Logistics Theory Building - Produktionsekonomi ... Theory Building 11 interpretivism, there are in...

7Logistics Theory Building

Logistics Theory Building

Building theory can be thought of as a never-ending journey. Theory building is particularly important todisciplines that are emerging and growing. Compared to older and more established academic disciplines,logistics does not have a rich heritage of theory development. This paper aims to construct a framework thatcombines different research paradigms with research approaches for logistics theory building. This frameworkcan be used for positioning studies that aim at building and articulating core logistics theories. The frameworkis illustrated by providing examples from logistics research adhering to different research paradigms (positivism,scientific realism and interpretivism) and using different research approaches (deduction, induction, andabduction). The paper discusses how these different research paradigms and research approaches contributeto theory building in their own way.

Gyöngyi Kovács* and Karen M Spens**

© 2007 The Icfai University Press. All Rights Reserved.

* Assistant Professor (Acting), Swedish School of Economics and Business Administration (Hanken), Departmentof Marketing, Helsinki, Finland; and the corresponding author. E-mail: [email protected]

* * Acting Professor, Swedish School of Economics and Business Administration (Hanken), Department ofMarketing, Helsinki, Finland. E-mail: [email protected]

IntroductionLogistics has neither a unified theory (Mentzer et al., 2004) nor a rich heritage of theorydevelopment (Stock, 1997). Theory building, however, is argued to advance a disciplineand profession, and is thus relatively more important to an emerging and growingdiscipline (Swanson, 2000). Logistics just like any other scientific discipline aims tocreate new, or modify existing theories (Arlbjørn and Halldórsson, 2002). Theorybuilding can take different forms and follow different paths (Sebeok, 1983; and Nesher,1999). As for logistics, borrowing theories from other disciplines has been advocated asa viable option for advancing the discipline (Stock, 1997; and Arlbjørn and Halldórsson,2002). Borrowing has the advantage of learning from others and avoiding the pitfall ofreinventing the wheel of building a theory.

At the same time, logistics research has often been discussed to be heavily footed inpositivism and predominantly employs a deductive research approach (Mentzer andKahn, 1995; Garver and Mentzer, 1999; Arlbjørn and Halldórsson, 2002; and Näslund,2002). This has, however, its limitations for theory building, as the deductive derivationof hypotheses can only modify existing theories (Gioia and Pitre, 1990; and Bjereldet al., 2002). Yet, other research approaches also find their way to logistics research.The use of the inductive research approach in logistics is gaining momentum (Kovácsand Spens, 2005). However, even this approach has been criticized by many, for instancePeirce (1934) stated that “[it] never can originate any idea whatever. No more candeduction”. Despite this criticism, we argue that the deductive and the inductiveresearch approaches are both relevant paths for advancing logistics knowledge. Kovácsand Spens (2005) also discuss the use of a third, the abductive, research approach in

The Icfai Journal of Supply Chain Management, Vol. IV, No. 4, 20078

logistics research. Looking at different research paradigms and approaches, this paperthereby advocates a paradigm-based approach to theory building. In this we adhere toGioia and Pitre (1990) who state that the grounding of theory in paradigm-appropriateassumptions helps in avoiding the common tendency of trying to force-fit functionalist/positivist theory building techniques as a universal approach. The aim of the paper isto construct a framework that combines different research paradigms with researchapproaches for logistics theory building. The framework can be used as a tool forpositioning studies that aim at building and articulating logistics core theories. Weexemplify the framework with articles using different research approaches in differentresearch paradigms in logistics.

The paper begins with a discussion on theory building and the views of differentresearch paradigms on theory. Thereafter, different research approaches are discussed forbuilding theory. Building on this, a framework is presented that combines differentresearch paradigms and research approaches for theory building in logistics. Next, thisframework is discussed through examples from articles in logistics journals. A summarizingdiscussion of theory building in logistics concludes the paper.

Theories and Research ParadigmsKnowledge can be described as a multilevel abstraction of reality, and different layers ofknowledge can be distinguished (Grønhaug, 2002). Its base consists of facts,observations, experiences and perceptions, which when recorded constitute the data.The data, when organized tabulated and presented logically and meaningfully becomesinformation. Information, when analyzed and processed with care can lead to inferences,generalizations and insights. Insights based on empirical observations are derivedinductively, but insights can be arrived at deductively also (Ahmad, 2000). Insights canlead to formulation of concepts that become the tools for thinking, analysis anddiscussion. Concepts are then the basic elements for building theory. (Ahmad, 2000).Figure 1 exhibits this sequence leading to theory building.

The top two boxes in Figure 1 indicate a separation between scientific theory andmanagerial problem solving. The ultimate goal of scientific research is to create newtheories or test and/or modify existing theories (Arlbjørn and Halldórsson, 2002).Logistics literature and research have so far focused on managerial problem-solvingmainly (Mentzer and Kahn, 1995). The body of knowledge for building theories exists,even concepts are fairly well developed but the step of building scientific theories is notyet accomplished.

But what is theory? The term ‘theory’ can mean different things, e.g., ordering-frameworks used for predicting and explaining empirical events, conceptualizations ortheorizing, or even just hypotheses or explanations (Sayer, 1992). Vafidis (2002), in areview of Nordic doctoral dissertations in logistics also concludes that theoriesseemingly mean different things to different authors. Many of the theories referred toand used as a basis in the dissertations were not (yet) mentioned in the Dictionary ofTheories (Bothamley, 2004) and even models, frameworks and concepts were labelled


theories. If we adhere to the definition of theory by Sutherland (1975, as referenced inWeick, 1989) which argues that theory is “an ordered set of assertions about a genericbehavior or structure assumed to hold throughout a significantly broad range of specificinstances”, then models and frameworks would not qualify as theory. The same goes forreferences, data, variables, diagrams and hypotheses (Sutton and Staw, 1995). On theother hand, Peter and Olson (1983) contend that “scientific ideas consist of inventedconstructs and hypothesized relationships among them”. Similarly, Gioia and Pitre(1990) provide a loose definition of theory as it is defined as “any coherent descriptionor explanation of observed or experienced phenomena”. They argue that this loosedefinition of theory is needed because different paradigms offer a wide scope oftheoretical representations (Gioia and Pitre, 1990). Turning again to the issue of theorybuilding, the authors advocate what they call a paradigm-based approach.They argue that the grounding of theory in paradigm-appropriate assumptions helps inavoiding the common tendency of trying to force-fit functionalist/positivist theorybuilding techniques as a universal approach (Gioia and Pitre, 1990).

