روابط احتیاطی در اندازه شرکت، مدت زمان مدیریت کیفیت جامع (TQM)، تشکیل اتحادیه و زمینه صنعت در پیاده سازی مدیریت کیفیت جامع (TQM) تمرکز بر اثرات کل

Past TQM literature has been inconclusive in resolving the validity of the universalistic approach versus the contingency approach of TQM implementation. In this paper we contribute to the resolution of this debate by using a total effects approach within the contingency perspective. We propose a culture-quality system design-outcomes framework for TQM implementation. We then use this framework to study the differences in total effects relationships among TQM constructs across four contingencies, firm size, TQM duration, unionization, and industry type. Data from a sample of 394 plants (SIC codes 20 through 39) are used to test the validity of our contingency model. Our results revealed that firm size, TQM duration, and industry type moderate the influence of total effects of culture (top management commitment, customer focus, and trust) on final outcomes (process quality, product quality, and customer satisfaction). These contingencies also moderate the influence of total effects of quality system design (design management, training, empowerment, quality information usage, supplier quality management, and process quality management) on final outcomes. The strongest contributor to variation in total effects across groups was industry type, followed by size and then TQM duration. To a lesser extent, unionization was a moderator in total effects relationships. These findings uniquely contribute to the emerging contingency theory of TQM implementation.

Total quality management (TQM) has been widely adopted by firms in the last 50 years and yet firms report less than optimal results (Sousa and Voss, 2008, Sila, 2007 and Powell, 1995). In a study of 225 electronics manufacturing firms in Hong Kong, Yeung et al. (2006) reported that advanced TQM firms and non-TQM firms do not differ in their organizational performance. Our study contributes to the contingency perspective research stream which argues that the success of TQM depends on contextual factors (e.g., Sousa and Voss, 2008, Sila, 2007, Shah and Ward, 2003, Powell, 1995 and Benson et al., 1991). Specifically, our research objectives are: (1) To study the effects of culture and quality system design on outcomes; and (2) To examine direct and total effect relationships among culture, quality system design and outcomes that are attributable to contextual factors. We propose a culture-quality system design-outcomes model that link key TQM constructs in a network of relationships. As relationships similar to this model have been established previously (see Section 2), we only provide a brief summary of this model and then move on to our main research intent of examining contingency relationships.

While there could be structural differences in baseline TQM frameworks (Bou-Llusar et al., 2009, Ghosh et al., 2003 and Anderson et al., 1995), we focused this study on the relatively new domain of contingency theory. Even within the contingency theory, the simultaneous consideration of four key contextual influences and a focus on total effects of culture and quality system design on outcomes is a novel contribution. Our study found support for an overall model that suggests a combinatorial influence of culture and quality system design on outcomes. Moreover, this study also found support for direct and total effects in a network of relationships linking culture, quality system design and TQM outcomes that varied according to contextual factors. Our research has two limitations. First, we used survey items which were perceptual and hence more prone to respondent bias than objective items. Nevertheless, the perceptual items included were based on past seminal works (see Appendix A), and the distribution of data did not suggest the possibility of systematic bias in our sample. Second, our list of contingency factors was not exhaustive as would be expected. But within this context, it is probably the most comprehensive examination of multiple contingencies that have received scant attention in the past literature (Table 1). Our cross-sectional design offers consistency across industry types but is still restrictive in terms of temporal maturity. In other words, the longitudinal aspects of TQM evolution cannot be captured in a static study such as ours. Caution is therefore urged for interpreting our findings only in light of “like” structural conditions. This study also paves way for further research to unravel the opposing directionality of some of our significant paths across multiple groups. This puzzling pattern of crossover effects pinpoints the importance of fine-grained contingency analyses that could lead to more robust and consistent results thereby strengthening efforts of theory building.