Examples of integration are the relationship between practices criteria and results and between strategic planning and leadership and workforce planning as well identifying customers and markets while developing strategy (NIST, 2015).
The model identifies the mechanism through which leadership can direct practices that determine performance (Prybutok et al., 2011;Wilson & Collier, 2000). In particular, it highlights the significant role of strategy in implementing TQM practices (NIST, 2015), as it plays a central role in harmonising and connecting such practices to achieve effective and consistent results (Kumar & Sharma, 2017). While the MBNQA model does not explicitly refer to the supportive and primary TQM factors, leadership, workforce, and customers criteria are consistent with the supportive TQM factors, and operations management and measurement, analysis and knowledge management correspond to primary TQM factors (Flynn et al., 1995;Ho et al., 2001).
Although studies have examined the relationship between TQM practices and organisational results, the findings have been inconsistent (Dubey & Gunasekaran, 2015;Psomas & Jaca, 2016;Rahman & Bullock, 2005;Sila, 2007). Studies have not provided clear findings on which factors have the most significant effect (primary, supportive, or strategy factors) on results, or which factors directly or indirectly affect results through other factors . Furthermore, such studies are limited in the BEMs domain (e.g., Calvo-Mora et al., 2015). Although the MBNQA model supports the QMS perspective on the interrelation and integration of the model's seven criteria, it does not clearly outline the relationship between process criteria or the relationship between such criteria and results (Curkovic et al., 2000;NIST, 2015). Therefore, a research model ( Figure 1) is proposed to identify the relationships between leadership, workforce, and customers factors and primary TQM factors as well as their relationships with organisational results in the MBNQA model.

The relationship between leadership, workforce, and customers and primary TQM factors
Organisations can achieve desired results with various methods such as benchmarking and defectprevention mechanisms. However, to successfully implement these primary TQM factors, they should focus on customers and workforce participation through empowerment and training. The essence of quality is a workforce that is sensitive to customers' needs and expectations (Ahire et al., 1996;Fotopoulos & Psomas, 2010). Therefore, an organisational leader should focus on customer-oriented strategies and targets. Leadership represents a guiding force for TQM factors through a flexible and creative direction connected with stakeholders (Rahman & Bullock, 2005). Supportive TQM factors (leadership, workforce, and customers) and primary TQM factors are not independent. Therefore, the following hypothesis is proposed: H1. Leadership, workforce, and customers (supportive TQM factors) are positively related to primary TQM factors.
2.2. The mediation effect of strategy in the relationship between leadership, workforce, and customers and primary TQM factors Kochan et al. (1995) indicated that quality should be considered as part of a broader strategy in organisational change. In this context, TQM should be considered as a key component of an organisation's strategy for survival and growth (Deming, 1982). In the MBNQA model, the strategy criterion focuses on the development of objectives and action plans; implementing those plans; modifying them when necessary; and measuring progress (NIST, 2015). Thus, TQM practices and targets should be integrated into the strategic plan in a systematic manner (Oakland, 2011).
Moreover, quality practices are considered an intangible strategic resource that cannot be imitated by competitors (Powell, 1995). Through quality implementation and improvement, a competitive advantage can be achieved (Zatzick et al., 2012). TQM can be used in different strategic situations such as a cost leadership strategy that is related to cost advantage (focus on operations) (Zatzick et al., 2012) or a differentiation strategy (focus on customers) (Prajogo & Sohal, 2006). Reiner (2002) illustrated that strategy, in the context of BEMs, sets a roadmap for reaching desired results by identifying stakeholders, market segments, targets, operations, and required resources. Evans and Lindsay pointed out that "A key role of strategic planning is to align work processes and learning initiatives with an organisation's strategic directions, thereby ensuring that improvement and learning prepare for and reinforce organisational priorities" (2014, 558). Therefore, the following hypothesis is proposed: H2. Strategy mediates the relationship between leadership, workforce, and customers and primary TQM factors.