In this paper the paradigm-based approach is adopted, however, many differentcategorizations exist for research paradigms. Arlbjørn and Halldórsson (2002)distinguish between positivism, post-positivism and hermeneutics; Gioia and Pitre(1990) adopt the framework distinguishing between interpretivism, radical humanism,radical structuralism and functionalism; Hirschman (1986) contrasts the positivisticmetaphysic to the humanistic one; Mentzer and Kahn (1995) distinguish betweenpositivism and interpretivism; Solem (2003) between positivism, critical realism andinterpretivism, etc. Common to all these distinctions are their two extreme ends,

-

Figure 1: Different Abstraction Levels

Scientific Theory ManagerialProblem-Solving

Concepts

InsightsKnowledge

Information

Data


constituted of positivism at one end and interpretivism at the other (Mangan et al.,2004). There is less agreement on what constitutes the ‘in-between’. While Arlbjørnand Halldórsson (2002); and Solem (2003) refer to critical realism only, Muncy andFisk (1987) name many different approaches within scientific realism in this‘in-between’. According to Hunt (1993), scientific realism is an umbrella term for allthese different versions of realism, ranging from social realism (closest tointerpretivism within scientific realism, Peter and Olson, 1983; and Sayer, 1992) tocritical realism (closest to positivism; Boyd, 1984; Leplin, 1984 and 1997; and Hunt,1990, 1991 and 1993). Other authors again call scientific realism for relativism(Muncy and Fisk, 1987).

The following discussion will start from the most common paradigm in logistics,namely positivism (Mentzer and Kahn, 1995; Arlbjørn and Halldórsson, 2002; andNäslund, 2002) and then proceed through scientific realism to interpretivism.

PositivismPositivism is by far the most dominant research paradigm (Kirkeby, 1990; Alvesson andSköldberg, 1994; and Indick, 2002), thus it is not surprising that it also dominates inlogistics research (Garver and Mentzer, 1999). Modern logical positivists call themselvesempiricists (Boyd, 1984) to indicate the high standards they demand of empiricalresearch (Indick, 2002). Positivism strives to discover the true nature of reality,believing in one universal absolute truth (Peter and Olson, 1983; and Mentzer andKahn, 1995). Research following this paradigm examines regularities and relationshipsthat lead to generalizations and ideally universal principles (Gioia and Pitre, 1990).These universal principles are not only uncovered, but can be used for normativepredictions (Anderson, 1983). The universality claim is supported by a strong belief inthe possibility to conduct research truly objectively and value-free (Peter and Olson,1983; and Hirschman, 1986). Objectivity is deemed possible because of the rationality-claims of positivism that include the possibility to detach science from cultural, social,political and economic factors, and to detach the measured subject from themeasurement process (Peter and Olson, 1983). Positivism can also be labelledfunctionalism (Gioia and Pitre, 1990).

Theories in positivism are judged by their predictive capabilities and formal elegance.Thus, the truth of a theory is judged by the truth of its predictions (Leplin, 1997).Research in a positivist paradigm is corroborated with empirical observations that arein line with its predictions, and falsified through any even minor deviating observation(Popper, 1959). Positivism in fact relies on empirical testing as the sole means of theoryjustification (Anderson, 1983). Any new hypothesis that is taken up and that enduredin empirical testing changes the original theory and thus creates a different theory(Peter and Olson, 1983).

Scientific RealismEven though scientific realism is used as an umbrella term for many different approaches(Hunt, 1993; and Muncy and Fisk, 1987) that lie ‘in-between’ positivism and


interpretivism, there are in fact some commonalities to all these approaches. Scientificrealism does not, indeed, refute the ideas of truth and falsity (that would be scepticism),just replaces the notion of “absolute truth” with that of a contextualized, ‘approximate’or “relative truth” (Boyd, 1984; and Muncy and Fisk, 1987). The questioning of theexistence of an absolute truth even leads to the notion that science is seen as a processwithout an ultimate goal (Anderson, 1983).

Common to scientific realists, however, is their shared belief that the world existsindependently of our knowledge of it (Leplin, 1984; and Sayer, 1992). Knowledge doesn’texist without any context, in fact all our knowledge products are theory-laden (Sayer,1992) and cannot be detached from the approach that produced them (Boyd, 1984; andAnderson Hudson and Ozanne, 1988). Scientific realists contextualize knowledge inmany ways, in essence data is created and interpreted in terms of a variety of theoriesand theoretical commitments (Peter and Olson, 1983; and Boyd, 1984). Some socialrealists see theories even as the outcome of a negotiation process between scientists ina particular research community (Peter and Olson, 1983). A problematic notion couldbe that what counts as scientific knowledge is determined by the community thatproduces this knowledge (Anderson, 1983). In other terms, “a successful theory is onewhich is treated seriously and studied by a significant portion of a research community”(Peter and Olson, 1983, p. 112). One important notion of scientific realism is itscommitment to pluralism (Solem, 2003). No theory, research approach, methodology ormethod is advocated as the sole path for science.

However, scientific realism should not be used as an excuse for ‘anything goes’ (Hunt,1990). What makes an explanation, and a theory, right, is that it gives us what we needto solve a problem in some context (Leplin, 1997). Not truth, but the pragmatic valueof a theory to explain and solve empirical problems should be the criterion for theoriesto be appraised in scientific realism (Anderson, 1983; and Halldórsson and Aastrup,2003). Nonetheless, scientific realism follows Popperian falsificationism in many ways.The explanatory, predictive, and pragmatic success of a theory is the factor providingevidence for the existence of its associated entities (Hunt, 1991).

Interpretivism

Interpretivism is that discipline of science that is concerned with the interpretationof meaning (Sayer, 1992). The goal of theory building in the interpretive paradigm isto generate descriptions, insights and explanations of events so that the system ofinterpretations and meaning and the structuring and organizing processes are revealed(Gioia and Pitre, 1990). In interpretivism, the researcher and phenomenon aremutually interactive (Hirschman, 1986) and the researcher becomes part of theevolving events studied (Gioia and Pitre, 1990). Interpretivism refutes the notion ofabsolute truth due to refuting the possibility for truly objective ways of conductingscience (Peter and Olson, 1983). In fact, seeking objectivity is even deemedundesirable in interpretivism (Hunt, 1993). Thus, interpretivism does not seek truthbut variety, i.e., wants to show the many versions of subjective impressions.


As Mentzer and Kahn (1995) put it, “interpretivism portrays reality as a collective ofmultiple socially constructed realities”. In sociology, interpretivism is commonlycriticized for its subject-ridden level advocating that sensate experiences require thepresence of an actor (Bertilsson, 2004).

The interpretative researcher collects data that are relevant to the informants andaims at preserving their unique representations. Analysis begins during data collection,thereafter analysis, theory generation and further data collection go hand in hand whichmeans that the theory building process is iterative, cyclical and nonlinear. Through theprocess, tentative speculations are confirmed or unconfirmed by consulting informants(Gioia and Pitre, 1990). Academics coming from a positivist background tend to claimthat theories without a predictive function are pre-scientific (Cox, 1996). Interpretivists,however, have no ambition to build theories for predictions (Mentzer and Kahn, 1995).The purpose of an interpretivist theory can be descriptive or explanatory but notpredictive.

