Magma topp logo Til forsiden Econa

Dr. Roy D. Howell is currently the J. B. Hoskins Professor of Marketing in the Rawls College of Business at Texas Tech University, a position he has held since 1982. His research interests are in the areas of Marketing Strategy and Research Methods, particularly structural equation modeling.

Sigurd Villads Troye er professor ved Institutt for organisasjon og ledelse ved Norges Handelshøyskole. Han er siviløkonom fra NHH og har doktorgrad i marketing fra University of Illinois at Urbana-Champaign.

Toward a new understanding of marketing: Gaps and opportunities

Marketing research has been subject to criticism from members of the discipline and from persons outside the discipline. The critical issues that have been raised can be classified into four different domains: first, marketing researchers have been criticized for their selection of phenomena to address (it has been argued that too much attention has been devoted to certain phenomena, while others have been ignored); a related second criticism holds that marketing research is managerially irrelevant, somehow fails to provide support and does not give direction to decision makers; third, marketing researchers have been accused of developing and applying ontological and theoretical frameworks that do not satisfactorily account for the phenomena being addressed; and, fourth, marketing research has been reproached for using research approaches and methods that do not properly challenge theories being put to test or do not provide an adequate description of the subject matter.

We posit that the content of each of these domains of criticism to some extent reflects the developments in other domains and give some examples of how these influences occur. Three of these domains correspond closely to Brinberg and McGrath's (1988) notion of the substantive domain, the conceptual domain and the methodological domain.

The substantive domain, according to Brinberg and McGrath (1988, p. 25), contains «the phenomena, processes, or focal problems of interest» and comprises aspects of reality with which we are concerned. The conceptual domain consists of the «ideas, concepts, and their relations as well as the philosophical assumptions underlying them.» It contains theories and models of how states and events in the substantive domain are interrelated as well as fundamental ontological assumptions about reality, such as «modernism» versus «post-modernism» (Firat and Venkatesh, 1995). The methodological domain «contains the methods, designs, and research strategies used to examine concepts and phenomena» and covers on one side specific sampling, design, measurement and analytical approaches and fundamental epistemological positions on the other.

The «managerial or decision domain» can be considered to be a subset of the phenomena referred to in the substantive domain, since the phenomena it contains are the decision makers' choices and decision strategies. We have chosen to treat it as being distinct from the other domains in order to emphasize that marketing practice may not be fully determined by developments in the other domains and vice versa: what managers and other decision makers do that affects the marketplace may not be fully absorbed by the theories that are developed and phenomena that are focused on as part of the discipline's substantive domain.

The purpose of this article is twofold. The first goal is to provide a framework that can help provide a systematic inventory of alleged shortcomings in the marketing discipline. In order to illustrate this framework we will give a review of some recent criticisms that have been raised, although it is obvious that such a review cannot be exhaustive. The second goal reflects the premise of this article that, in order to meet the challenges and fill the gaps implied by the criticisms, there is a need to understand the dynamics of knowledge generation in a discipline, such as marketing. Using the Brinberg and McGrath (1988) framework discussed above, we discuss the interplay among the domains in the progress of scientific advancement in marketing.

Criteria for evaluating the adequacy of marketing research

While a long array of scientific «criteria» have been suggested in the philosophy of science literature, we follow Larry Laudan (Laudan, 1977, 1989) who uses the term «adequacy» to evaluate research traditions: «We are'essentially asking here how effective theories within the research tradition are at solving problems» (Laudan, 1989, p. 375). «Problems» can be puzzles of understanding (e.g., whether theories at hand help us understand and predict the phenomena we are concerned with). The ability of theories to solve problems can however also be tied to managerial issues: Can theories adequately prescribe managerial action? A theory (e.g., attitude theory) that does well in terms of accounting for individual behaviour (e.g., brand choice) may thus not provide a sufficient basis for managerial decisions (e.g., with respect to positioning or advertising). The adequacy requirement can be extended to other aspects of a paradigm, or discipline, which we will return to in closer detail below.

Theoretical adequacy of a discipline can be obtained at the cost of the inadequacies of its substantive domain in the sense that researchers may be reproached for solving the «wrong» problems (i.e., for not attending to phenomena which from some normative perspective they should be concerned with). At the same time it is conceivable that a discipline, regardless of the adequacy of its theoretical foundation and span of attention, may not be equipped with a methodology that adequately helps us capture developments in the substantive domain and allows us to test and refine existing theories. 1

The criticisms against marketing can be grouped as indicated above in three different categories that correspond to three general criteria for judging marketing as an academic field: attentional, theoretical and methodological adequacy.

Attentional adequacy

The first issue addresses whether thescope or attention of marketing research is adequate. Does marketing research address phenomena on which attention should be focused? One requirement is that empirical research should (at least) capture the phenomena suggested by our theories; otherwise, marketing could be considered as an abstract exercise limited to armchair reasoning. Whereas such a criticism might be justifiable in the early period of marketing and still may apply to certain isolated fields in marketing, we do not believe this in general is a valid criticism. Most theories introduced in marketing are put to empirical tests with little delay.

