Design: an overview of key research stages, methods and approaches

The study was then designed by breaking the entire process of construct development into 5 stages. Each of these 5 stages corresponded to tackling the essential questions mentioned in the previous section. The overall set of approaches, methods and techniques that were relevant for construct development, and essential questions that each stage sought to answer in the process are the subject of description in this section. Fig. 3. is provided here from this point of view, in order to illustrate, and to give a good overview of the entire construct development and validation process. Given its research objectives, all 5 stages were

9See Appendix A.1. for a brief abstract of this publication.

10See Appendix A.2. for a brief abstract of this publication.

envisaged to be completed within the present dissertation. A stage-wise explanation, and brief descriptions are as follows. Since all of these stages shall be given a more in-depth treatment in each individual chapter that covers these, the reader is referred to their original locations in this dissertation for the bulk of the discussion regarding their execution.

Figure 3 An overview of key research stages, methods and approaches for construct development

1.10.2.1. Stage 1: type of decision factors required for operationalising environmental complexity

The opening stage corresponded to the question of “how do we know which factors”

operationalise environmental complexity. More precisely, this stage sought to look into the broader subject area of environmental complexity research in order to find out the type or categories of factors that have been used in literature. Since an explicit operationalisation of the construct of environmental complexity was new to the dissertation domain literature of logistics and SCM, literature reviews at this stage were meant to look beyond into the broader organisational research literature, and methods used therein in order to see how the construct had been operationalised. This stage is given an exclusive treatment in Chapter 3, where the

“Methods & issues in construct measurement” are discussed.

1.10.2.2. Stage 2: type of decision factors required for operationalising supply chain logistics environmental complexity (1)

This stage may be referred to as the first of the two sub-stages that sought to answer the question of “how do we know which SCM factors”. For this, this stage sought to look more specifically into the domain literature of logistics and supply chain management in order to find relevant studies, and to build correspondences with those studies that most closely matched the description of environmental complexity research within the domain literatures.

In other words, a starting point for embarking on a more detailed literature review process was sought at this stage. This stage is given exclusive treatment in Chapter 5, where

“Specific categories and factors of supply chain logistics environments”, are discussed. Once again, literature reviews were to represent the bulk of the approach employed here in order to identify and short-list studies, which were then meant to serve as a starting point for the next stage involving more in-depth reviews of the domain literature. For this purpose, since the construct was new to the domain literature of logistics and supply chain management, a broader mapping of the construct was envisaged not only within ‘parent’¹¹ literature in operations management, but also within the context of the parent construct of environmental complexity, which is environmental uncertainty. The reader may refer to literature reviews and the meta-analysis presented in Chapter 3 in order to get a grasp of these issues.

Furthermore, the reader is also referred to Chapter 5, where parity between various categories of decision factors is established.

11 One may argue for a different term than ‘parent’ based on Mentzer et al. (2008) and Frankel et al. (2008), who delve into the relationship between the disciplines of logistics/SCM and OM.

1.10.2.3. Stage 3: decision factors required for operationalising supply chain logistics environmental complexity (2)

This stage was envisaged as the second of the two sub-stages that sought an answer to the question of “how do we know which SCM factors”; the first has just been described in the previous sub-section. Based on the input from the previous stage, a more detailed literature review in the form of a content analysis was to be performed here. Furthermore meta analyses were to be employed for guiding these literature reviews and for concentrating the results of the content analyses in a meaningful manner. The reader is referred to Chapter 6, where the initial and first rounds of content analyses are described, and where the specific techniques employed in order to short-list the decision factors that were deemed as being important for operationalising the construct, are described in more detail.

1.10.2.4. Stage 4: (information) measures for measuring the decision factors

Stage 4 was to mark the beginning of data collection activities in this study, as specific

(information) measures, that provided information on supply chain environmental complexity through their respective decision factors, were to be sought within the objectives of this study. It was envisaged that undertaking an in-depth content analysis of relevant material within the problem domain, would bring out a range of measures that could be linked to each decision factor. The ensuing content analysis, and its findings are described in detail in Chapter 6. Furthermore, it was envisaged that a theoretical model on the construct, with a hierarchy of decision factors and (information) measures would emerge at the conclusion of this stage.

1.10.2.5. Stage 5: construct validation - content validity

Stage 5 was envisaged to deal with aspects of content validity and construct validity. More specifically it was to deal with aspects of internal validity of the theoretical model on the construct that would have emerged from the previous stage. This was envisaged in the form of the following objectives:

• Validating the theoretical model in terms of whether each decision factor was important for operationalising the construct.

• Short-listing only those measures from the previous stage, that provided important information for assessing or measuring each decision factor.

It was therefore envisaged that a small (e.g. pilot) test and screening empirical study in the spirit of Lewis et al. (2005), could check for internal validity of the model by garnering responses from subject matter experts. It was also envisaged that aiming for more than

internal validity of the theoretical model, e.g. external validity, would remain beyond the confines of this dissertation because of the methodological delimitations outlined earlier. The reader is referred to Chapter 7 in order to gain more insight on this stage, where more

detailed objectives and findings of this test study are documented.