Theory Building through Different Research ApproachesLogistics research is largely following the hypothetico-deductive research approach(Mentzer and Kahn, 1995; Arlbjørn and Halldórsson, 2002; and Näslund, 2002). Itsdominance is so striking that Kovács and Spens (2005) conclude that researchapproaches are usually discussed in logistics research if deviating from deduction. Infact, Cox (1996) even views any non-deductive study as pre-scientific, and Mentzerand Kahn (1995) argue with the dominance of hypothetico-deductive research inlogistics for disregarding other research approaches when suggesting a framework forlogistics research.

In the following, we will discuss the three research approaches in the order of theirusage in logistics (Kovács and Spens, 2005), deduction, induction and abduction, beforepresenting a framework of their application for theory building within different researchparadigms.

Deduction

The deductive research approach follows the path from theoretical advances to theirempirical testing. The process starts with an established theory or generalization, andseeks to test whether the theory applies to specific instances (Hyde, 2000), thusfollowing the path from a general law to a specific case (Alvesson and Sköldberg, 1994;and Johnson, 1996), in other words from rule to case to result (Kirkeby, 1990; andDanemark, 2001). More specifically, deductive research starts with logical conclusionsthat were derived from theory and are presented in the form of hypotheses orpropositions. Apart from logic, a good operationalization of the studied phenomena isrequired for deduction (Bjereld et al., 2002). The hypotheses or propositions aresubsequently tested in an empirical setting, and general conclusions are presented basedon the corroboration or falsification of the prior self-generated hypotheses/propositions(Kirkeby, 1990; and Arlbjørn and Halldórsson, 2002).


The deductive research approach has often been criticized for not being able togenerate new theories (Peirce, 1931, etc.) but just modify or refine theories (Gioia andPitre, 1990; and Bjereld et al., 2002). Nevertheless, deductive research contributes totheory building in its own way. Gioia and Pitre (1990) see the tested hypotheses orpropositions along with their verifications or falsifications as revising or extendingthe original theory. According to Peter and Olson (1983, p. 112), “any change in atheory creates a modified product—i.e., a different theory”. Thus creating newhypotheses that are logically deduced from a theory and testing them also qualifies astheory building.

Even though the deductive approach is most commonly associated with a positivistparadigm (Kirkeby, 1990; Alvesson and Sköldberg, 1994; Mentzer and Kahn, 1995;Arlbjørn and Halldórsson, 2002; and Näslund, 2002), it finds its application in otherparadigms as well. Scientific realism uses the deductive research approach for refiningits theories as well, only the testing is not seen as asserting its hypotheses orpropositions absolute truth (Boyd, 1984; and Muncy and Fisk, 1987). This puts thepredictive force of deduction into perspective, as it is questionable how the outcome ofindividual phenomena could be predicted based on universal facts without the questionof probabilities (Bertilsson, 2004). Thus, while the deductive research approach whenused in the positivist paradigm can be described as deterministic, it is only probabilisticin scientific realism. The purpose of the deductive approach can be propositional orpredicative (Danemark, 2001). Propositions are related to one another with theconjunctions ‘not’, ‘and’, ‘or’, ‘if…then’, while predicative logic also examines ‘all’ and‘no’ junctions.

Research using deductive reasoning can be purely theoretical, thus skipping thephase of empirical testing, however, the deductive research approach asks for a testingphase. Different research methods can be used for this, ranging from (quantitative)methods including simulations, model building and statistical surveys (Halldórsson andAastrup, 2003) to (qualitative) structured interviews (Hyde, 2000).

InductionThe inductive approach follows the path from empirical observations to theoreticaladvances. The process starts with a specific empirical case or a collection of observationsto general law, i.e., from facts to theory (Alvesson and Sköldberg, 1994), in other wordsfrom case to result to rule (Kirkeby, 1990; and Danemark, 2001). Previous theoreticalknowledge is not a necessity (Gioia and Pitre, 1990), and even if not ruled out, notsystematic (Bjereld et al., 2002). Instead, empirical observations lead to emergingpropositions and their generalization in a theoretical frame.

Generalization may, however, not be the purpose of all inductive research.Interpretive theory building tends to be inductive in nature (Gioia and Pitre, 1990)but does not aim at generalizations but to show subjective varieties (Hirschman,1986). Thus the purpose of inductive research forms a continuum from subjectivevariety in interpretivism towards categorizations and the development of taxonomies


to generalizations in scientific realism (Solem, 2003). Natural science research usesthe inductive approach also in positivism and even adheres to the ideas ofverification (Danemark, 2001). In business logistics research, however, induction(and abduction) are seen as deviating from the dominating deductive positivism(Mentzer and Kahn, 1995; and Kovács and Spens, 2005) and thus turned to whenleaving the positivist paradigm. The outcome of inductive research is a theory thatis tentatively accepted a priori and has been deemed as holding stand with a highprobability for all possible further experiences (Anderson, 1983). Nonetheless,inductive research has also been criticized for not being able to originate radicallynew theories (Peirce, 1934), because once rejecting a proposition, it lacks the powerof modifying action (Nesher, 1999).

Inductive research is often confused with qualitative research and vice versa.While not all qualitative research is inductive (see, e.g., structured interviews indeduction), inductive research can also encompass quantitative methods. This is thecase in data mining, where researchers engage in reinterpreting data from a newperspective. The purpose of this type of inductive study is to analyze data from adifferent angle than the original purpose of its collection, or recontextualize andreinterpret this data. A reason for data mining can be if a previous quantitative studyis non-conclusive or shows very interesting outliers. While this delimits the role ofquantitative studies for the interpretivist paradigm, interpretivists claim that evenstatistical data has to be interpreted when analyzed, and that in fact all research wouldbe interpretive (Gummesson, 2003).

AbductionThe abductive research approach was introduced by Peirce (1931). While somescientific realists see abduction as rooted in empiricism (Boyd, 1984), Peirce himselfhas been described as a scientific realist (Bertilsson, 2004). The outcome of theabductive research process is a “relative truth” (Sebeok, 1983) like in scientificrealism (Boyd, 1984; and Muncy and Fisk, 1987), which can be seen in Peirce’s (1934,p. 106, § 171) following description: “Deduction proves that something must be;Induction shows that something ‘actually’ is operative; Abduction merely suggeststhat something ‘may be’.”

There are two ways to initiate abductive research. In the first, the researcher makesan empirical observation that deviates from a previously used theoretical frameworkand thus challenges the usability of this framework. This empirical observation iscommonly called a “puzzling point” (Eco, 1983; Alvesson and Sköldberg, 1994; Duboisand Gadde, 2002; and Kovács and Spens, 2005). In this case, the puzzling pointinitializes a search for matching theories that can be used to explain the deviatingobservation (Kirkeby, 1990; and Dubois and Gadde, 2002).