The mediating effect of primary TQM factors
Academic literature shows the mediating role of primary TQM factors between leadership, workforce, and customers and results (Calvo-Mora et al., 2014;Fotopoulos & Psomas, 2010;Peng & Prybutok, 2015). Prior studies also indicate that leadership, workforce, and focus on customers are positively related to an organisation's results (Abdullah et al., 2008;Badri et al., 2006;He et al., 2011;Peng & Prybutok, 2015;Rahman & Bullock, 2005). These supportive TQM factors are essential prerequisites for effectively and efficiently developing and implementing other TQM practices, as they focus efforts toward continuous improvement (Abdullah et al., 2008). Specifically, leadership plays a critical role in determining values, orientations, and communications and establishing an organisation's focus on actions, including transforming an organisation's structure and culture (NIST, 2015). In addition, research indicates that TQM success is strongly associated with workforce and its related practices such as empowerment, training, and engagement (Ahire et al., 1996;Badri et al., 2006;Oakland, 2011). Without employee commitment and engagement, initiatives would fail (Ahire et al., 1996). On the other hand, focusing on customers is one of the most significant practices, as it is a critical predictor of results (Prybutok et al., 2011;Terziovski et al., 2003).
In this context, the MBNQA model emphasises the importance of engaging customers, including practices such as listening, building relationships, and effectively using customer information-all of which highlight the importance of customer satisfaction (NIST, 2015). Primary TQM factors such as operations management and measurement, analysis, and knowledge management have a significant effect on organisational results. Operations management focuses on designing, managing, improving, and innovating main products, services, and processes to create value for customers (NIST, 2015) and has positive impacts on results (Kaynak, 2003).
Furthermore, using quality information has a direct effect on results, as the purpose of measurement, analysis, and knowledge management is to direct operations and activities toward desired results (NIST, 2015). Therefore, leaders should strive to obtain employees' commitment (Tutuncu & Kucukusta, 2007) in decision-making by integrating workforce management, operations management, and quality information. In short, effective and integrated management of the organisation's resources represented by supportive and primary TQM factors have positive impacts on productivity as well as financial and non-financial results (Kaynak, 2003). Therefore, the following two hypotheses are proposed: H3. Primary TQM factors mediate the relationship between leadership, workforce, and customers and results.
H4. Strategy and primary TQM factors consecutively mediate the relationship between leadership, workforce, and customers and results.

Sample
The sample was provided by the Saudi Ministry of Commerce and Investment and consisted of 941 companies that have implemented TQM programmes in Saudi Arabia. Phone calls to the firms confirmed the accuracy and completeness of the list. A questionnaire was emailed to a random sample of 700 CEOs or quality managers. Of the 228 returned questionnaires, 217 were usable (RR: 31%) which is not uncommon in the Saudi context (e.g., Abosag & Peter, 2014;Al-Faraj & Alidi, 1992) and in BEM studies (e.g., Bou-Llusar et al., 2009;Moon et al., 2011;Wilson & Collier, 2000). Both response waves were analysed for non-response bias (Armstrong & Overton, 1977). Based on demographic variables, no differences were found between the two waves of data collection. Table 1 shows a summary of the sample demographics, firm sizes, and various sectors and activities (primary, secondary, and tertiary).

Measures
The literature (Bou-Llusar et al., 2009;Calvo-Mora et al., 2013) has suggested that models based on quality awards (e.g., MBNQA) fit the definition of TQM and could therefore be considered valid frameworks for TQM. In addition, previous academic studies have maintained that BEM/quality award material is an appropriate and comprehensive input for designing a questionnaire and considered as comprehensive (Calvo-Mora et al., 2018Bou-Llusar et al., 2005;Santos-Vijande & Alvarez-Gonzalez, 2007). While the MBNQA model does not explicitly refer to supportive and primary TQM factors, leadership, workforce, and customers criteria are consistent with supportive TQM factors, and operations management and measurement, analysis and knowledge management correspond to primary TQM factors (Flynn et al., 1995;Ho et al., 2001). The research model comprises six variables (Figure 1 and Appendix 1). Given that using pre-tested measures from previous empirical research were emphasised wherever possible (Bryman & Bell, 2015;Tata et al., 1999), Peng and Prybutok (2015) research served as the reference to identify the measures of leadership, workforce, customers, primary TQM factors, strategy, and results in the MBNQA context. The questionnaire was finalised based on several iterations that involved observations from four academics and four industry experts, pre-testing, and a pilot study to ensure the items were clearly stated.
To gauge the sub-criteria, items sharing the same sub-criterion were averaged to create composite measures. This is an established procedure (Landis et al., 2000;Little et al., 2002) that has been used in BEM studies (Bou-Llusar et al., 2009;Meyer & Collier, 2001) as well as in other fields (e.g., Zhang & Bartol, 2010;Zhou et al., 2010). In the scale validation process, composite measures are used to weigh more than one item and create score aggregates that were then introduced to confirmatory factor analysis (CFA) as indicator variables (Bagozzi & Edwards, 1998). Hence, each multi-item dimension was averaged to form a composite that served as an indicator variable of the latent construct to test the model.