The second way to start an abductive process is more conscious. Here theresearcher applies a new theory, or a new framework, to already existing observations(Kirkeby, 1990). The novelty of the theory does not need to be universal, the claim


is that it has not been used in this stream of research (e.g., introducing chaos theoryto logistics research). Borrowing theories from other disciplines thus can initiate anabductive research process. However, doing so implies a belief in the possiblecoexistence of theories, as in scientific realism (Anderson, 1983; and Leplin, 1997).

The aim of abductive reasoning is to understand the new phenomenon (Alvessonand Sköldberg, 1994) and to suggest new theory (Kirkeby, 1990) in the form of newhypotheses or propositions. The abductive research process closes with the applicationof these hypotheses/propositions in an empirical setting (Alvesson and Sköldberg,1994). This last step can also be characterized as a deductive part of the research.

While both induction and deduction have been largely criticized for theirrestrictions in creating (radically) new theories (Peirce, 1934; Gioia and Pitre, 1990; andDanemark, 2001), abduction is seen as being the most creative research approach(Kirkeby, 1990) for theory building. With this creative power, abduction can introduceradically new ideas and is thus best suited for the (in Kuhn’s, 1970, terms) revolutionaryphase of theory building. The abductive approach stems from a pragmatic view of science(Nesher, 1999; and Bertilsson, 2004). Hence, it is most suitable for the scientific realistparadigm that evaluates its theories in terms of their pragmatic applicability (Anderson,1983) and usefulness (Hunt, 1991; and Leplin, 1997).

Theory Building in Logistics ResearchThe division into three research paradigms, positivism, scientific realism andinterpretivism, provides a good overview of the differences to theory building in general.Turning to studies in logistics and Supply Chain Management (SCM), we candistinguish different schools and research communities along the three main paradigmsor in Arbnor and Bjerke’s (1997) terms, three methodological approaches (Gammelgaard,2004). These schools differ along their methodological, ontological and metaphysicalcommitments (Anderson, 1983). As for methodological commitments, the positivistparadigm largely corresponds with the analytical approach, the scientific realistparadigm with the systems approach, and the interpretivist paradigm with the actorsapproach (Arbnor and Bjerke, 1997; and Gammelgaard, 2004). The analytical approachin unison with positivism emphasizes the need for objectivity in research (Arbnor andBjerke, 1997), while interpretivism specifically requires the presence of an actor(Bertilsson, 2004) and advocates subjectivism.

These different paradigms and their respective theories should not be seen ascompeting with each other, they rather fulfil complementary purposes (Solem,2003). Even though, or maybe even because research in logistics is strongly footedin the positivist paradigm (Mentzer and Kahn, 1995; Garver and Mentzer, 1999;Arlbjørn and Halldórsson, 2002; and Näslund, 2002), logistics researchers haveoften called for complementary approaches (Seaker et al., 1993; Arlbjørn andHalldórsson, 2002; and Näslund, 2002). The different paradigms shed differentlight on empirical phenomena and can help to solve different (managerial andtheoretical) problems.


Similarly, different research approaches contribute to theory building in different ways.Even though the deductive research approach dominates in logistics research, in order tobuild logistics theory, a combination of all three approaches would be needed, however,this does not mean that they should all be used in one study. Instead, the trio of abduction,deduction and induction complement each other (Sebeok, 1983; and Nesher, 1999).Meredith (1993) describes the cycle of research to go from description to explanation totesting. Different research approaches are suitable for different phases (Nesher, 1999) inthis cycle. Scientific revolutions (Kuhn, 1970) can be driven through the creative powerof abductive research, which introduces radically new ideas and theories (comp. Nesher,1999). To fortify these yet rather loose theories (Lakatos, 1970; and Arlbjørn andHalldórsson, 2002), induction can be used to further develop them, expanding existingknowledge. Once these theories have been established, after the transition from loose tosolid theories (Arlbjørn and Halldórsson, 2002; and Lakatos, 1970), deductive researchhelps to test them and further establish them. In other words, the abductive approachdiscovers new concepts and rules, which the inductive approach can evaluate and that areinferred in the deductive approach (Nesher, 2002). Nesher (2001) also distinguishesbetween the logic of discovery (abduction), the logic of consequence (deduction) and thelogic of evaluation (induction). The scrutiny of falsification is needed in all phases(Popper, 1959), however, in non-positivist paradigms bearing in mind the probabilitiesassociated with any suggested hypothesis or proposition.

Thus, the use of the three different research approaches adheres to different worldviews, and different underlying research paradigms, are yielding different results in

Figure 2: Framework for Theory-Building in Logistics

Paradigm

ResearchApproach

Inductive

Deductive

Abductive

Interpretivism Scientific Realism Positivism

Process

Method

Purpose

Process

Method

Purpose

Process

Method

Purpose

from empirical observations to theoretical propositionsgrounded –––––– qualitative –––––––– non-structured

quantitative

variety ––––––––––––––––––––––––– generalizationdescriptive/exploratory/explanatory ––––– prescriptive

Notapplicable

Notapplicable

from theoretical hypotheses to empirical observationssemi-struct. ––– qualitative –––struct.

quantitative

probabilistic –––––– generalization ––––– deterministicdescriptive/explanatory/predictive

from emp.observationsto theor.propositions

to applicationqualitativequantitative

theory suggestiondescriptive/exploratory/explanatory/predictive

in respective phases

Notapplicable

Notapplicable


building logistics theory. While following a deductive research approach inpositivism results in universal claims, the findings of deductive research in scientificrealism are deemed probable but not universal. Nonetheless, Nesher’s (2002) trio ofresearch approaches complement each other for the purposes of theory building.Figure 2 combines the presented paradigms with research approaches in a frameworkfor logistics theory building. In doing so, representations of the typical researchprocesses, methods and purposes are indicated for each combination of researchparadigms and approaches.

For example, in the inductive research approach research processes in theinterpretivist and scientific realism paradigms proceed through empirical observationsto theoretical propositions. The methods used e.g., in the scientific realist paradigm canrange from quantitative to qualitative, purposes vary, however the ultimate goal, thatis the far end of the continuum, stresses generalization of the findings.

Illustrating the Framework in Logistics ResearchExamples from current logistics research are used to illustrate the framework for theorybuilding for this particular discipline. In the following, we will discuss each of theresearch approaches in the different research paradigms; starting with deduction andinduction in positivism; deduction, induction and abduction in scientific realism, andinduction in interpretivist logistics research.