Data analysis
Structural equation modelling (SEM) and maximum likelihood estimation were used to test the research model using Amos 25.0 (Arbuckle, 2014). Accordingly, Anderson and Gerbing (1988) comprehensive two-stage analytical strategy was followed. Further, SEM was adopted to test for mediation effects following the procedure recommended by Preacher and Hayes (2008). In this context, Amos was used to assess the whole model at once. This procedure does not depend on the assumption of normality for indirect effects, the significance of which was examined using a bias-corrected bootstrapping procedure with 10,000 samples (Preacher & Hayes, 2008).

Measurement model
The measurement model assessment is based on fit indices, convergent validity, and discriminant validity (Hair et al., 2010). A formalised, iterative process was followed to determine which items should be eliminated. Using modification indices and other diagnostics provided by Amos 25.0, item deletion was based on weak loadings, cross-loadings, communalities, error residuals, and theoretical determination (Hair et al., 2010;Prahinski & Benton, 2004). Consequently, 57 out of the 68 proposed items in the measurement scales were retained. Appendix 1 shows the factor loadings and dropped items.
The fit indices fell within an acceptable range for the purified measurement model, as shown in Table 2. It is of note that, whilst the λ 2 test was statistically significant, this test is well-known to be sensitive to sample size and may be found to be significant even when the differences between modelimplied and observed covariances are relatively small (Kline, 2010). Thus, multiple indices were used to evaluate model fit, as recommended by the SEM literature (Bollen, 1989;Schumacker & Lomax, 2010).
In terms of convergent validity, all average variance extracted (AVE) values, as presented in Table 3, are higher than the 0.50 cut-off (Fornell & Larcker, 1981;Shook et al., 2004). The factor loading for each indicator was also good (> 0.6) (Byrne, 2010;Hair et al., 2010). Construct reliability was assessed using Cronbach's alpha (Cronbach, 1951) and composite reliability (Fornell & Larcker, 1981). The resulting values are shown in Table 3. All constructs exceed the recommended threshold of 0.7 (Anderson & Gerbing, 1988). Discriminant validity was assessed by testing the square root of the AVE for each construct, which should be higher than its correlations with the other latent constructs (Fornell & Larcker, 1981). Table 3 shows that the square root of the AVE for each latent variable is higher than its correlations with other latent constructs. Discriminant validity was assessed by comparing Cronbach's alpha with the average inter-scale correlation (AVISC). Cronbach's alpha should show values greater than the AVISC for discriminant validity to be confirmed (Bou-Llusar et al., 2009;Bagozzi & Phillips, 1982). For all scales, this was the case and Cronbach's alpha was higher than the respective AVISC. Table 3 also reports the mean, standard deviation, and correlations of the unidimensional measurement constructs identified in the measurement model validation process. All correlations  (Browne & Cudeck, 1989;Bagozzi & Youjae, 1988;Kline, 2010;Medsker et al., 1994;Mulaik et al., 1989) Note: RMSEA: root mean square error of approximation; SRMR: standardised root mean square residual; CFI: comparative fit index; IFI: incremental fit index. were positive and significantly different from zero. This was the expected result, as the constructs are part of an integrated approach to TQM (Flynn & Saladin, 2001).
To examine common method bias, a number of statistical techniques were applied, such as Harman's one-factor test and different CFA specifications for the research model, such as the more comprehensive CFA with a single-factor model (Chang et al., 2010;Podsakoff et al., 2003) as well as indicators measuring strategy and leadership loading on the same factor, whereas all the other indicators loaded onto their intended factors (see for example, Kautonen et al., 2015). The result of an unrotated exploratory factor analysis showed that a single method factor does not explain the majority of the variance. In all the CFA specifications, the fit of the model was worse than that where all the indicators loaded onto their theoretically specified factors. This indicates that common method bias is unlikely to be a threat.