Positivism in Logistics ResearchLogistics literature often makes a link between the positivist paradigm and a ‘deductiveresearch approach’ (Mentzer and Kahn, 1995; Garver and Mentzer, 1999; Arlbjørn andHalldórsson, 2002; and Näslund, 2002). Typically, quantitative methods are employed inthis category, mainly regression analysis (Keller, 2002; Richey et al., 2002; Bhatnagaret al., 2003; and Zsidisin et al., 2003), or Structural Equation Modelling (SEM) (Autryand Daugherty, 2003; Kent and Mentzer, 2003; and Wisner, 2003).

In natural sciences, the ‘inductive research approach’ adheres to the idea ofverification and can thus be used within a positivist paradigm (Danemark, 2001).However, in logistics, the inductive research approach is mainly turned to whenpurposefully leaving the positivist paradigm. Solem (2003) discusses inductive researchin scientific realism and interpretivism only. ‘Abduction’ is also used for leaving thedominant research paradigm in logistics. Any theory-matching process implies apluralistic belief in the coexistence of theories, which can be found in scientific realism(Anderson, 1983; and Leplin, 1997).

Scientific Realism in Logistics ResearchEven though the ‘deductive research approach’ is often linked to positivism, it is notconfined to this research paradigm. Deductive hypothesis-testing in scientific realism,however, asserts hypotheses only relative but not absolute truth (Boyd, 1984; and Muncyand Fisk, 1987). This relativizes the generalizability or universality of the findings of


such research (Bertilsson, 2004). In other words, deduction in positivism is deterministic,but the same research approach is only probabilistic in scientific realism.

This is probably most striking for the data analysis method of SEM. Most articlesthat use SEM in logistics research (which currently dominates the publications ofthe Journal of Business Logistics) reportedly use maximum likelihood in theirconfirmatory factor analysis, which is presumably a sign for a positivist background.At the same time, though, many articles using SEM in logistics test alternativemodels as well and/or discuss the exploratory nature of their survey. Positivistsproclaiming absolute truth for one theory (Hirschman, 1986; and Gioia and Pitre,1990) would, however, not be inclined to seek alternative models and hypothesesto their theory, this is rather a sign of scientific realism (Solem, 2003). Alternativemodel testing and exploratory surveys can therefore be considered as indicators fora scientific realist paradigm.

Qualitative deductive studies adhering to scientific realism can also be found inlogistics. As an example, Harrison (1999) e.g., presents a very deductive, dual macro-micro model when describing different logistics systems, which is then applied in hisstudy. According to his article, two quantitative studies precede the discussed study,which is nonetheless a qualitative one, using semi-structured interviews as its datacollection technique. A predetermined set of codes is then used to categorize thetranscribed quotes of these interviews. As the data analysis strictly follows prior setpropositions and categories, this research is purely deductive. Harrison (1999) evensets system boundaries for the data analysis and questions the possibility of absolutegeneralization of his findings. This indicates that the study appertains to the scientificrealist paradigm.

Qualitative data collection can also precede a quantitative data analysis inscientific realism. Moberg and Speh (2003) use structured interview guides fortelephone interviews to evaluate the strength of relationships in a supply chain, butthen use factor analysis to analyze this data. Zacharia and Mentzer (2004) even usetheir data from “in-depth” interviews to develop a structural equation model for thesalience of logistics.

Inductive scientific realist research can also be found in logistics. According toSolem (2003), the purpose of inductive research ranges from manifesting subjectivevarieties in interpretivism to arriving at taxonomies and (relative) generalizations inscientific realism. Miemczyk and Holweg (2004) e.g., explore the effects of customizationon inbound logistics in a study that uses semi-structured interviews and processmapping alongside quantitative data about inventory levels prior to developing theirpropositions. To be able to generalize their findings within particularly set systemboundaries, Miemczyk and Holweg (2004) triangulate their data from different datacollections. Hingley (2001) also uses a multi-case, multi-site approach to allow for thegeneralization of his findings within its industrial and market context when examiningrelationship management in the food industry. But while both these studies are


qualitative, there are also examples of quantitative inductive research in scientificrealism. Mathematical models are deductive if a model is only applied and refined in theprocess, however, they are inductive if they are built from scratch e.g., in a case study(de la Fuente and Lozano, 1998).

Some authors combine two studies in one article before presenting their supposedlygeneral conclusions. Dadzie (1998) e.g., sets survey categories deductively a priori to hisstudy but then goes on investigating inductively emerging clusters in his data analysis.Svensson (2001) derives his propositions about disturbances in logistics flows from aninductive (and in this case, qualitative) study before evaluating these in a deductivesurvey. The combination of two such fundamentally different studies is only possiblewithin scientific realism that allows for pluralistic thinking.

The theory-matching element of abduction implies pluralistic thinking. Consciouslyapplying a new theoretical framework to investigate known empirical observations ina field can be seen as borrowing theories. According to Stock (1997), theory borrowingis common in logistics research. It even helps in avoiding the pitfall of “reinventingthe wheel”, i.e., having to develop these theories again even though they are alreadyknown in other disciplines. This, however, is not the only reason abductive researchpertains to scientific realism. The aim of abduction to suggest theories instead ofarriving at generalizations and universal claims points at the relativistic view of thisresearch approach on its outcome. This resembles scientific realism (Sebeok, 1983;Boyd, 1984; and Muncy and Fisk, 1987). This supports the proposition of ourframework that the abductive research approach adheres to the scientific realistparadigm (Figure 2).

Examples for the abductive research approach in logistics originate from mathematicalmodeling but also action, and constructive research. Westwood (1999) starts out withthe puzzling point of forecasting methods being insufficient and not applicable for aparticular case company. He then suggests a solution from this empirical backgroundwhile negotiating between theory and data from his case study. This solution is toinclude inter-organizational transfers in forecasting. He then develops a mathematicalmodel to suggest his theory, and reapplies the model to the data from his case study toevaluate its use. Campbell et al. (2001) also develop a mathematical model for a city astheir case study, before applying their findings to a larger geographical area andpresenting conclusions on the basis of this application. Holmström et al. (1999) studyprocess networks creating their own examples, i.e., companies are asked to implementa process network design while the implementation process is studied. They negotiatebetween different alternative theoretical frameworks before developing guidelines forprocess networks that are subsequently implemented in their prior examples beforearriving at their conclusions.

Interpretivism in Logistics ResearchLogistics researchers often make a link between interpretivism, inductive research, andqualitative studies. More and more inductive research is indeed published in logistics


journals (Spens and Kovács, 2006). However, Hilmola et al. (2005) found that about halfof inductive case studies in the discipline use quantitative research methods. But whilenot all inductive research is qualitative, and does not necessarily adhere to aninterpretivist paradigm, examples where this triple linkage holds stand can be found inlogistics literature.

In an article by Golicic et al. (2002), the authors set out to “examine the impact ofthe dimensions of e-commerce on managing relationships in the supply chain” (Golicicet al., 2002, p. 852). In order to do so, they interview employees of eight companies whoare responsible for activities related to the company’s supply chain. They stress theimportance of on-site, in-depth interviews to seek the perceptions of the intervieweeson e-commerce. Codes and categories for the collected data emerged during thetranscription of interviews, and the conceptualization of main themes in the researchis then purely based on the empirical data.