Structural model and hypotheses testing
Several nested models (Anderson & Gerbing, 1988;Eddleston & Kellermanns, 2007) were developed and nested model comparisons were performed (Byrne, 2010). The fit of the research model was compared with two alternative models, as shown in Table 4. In the first alternative (Model 2 in Table 4), the direct paths from leadership to PTQMP (primary TQM practices), from workforce to PTQMP, and from customers to PTQMP were removed. The χ 2 difference tests show that the research model is not significantly better than Model 2 (Δ χ 2 (2) = 2.308, n.s.). However, Model 2 has an inadequate fit to the data (χ 2 = 557.420, df = 222, χ 2 /df = 2.970, p = 0.00, CFI = 00.880, SRMR = 00.058, and RMSEA = 0.096), while the research model provides a more conservative test and is therefore retained for further analysis (Hayes, 2017). In the second alternative (Model 3 in Table 4), the direct paths from leadership to strategy, from workforce to strategy, and from customers to strategy were removed. This model has an adequate fit to the data (χ 2 = 561.579, df = 221, χ 2 /df = 2.821, p = 0.00, CFI = 00.908, SRMR = 00.0537, and RMSEA = 0.084) but is not significantly better than the research model (Δ χ 2 (2) = 6.467, p < 0.05). Overall, these results suggest that the research model has the best fit to the data (Model 1 in Table 4).
The structural model (Figure 1) was evaluated based on fit indices, magnitude, and significance of the structural path coefficients in combination with the R 2 values (Hair et al., 2010). A number of indices were used to determine the fit of data to the model (e.g., χ 2 /df, CFI, RMSEA, and SRMR), as shown in Table 4. These indicate that the overall model demonstrates an adequate fit (χ 2 = 555.112, df = 219, χ 2 /df = 2.535, p = 0.00, CFI = 0.908, SRMR = 0.0511, and RMSEA = 0.084).  Furthermore, the endogenous constructs (PTQMP and results) attain R 2 values of 0.840 and 0.770, respectively. This is higher than the substantial level of R 2 (Chin, 1998). However, as shown in Table 5, the strategy construct exhibits a value close to that level (0.539). Table 6 shows the results for hypotheses H2, H3, and H4 and indicates that the indirect effects of both leadership and workforce on PTQMP (the mediating role of strategy) are statistically significant: (indirect effect (A 1 C) = 0.075, p ≤ 0.050) and (indirect effect (A 2 C) = 0.046, p ≤ 0.050), respectively. The indirect effect of customers on PTQMP (the mediating role of strategy) is not statistically significant (indirect effect (A 3 C) = 0.017, p > 0.050). These results provide partial support of H2. More specifically, H2 is supported for the indirect effect of leadership and workforce but not supported for the indirect effect of customers.
In addition, the results show a lower direct effect of leadership on PTQMP (Table 6) compared with the greater direct effect of leadership on PTQMP when the mediating role of strategy is not introduced (Appendix 2). Such results may indicate that strategy partially mediates the relationship between leadership and PTQMP. Similarly, the results show a lower direct effect of workforce on PTQMP (Table 6) compared with the greater direct effect of workforce on PTQMP when the mediating role of strategy is not introduced (Appendix 2). Such results may indicate that strategy partially mediates the relationship between workforce and PTQMP.
For H4, the results show that the indirect effects of leadership, workforce, and customers on results (the consecutive mediating role of strategy and PTQMP) are statistically significant: indirect effect (A 1 CD) = 0.283, p ≤ 0.010; indirect effect (A 2 CD) = 0.208, p ≤ 0.050; and indirect effect (A 3 CD) = 0.397, p ≤ 0.001, respectively (Table 6). These results support H4.