Another example for interpretivist studies in logistics can be found in Pålsson(2006), in which he describes the four phases he used in a participant observatorystudy going from a preparation phase especially focusing on the finding of gatekeepersto the empirical study, data collection through observations and interviews on site,analysis and reflection emphasising the aspect of emerging interpretations of the data,and finally, the phase of writing up the research. The study is about theimplementation of Radio Frequency Identification (RFID) technology in packaging andfood manufacturing companies in Sweden, apart from the technology provider. Noteven the goal of the RFID implementation is given a priori but determined throughinterviews with the actors. Conclusions at the end of the research are presented asin comparison with existing literature in the field. Pålsson (2006, p. 6) highlights animportant aspect of interpretivist studies when commenting on the resulting articlebeing his own “interpretation of many interpretations” by the interviewees.

ConclusionThe arena of theory building research can be thought of as a never-ending journey(Swanson, 2000). Logistics has no unified theory (Mentzer et al., 2004). However,looking at different ways to theory building we can contend that not one but manypossibilities exist to develop logistics theory, or indeed even theories. Theory buildingis a very important discipline that is emerging and growing. Theory building in itself,as important as it is, should not be taken lightly. The aim should always be to developa good theory. Theories are judged by different criteria depending on their underlyingresearch paradigms (Leplin, 1997). Good theory is a plausible theory, and it is judgedto be more plausible and of high quality if it is “interesting rather than obvious,irrelevant or absurd, obvious in novel ways, a source of unexpected connections, highin narrative rationality, aesthetically pleasing or correspondent with presumedrealities” (Weick 1989).

What makes an explanation, and a theory, right, is that it gives us what we need tosolve a problem in some context (Leplin, 1997). All judgments about theories are thus


context-dependent (Sayer, 1992). Different research paradigms and research approachescontribute to theory building in their own (but very different) way.

In most fields, even natural science, many paradigms can coexist (Anderson,1983). It is thus important to realize that different research communities coexist alsoin logistics. While it is important to distinguish between these schools, they cannotbe evaluated as better or worse than the other. So far, no consensus in science existsas to the nature or even the very existence of a unique scientific method (Anderson,1983), paradigm or research approach. This even suggests that it is inappropriate toseek a single best method (Anderson, 1983), paradigm or research approach for theevaluation of logistics theory. Often many theories can come to the same predictionabout particularities of an observation (Leplin, 1997). Which theory is confirmed, orcorroborated through a particular observation is yet to be determined (Leplin).Depending on its underlying paradigm, scientific theories can be evaluated in termsof their truth content, or in terms of their usefulness (Anderson, 1983; and Peter andOlson, 1983).

Although theory building can evolve from mainstream approaches to research,rebellious behavior has and will continue to result in new insights and theories beingdeveloped (Trim and Lee, 2004). The dominance of a research paradigm or researchapproach in a discipline poses strong limitations on how theory in that discipline canevolve. Meredith (1993) criticizes operations management research for skipping thedescriptive phase of research and thus becoming more and more unrealistic. Ifmanagement researchers are to produce new insight into complex managementproblems they need to be innovative and have the confidence to question whatconstitutes appropriate research (Trim and Lee, 2004). However, introducing radicallynew theories, perspectives and research approaches can be compared to launchingdiscontinuous innovations (Peter and Olson, 1983). Theory building research oftenreceives rough treatment from journals and reviewers and has an uncomfortablejourney getting published (Swanson, 2000; and Halldórsson and Aastrup, 2003). Peterand Olson (1983) conclude that new scientific streams have to be carefully ‘marketed’before being introduced in their fields. This paper markets the paradigm-basedapproach to theory building and presents the abductive approach in order tocomplement the research approaches that have already established their existence inthe logistics literature.

References1. Ahmad Abad (2000), “Research with a Purpose”, Journal of Management Research,

Vol. 1, No. 1, pp. 5-11.

2. Alvesson Mats and Sköldberg Kaj (1994), Tolkning och reflektion, Vetenskapsfilosofi ochkvalitativ metod, Studentlitteratur, Lund.

3. Anderson Hudson Laurel and Ozanne Julie L (1988), “Alternative Ways of SeekingKnowledge in Consumer Research”, Journal of Consumer Research, Vol. 14 (March),pp. 508-521.


4. Anderson Paul F (1983), “Marketing, Scientific Progress, and Scientific Method”,Journal of Marketing, Vol. 47 (fall), pp. 18-31.

5. Arbnor Ingeman and Bjerke Björn (1997), Methodology for Creating Business Knowledge,Sage Publications, London.

6. Arlbjørn Jan Stentoft and Halldórsson Árni (2002), “Logistics Knowledge Creation:Reflections on Content, Context and Processes”, International Journal of PhysicalDistribution and Logistics Management, Vol. 32, No. 1, pp. 22-40.

7. Autry Chad W and Daugherty Patricia J (2003), “Warehouse Operations Employees:Linking Person-Organization Fit, Job Satisfaction, and Coping Responses”, Journal ofBusiness Logistics, Vol. 24, No. 1, pp. 171-197.

8. Bertilsson Thora Margareta (2004), “The Elementary Forms of Pragmatism: OnDifferent Types of Abduction”, European Journal of Social Theory, Vol. 7, No. 3,pp. 371-389.

9. Bhatnagar Rohig, Jayaram Jayanth and Phua Yue Cheng (2003), “Relative Importanceof Plant Location Factors: A Cross National Comparison Between Singapore andMalaysia”, Journal of Business Logistics, Vol. 24, No. 1, pp. 147-170.

10. Bjereld Ulf; Demker Maria and Hinnfors Jonas (2002), Varför Vetenskap?,Studentlitteratur, Lund.

11. Bothamley Jennifer (2004), Dictionary of Theories: More than 5000 Theories, Laws andHypotheses Described, Barnes and Noble, USA.

12. Boyd Richard N (1984), “The Current Status of Scientific Realism”, in Leplin Jarrett(Ed.), Scientific Realism, University of California Press, pp. 41-82, Berkeley.

13. Campbell James F, Labelle Alain and Langevin Andre (2001), “A Hybrid TravelDistance Approximation for a GIS-based Decision Support System”, Journal ofBusiness Logistics, Vol. 22, No. 2, pp. 165-181.

14. Cox Andrew (1996), “Relational Competence and Strategic Procurement Management,Towards an Entrepreneurial and Contractual Theory of the Firm”, European Journalof Purchasing and Supply Management, Vol. 2, No. 1, pp. 57-70.

15. Dadzie Kofie Q (1998), “Management Education for Physical Distribution andLogistics”, International Journal of Physical Distribution and Logistics Management, Vol. 28,No. 4, pp. 259-271.

16. Danemark Berth (2001), Explaining Society: An Introduction to Critical Realism in theSocial Sciences, Routledge, Florence, KY.

17. de la Fuente David and Lozano Jesús (1998), “Determining Warehouse Number andLocation in Spain by Cluster Analysis”, International Journal of Physical Distribution andLogistics Management, Vol. 28, No. 1, pp. 68-79.

18. Dubois Anna and Gadde Lars-Erik (2002), “Systematic Combining: An AbductiveApproach to Case Research”, Journal of Business Research, Vol. 55, Issue 7, pp. 553-560.


19. Eco Umberto (1983), Horns, Hooves, Insteps. Some Hypotheses on Three Types ofAbduction, in Eco Umberto and Sebeok Thomas A (Eds.), The Sign of Three, Dupin,Holmes, Peirce, Indiana University Press, pp. 198-220, Bloomington.

20. Gammelgaard Britta (2004), “Schools in Logistics Research? A MethodologicalFramework for Analysis of the Discipline”, International Journal of Physical Distributionand Logistics Management, Vol. 34, No. 6, pp. 479-491.

21. Garver Michael S and John T Mentzer (1999), “Logistics Research Methods:Employing Structural Equation Modelling to Test for Construct Validity”, Journal ofBusiness Logistics, Vol. 20, No. 1, pp. 33-57.

22. Gioia Dennis A and Evelyn Pitre (1990), “Multiparadigm Perspectives on TheoryBuilding”, The Academy of Management Review, Vol. 15, No. 4, pp. 584-602.

23. Golicic Susan L, Davis Donna F, McCarthy Teresa M and Mentzer John T (2002),“The Impact of E-commerce on Supply Chain Relationships”, International Journal ofPhysical Distribution and Logistics Management, Vol. 32, No. 10, pp. 851-871.

24. Grønhaug Kjell (2002), “Is Marketing Knowledge Useful?”, European Journal ofMarketing, Vol. 36, No. 3, pp. 364-372.

25. Gummesson Evert (2003), “All Research is Interpretive!”, Journal of Business andIndustrial Marketing, Vol. 18, Nos. 6/7, pp. 482-492.

26. Halldórsson Árni and Aastrup Jesper (2003), “Quality Criteria for QualitativeInquiries in Logistics”, European Journal of Operational Research, Vol. 144, Issue 2,pp. 321-332.

27. Harrison Alan (1999), “Horses for Courses: Human Aspects of DifferentLogistics Systems”, International Journal of Logistics: Research & Applications, Vol. 2,No. 3, pp. 247-261.

28. Hilmola Olli-Pekka, Hejazi Annukka and Ojala Lauri (2005), “Supply ChainManagement Research Using Case Study Analysis: A Literature Analysis”,International Journal of Integrated Supply Management, Vol. 1, No. 3, pp. 294-311.

29. Hingley Martin (2001), “Relationship Management in the Supply Chain”, InternationalJournal of Logistics Management, Vol. 12, No. 2, pp. 57-71.

30. Hirschman Elizabeth C (1986), “Humanistic Inquiry in Marketing Research:Philosophy, Method and Criteria”, Journal of Marketing Research, Vol. 23 (August),pp. 237-249.

31. Holmström Jan, Hoover William E Jr., Eloranta Eero and Vasara Antti (1999), “UsingValue Reengineering to Implement Breakthrough Solutions for Customers”, InternationalJournal of Logistics Management, Vol. 10, No. 2, pp. 1-11.

32. Hunt Shelby D (1990), “Truth in Marketing Theory and Research”, Journal ofMarketing, Vol. 54, No. 3, pp. 1-15.


33. Hunt Shelby D (1991), “Positivism and Paradigm Dominance in Consumer Research:Toward Critical Pluralism and Rapprochement”, Journal of Consumer Research, Vol. 18(June), pp. 32-44.

34. Hunt Shelby D (1993), “Objectivity in Marketing Theory and Research”, Journal ofMarketing, Vol. 57, No. 2, pp. 76-91.

35. Hyde Kenneth F (2000), “Recognising Deductive Processes in Qualitative Research”,Qualitative Market Research: An International Journal, Vol. 3, No. 2, pp. 82-90.

36. Indick William (2002), “Fight the Power: The Limits of Empiricism and the Costsof Positivistic Rigor”, The Journal of Psychology, Vol. 136, No. 1, pp. 21-36.

37. Johnson Colleen (1996), “Deductive Versus Inductive Reasoning: A Closer Look atEconomics”, Social Science Journal, Vol. 33, No. 3, pp. 287-299.

38. Keller Scott B (2002), “Driver Relationships with Customers and Driver Turnover:Key Mediating Variables Affecting Driver Performance in the Field”, Journal ofBusiness Logistics, Vol. 23, No. 1, pp. 39-64.

39. Kent John L and Mentzer John T (2003), “The Effect of Investment inInterorganizational Information Technology in a Retail Supply Chain”, Journal ofBusiness Logistics, Vol. 24, No. 2, pp. 155-175.

40. Kirkeby Ole Fogh (1990), “Abduktion”, in Andersen Heine (Ed.), Vetenskapsteori ochmetodlära, Introduktion, Translated from Danish by Carl G Liungman, 1994,Studentlitteratur, Lund.

41. Kovács Gyöngyi and Karen M Spens (2005), “Abductive Reasoning in LogisticsResearch”, International Journal of Physical Distribution and Logistics Management, Vol. 35,No. 2, pp. 132-144.

42. Kuhn Thomas S (1970), The Structure of Scientific Revolutions, University of ChicagoPress, 2nd Edition, Chicago, IL.

43. Lakatos Imre (1970), “Falsification and the Methodology of Scientific ResearchProgrammes”, in Lakatos Imre and Musgrave Alan (Eds.), Criticism and the Growth ofKnowledge, Proceedings of the International Colloquium in the Philosophy of Science, 1965,Vol. 4, pp. 91-196, London.

44. Leplin Jarrett (1984), “Introduction”, in Leplin Jarrett (Ed.), Scientific Realism,University of California Press, pp. 1-7, Berkeley.

45. Leplin Jarrett (1997), A Novel Defense of Scientific Realism, Oxford University Press,Oxford.

46. Mangan John, Lalwani Chandra and Gardner Bernard (2004), “CombiningQuantitative and Qualitative Methodologies in Logistics Research”, InternationalJournal of Physical Distribution and Logistics Management, Vol. 34, No. 7, pp. 565-578.


47. Mentzer John T and Kahn Kenneth B (1995), “A Framework of Logistics Research”,Journal of Business Logistics, Vol. 16, No. 1, pp. 231-250.

48. Mentzer John T, Soonhong Min and Michelle Bobbitt L (2004), “Toward a UnifiedTheory of Logistics”, International Journal of Physical Distribution and LogisticsManagement, Vol. 34, No. 8, pp. 606-627.