Discussion
In general, the results supported the four hypotheses (H1, H2, H3, H4). Hence, the study advances the extant knowledge on how TQM practices can be aligned and integrated to achieve the desired results, especially in a BEM context. In this respect, supportive TQM factors (leadership, workforce, and customers) play a central role within QMS and organisational results are the starting point for initiating and managing TQM practices. They also have a different impact on other QMS components. Furthermore, the study findings explain the pattern in which strategy and primary TQM factors mediate the relationships with the other QMS components, indicating that they play a central mediating role in a sequential and consistent manner (H4) at a system level as a whole, and variably both at the level of the system (QMS) and according to a specific mediator (strategy or primary TQM factors) (H2 and H3). The study's main findings are as follows.
First, the results support H1. The analysis indicates that supportive TQM factors drive the QMS in the organisation, where leadership, workforce, and customer factors directly and positively influence primary TQM factors. This finding is consistent with those of prior research (e.g., Rahman &  Notes: Confidence intervals are bias-corrected based on 10,000 bootstrap samples. CI refers to confidence interval, and SE to standard error. Results based on two-tailed tests. All path coefficients reported in standardised form. Bullock, 2005), suggesting that effective leadership in the organisation oriented towards customers and an engaging and empowering workforce improve the effectiveness of primary TQM factors.
Second, support for H2 is found whereby the impact of supportive TQM factors (leadership and workforce) on primary TQM factors is enhanced via the mediating role of strategy. This means that developing, implementing, and reviewing strategy play a significant role in improving the effect of supportive TQM factors on primary TQM factors. In addition, information and tools improve over time and become more effective if integrated and connected with strategic planning. Indeed, the MBNQA model emphasises this issue by focusing on linkages: "The linkages among the criteria categories are an essential element of the systems perspective provided by the Baldrige framework. Some examples of these linkages are … the connection between workforce planning and strategic planning … the need for customer and market knowledge in establishing your strategy and action plans" (NIST, 2015, iii). In the context of BEMs, few studies have focused on the mediating role of strategy within QMS, although one study that does so was Suarez et al. (2016) using the EFQM model. The findings of the current study concur with these authors' findings. However, this study differs from theirs in two ways. First, it uses the MBNQA model. Second, it analyses the mediating role of strategy by operationalising supportive TQM factors as independent constructs, rather than as a combined construct. In other words, the results of this study show that integration with strategy (via its mediation role) is more pivotal and influential over time for leadership and workforce factors than for the customer factor in terms of the positive impact on primary TQM factors.
Third, support for H3 is found, as it is shown that primary TQM factors play a central role in explaining how supportive TQM factors affect organisational results. Specifically, the effective management of operations and quality tools and information improve the association between supportive TQM factors and organisational results. This finding is in line with the results of previous studies (Ho et al., 2001;Rahman & Bullock, 2005;Suarez et al., 2016). However, this study provides a deeper understanding of how the mediation effect differs according to the various supportive TQM factors. It is found that primary TQM factors play a greater mediating role between customers and results than between leadership and results. The results also indicate that primary TQM factors play a greater mediating role between leadership and results than between workforce and results.
Fourth, H4 is supported. Strategy and primary TQM factors play a consistent and sequential mediating role in improving the effect of supportive TQM factors (leadership, workforce, and customers) on results. In other words, the synergic and sequential mediating role of strategy and primary TQM factors facilitate the process through which supportive TQM factors may lead to desired results.
An interesting insight gained from this study is the need for integration among QMS components to succeed. That is, the results illustrate that (1) the mediation role of strategy in the association between customers and primary TQM factors is not supported and (2) both strategy and primary TQM factors play a relatively less significant mediation role in enhancing the indirect effects of workforce. However, sequentially and synergically, strategy and primary TQM factors play a greater mediation role in the association between supportive TQM factors (leadership, workforce, and customers) and results. Such findings show the need to recognise and accomplish integration among QMS components to achieve desired results.
In short, this study provides insights into the positive impact of TQM practices on organisational results as well as on the need for these practices to be conducted in an interrelated and integrated way. Specifically, it shows how the relationships among such practices can be managed as a system in the context of the MBNQA model according to the sequence in Figure 3.

Managerial implications
The business conditions usually lead managers to ask the following questions: Where should we start organisational improvement practices? What are the relationships of such practices? What are their relative significance and sequence? This study provides useful answers for such enquiries. In this context, the study findings emphasise the importance for managers and practitioners to properly implement TQM practices to achieve desired results. TQM failures are not necessarily caused by the approach itself, as the effective implementation of the TQM approach has a significant role in achieving success (as indicated by the different results of the studied relationships). In this vein, adopting and implementing TQM practices and tools is not the only requirement. Rather, there should be a clear progression starting from leadership's commitment and direction, which engages a qualified workforce that understands and effectively participates in managing operations, quality tools, and information through an integrated strategy that leads to desired results. This finding better equips firms to determine their priorities in investing resources and focus their efforts. In addition, the mediating role of strategy in the success of the integrative relationship among TQM practices indicates the long-term nature of TQM success that managers should take into account.