49. Meredith Jack (1993), “Theory Building Through Conceptual Methods”, InternationalJournal of Operation and Production Management, Vol. 13, No. 5, pp. 3-11.

50. Miemczyk Joe and Holweg Matthias (2004), “Building Cars to Customer Order—What Does it Mean for Inbound Logistics Operations?”, Journal of Business Logistics,Vol. 25, No. 2, pp. 171-198.

51. Moberg Christopher R and Speh Thomas W (2003), “Evaluating the RelationshipBetween Questionable Business Practices and the Strength of Supply ChainRelationships”, Journal of Business Logistics, Vol. 24, No. 2, pp. 1-19.

52. Muncy James A and Fisk Raymond P (1987), “Cognitive Relativism and the Practiceof Marketing Science”, Journal of Marketing, Vol. 51 (January), pp. 20-33.

53. Näslund Dag (2002), “Logistics Needs Qualitative Research—Especially ActionResearch”, International Journal of Physical Distribution and Logistics Management, Vol. 32,No. 5, pp. 321-338.

54. Nesher Dan (1999), “Peirce’s Theory of Signs and the Nature of Learning Theory”,in Shapiro M (Ed.), The Peirce Seminar Papers, Berghahn Books, pp. 349-388, Oxford:Berghahn Books,New York.

55. Nesher Dan (2001), “Peirce’s Essential Discovery: ‘Our Senses as ReasoningMachines’ Can Quasi-Prove Our Perceptual Judgments”, Transactions of the Charles SPeirce Society, Vol. 37, No. 1 (Winter), pp. 23-57.

56. Nesher Dan (2002), “Peircean Epistemology of Learning and the Function ofAbduction as the Logic of Discovery”, Transactions of the Charles S Peirce Society,Vol. 38, Nos. 1/2 (Winter), pp. 175-206.

57. Pålsson Henrik (2006), “Participant Observation in Logistics Research: Experiencesfrom an RFID Implementation Study”, in Jahre Marianne (Ed.), Logistics in the NewValley, CD-ROM Proceedings of the 18th Annual Conference for Nordic Researchersin Logistics, NOFOMA 2006, Oslo, Norway.

58. Peirce Charles Sanders (Santiago) (1931), Collected Papers of Charles Sanders PeirceVolume I: Principles of Philosophy, Hartshorne Charles and Weiss Paul (Eds.), HarvardUniversity Press, Cambridge, MA.

59. Peirce Charles Sanders (Santiago) (1934), Collected Papers of Charles Sanders PeirceVolume V: Pragmatism and Pragmaticism Logic, Hartshorne Charles and Weiss Paul(Eds.), Harvard University Press, Cambridge, MA.


60. Peter J Paul and Olson Jerry C (1983), “Is Science Marketing?”, Journal of Marketing,Vol. 47 (Fall), pp. 111-125.

61. Popper Karl (1959), The Logic of Scientific Discovery, Routledge Classics (2002 Reprint),London.

62. Richey R Glenn, Daugherty Patricia J, Genchev Stefan E and Autry Chad W (2004),“Reverse Logistics: The Impact of Timing and Resources”, Journal of Business Logistics,Vol. 25, No. 2, pp. 229-250.

63. Sayer Andrew (1992), Method in Social Science, A realist approach, Routledge, 2nd Edition,London.

64. Seaker Robert F, Waller Matthew A and Dunn Steven C (1993), “A Note on ResearchMethodology in Business Logistics”, Logistics and Transportation Review, Vol. 29, No. 4,pp. 383-387.

65. Sebeok Thomas A (1983), “One, Two, Three Spells UBERTY”, in Eco Umberto andSebeok Thomas A (Eds.), The Sign of Three, Dupin, Holmes, Peirce, Indiana UniversityPress, pp. 1-10, Bloomington.

66. Solem Olav (2003), “Epistemology and Logistics: A Critical Overview”, SystemicPractice and Action Research, Vol. 16, No. 6, pp. 437-453.

67. Spens Karen M and Kovács Gyöngyi (2006), “A Content Analysis of ResearchApproaches in Logistics Research”, International Journal of Physical Distribution andLogistics Management, Vol. 36, No. 5, pp. 374-390.

68. Stock James R (1997), “Applying Theories from Other Disciplines to Logistics”,International Journal of Physical Distribution and Logistics Management, Vol. 27,Nos. 9/10, pp. 515-539.

69. Sutton Robert I and Staw Barry M (1995), “What Theory is Not”, AdministrativeScience Quarterly, Vol. 40, No. 3, pp. 371-384.

70. Svensson Göran (2001), “Firms’ Preventive Activities and the Occurrence ofDisturbances in the Inbound and Outbound Logistics Flows”, International Journal ofLogistics: Research and Applications, Vol. 4, No. 2, pp. 207-236.

71. Swanson Richard A (2000), “Theory and Other Irrelevant Matters”, Human ResourceDevelopment International, Vol. 3, No. 3, pp. 273-278.

72. Trim Peter R J and Lee Yang-Im (2004), “A Reflection on Theory Building and theDevelopment of Management Knowledge”, Management Decision, Vol. 42, Nos. 3/4,pp. 473-480.

73. Vafidis Dimitrios (2002), “Methodological Tendencies in Logistics Research—Empirical Evidence form 25 Finnish and Swedish Doctoral Dissertations 1994-1998”, Publications of the Turku School of Economics and Business Administration,Series D-1:2002, Turku.


74. Weick Karl E (1989), “Theory Construction as Disciplined Imagination”, Academy ofManagement Review, Vol. 14, No. 4, pp. 516-531.

75. Westwood John B (1999), “Retail Inventory Movement—A Case Study inRationalisation”, International Journal of Physical Distribution and Logistics Management,Vol. 29, Nos. 7/8, pp. 444-452.

76. Wisner Joel D (2003), “A Structural Equation Model of Supply Chain ManagementStrategies and Firm Performance”, Journal of Business Logistics, Vol. 24, No. 1,pp. 1-25.

77. Zacharia Zach G and Mentzer John T (2004), “Logistics Salience in a ChangingEnvironment”, Journal of Business Logistics, Vol. 25, No. 1, pp. 187-210.

78. Zsidisin George A, Ellram Lisa M and Ogden Jeffrey A (2003), “The RelationshipBetween Purchasing and Supply Management’s Perceived Value and Participation inStrategic Supplier Cost Management Activities”, Journal of Business Logistics, Vol. 24,No. 2, pp. 129-154.

Reference # 34J-2007-12-01-01

Logistics Theory Building - Produktionsekonomi ... Theory Building 11 interpretivism, there are in...

Documents

Transcript of Logistics Theory Building - Produktionsekonomi ... Theory Building 11 interpretivism, there are in...