Limitations and future research
The cross-sectional nature of this study prevents definitive statements about the causal relationships (Hair et al., 2010;Kline, 2010). Therefore, further research following a longitudinal study design is necessary to further confirm the causality of the suggested relationships. In addition, perceptual data were used to gauge the BEM criteria. It is important to recognise the possibility of same-source bias. This implies that future research is needed that uses multiple sources of information to further test this study's findings.

Funding
The author received no direct funding for this research.

Disclosure statement
No potential conflict of interest was reported by the author.  The organisation achieves high quality performance that applies through all facets of the organisation.

Citation information
The organisation maintains effective communication channels to deliver the values and expectations of senior leaders to employees.
Management in the organisation sets strategy, goals, and objectives.
Management in the organisation establishes and reinforces an environment that fosters empowerment and innovation.
Management encourages and supports organisational and employee learning.

Communication and organisational performance-0.627
The organisation regularly evaluates all functions' performance and capabilities.* The organisation utilises performance reviews to identify opportunities for improvement and innovation.

Governance and social responsibility-0.871
Management in the organisation cares about the impacts of its products, services, and operations on society.
The organisation actively supports and strengthens the relationships with key communities, such as religious and educational organisations and professional associations.

Strategy Development Process-0.856
The organisation establishes a short-term (1-2 years) plan to help achieve goals and objectives.
The organisation establishes a long-term (2-5 years) plan to help achieve goals and objectives.

Strategic Objectives-0.903
The organisation establishes a strategy/plan to improve customer satisfaction.
The organisation establishes human resource requirements and plans considering employees' capabilities and needs.
The organisation establishes a strategy/plan to strengthen supplier-partner relationships.

Strategy Implementation-0.833
The organisation establishes a strategy/plan to address key goals and objectives.
The organisation uses performance measures to track the progress of action plans.
The organisation allocates resources to achieve overall action plans.

Voice of the Customers-0.879
The organisation measures and analyses customer satisfaction and dissatisfaction.
The organisation compares its customer satisfaction results with its competitors or other benchmarks.

Product Offerings and Customer Support-0.893
The organisation has an official method to determine current product/service requirements and customer expectations.
The organisation has an official method to determine future product/service requirements and customer expectations.
The organisation has an official method to identify customer groups and market segments.

Customer Relationships-0.853
The organisation implements effective customer relationship management practices to ensure customers seek assistance.
The organisation continuously improves its customer relationship management practices.* The organisation determines key customer requirements and delivers them to all employees in the response chain.* The organisation resolves customer complaints promptly and effectively. (Continued)

(Continued)
The organisation officially examines customer complaints to improve its processes.* 4. Primary TQM Practices 4.1. Measurement, Analysis, and Improvement of Organisational Performance-0.853 The organisation offers effective measurement techniques to ensure the following qualities of data and information are achieved (1-5): Reliability Consistency Accessibility Review* Timely update The organisation regularly compares its performance with its competitors or other benchmarks to support its performance, evaluation, and improvement.* 4.2. Knowledge Management, Information, and Information Technology-0.863 The organisation systematically analyses performance data and information collected internally to support its overall quality goals.
The organisation systematically analyses performance data and information collected externally to support its overall quality goals.

Operations-0.607
The organisation establishes a systematic method to introduce new products/services, including the following items (1-3): Including customer requirements in the design. Addressing quality issues early in the design cycle.
Analysing relevant process capabilities.
The organisation monitors the processes producing products/services to identify necessary actions to make corrections.
The organisation continuously improves the processes used to provide its products and services.
The organisation formally assesses the quality of its (6-8): Products/services Production/delivery systems Goods/services supplied by external suppliers/partners The organisation communicates its quality requirements to all external suppliers of goods/services.* 5. Workforce 5.1. Workforce Capability and Capacity-0.715 The organisation establishes human resource plans derived from its strategic plans to achieve the full potential of its workforce.

Workforce Climate-0.804
The organisation tries to build a work environment beneficial to the followings items (2-5): Performance excellence* Full involvement Personal growth Organisational growth* The organisation supports a work environment that is beneficial to the well-being and growth of all employees.

Organisational Culture-0.712
The organisation promotes cooperation, individual initiatives, innovation, and flexibility.

Assessment of Engagement-0.806
The organisation regularly examines employee satisfaction and utilises the results to support quality and innovation.

Performance Management-0.811
The organisation's compensation, recognition, and related reward practices encourage high performance.