Another requirement is that marketing researchers should also be equipped to capture phenomena that may not follow from extant theories or theoretical frameworks, but that nevertheless may impinge on important marketing aspects. Such phenomena as the Internet, cellular phones, just-in-time manufacturing methods and so on do not follow from our theoretical perspectives, but they still need to be considered and incorporated in order to allow an understanding of how marketing processes evolve. Inclusion of new phenomena can challenge existing theories, can make us more aware of their limitations and provide the basis for amending them. For example, the development of the Internet and e-marketing challenged the existing theories of attention, advertising effects and market information, as well as methods for measuring advertising reach, frequency and effectiveness.

Marketing as a management field will, however, not only be judged for its descriptive and predictive power but also for its ability to provide managerially relevant information, tools and conceptual guidelines. A third requirement, therefore, is that the attentional field of marketing researchers should at least cover phenomena considered important for marketing decisions (see Bass and Wind, 1995).

It appears reasonable that these different attentional fields may not entirely overlap. Malhotra et al. (1999, p. 178) point out that a certain gap exists between academic and commercial market research and that this gap ought to be bridged. Such a gap can exist for several reasons. One potential reason for this is that commercial market researchers are not aware of the advancement of tools used by academics. It is, however, also conceivable that the information needs perceived by practitioners call for methods and approaches other than the ones utilized in academic research.

Marketing researchers may to some extent be concerned with issues that are not considered important or relevant by practitioners who from their side may use concepts and procedures that appear inappropriate by academics. Marketing researchers for their part may be occupied with phenomena that are uninteresting from the point of view of more theoretically oriented colleagues. The constituencies of marketing are however not restricted to marketers but also include consumer welfare organizations, legislators and bureaucrats (cf.; Hirschman, 1991, p. 4; Wells, 1993). Marketing research has thus also been criticized for not addressing the critical aspects of marketing practice. Consequently, the discipline does not provide sufficient information for regulation of undesired marketing practice. Hirschman (1991), citing homelessness, drug addiction, credit card abuse and alcoholism, asks, «How much more could we accomplish if we turn even a portion of our talents toward ameliorating the dark side of consumer behavior?» (p. 4).

Theoretical adequacy

The second issue is whether marketing research is theoretically adequate. In other words, do the theories we develop and apply in fact explain and predict the phenomena they address? For example, does attitude theory account for behaviour and can satisfaction models predict loyalty? Do the theories we use help us attend to phenomena in need of explanation or do the theories we employ cause us to ignore important issues? A «third issue» is whether our theoretical frameworks lend themselves to empirical, testing or whether they somehow are too complex (e.g., criticisms of the Howard and Sheth model) or too imprecise to allow testing. A fourth aspect relates to the normative--descriptive dimension (Hunt, 1991). Whereas marketing theories may be adequate in the sense of providing reasonable explanation and precise predictions, they may not be considered particularly useful from a practitioner's point of view.

Methodological adequacy

The third criterion focuses on the methodological adequacy of the discipline. Does, for example, the structure of available methods match the sophistication of the theoretical models we put to test? Prior to the advent of structural equation modelling there was a mismatch between the complexities of consumer behaviour models, such as the Howard and Sheth (1969) model, and such techniques as multiple regression and MANOVA that were available for «testing» such theories (see, e.g., Farley and Ring, 1970; Lehman et al., 1974) that might have hampered further development of complex models.

In addition, the methodological toolbox should ideally be flexible and sensitive enough to allow us to detect phenomena and events not accounted for or emphasized in current theories. A third subcriterion is whether our methodological approaches help improve decision making. Do satisfaction surveys offer prescriptive advice to managers? Does conjoint analysis aid in the new product development process? Do positioning studies inform the choice of a marketing mix?

A summary of the criteria for addressing the adequacy of research in marketing is presented in Table 1.

Types of criticism against marketing

Criticisms of marketing are abundant; this is not surprising, because any new contribution by necessity points to a gap in our understanding or to an inadequacy as its raison d'etre. We need look no further than the 1999 millennium issue of theJournal of the Academy of Marketing Science or the 1999 special issue of theJournal of Marketing for a plethora of criticisms of the current state of the marketing discipline. Table 2 provides an incomplete list of recent criticisms. The list of things marketing needs to consider (phenomena addressed) is lengthy, including calls for more research in a variety of areas and issues and suggestions that too much attention has been paid to other phenomena (presumably at the expense of devoting attention to other, more relevant or important phenomena).

As noted in Table 2 the relevance of marketing research has also been questioned. In a recent discussion of the «relevance and rigor» debate, Varadarajan (2003, p. 369) «... questions and concerns regarding the relevance of scholarly research in marketing persist.» Similarly, specific issues regarding the theoretical and methodological adequacy of research in marketing are also detailed in Table 2. It is interesting to note that of the criticisms in Table 2 few claim that marketing is blatantly wrong, but that it is too narrow in some respects.

In sum, it appears that marketing is criticized more for what it does not do than for what it does.

The interaction of the four domains

Our thesis is that the four domains - Brinberg and McGrath's substantive, theoretical and methodological domains along with the managerial domain - interact to bring about changes in the nature of marketing as an academic discipline; this does not imply that the interdependence of the four domains necessarily is due to «causal» influences between the various domains. The use of qualitative, «Verstehen»-like approaches will typically be implied by a given ontological stance (e.g., postmodernism) in very mach the same manner that experimental approaches may be dictated by a behaviourist or operationist point of departure in the theoretical domain. Our choice in one domain may thus not be «caused» by our choices in other domains, but will frequently follow as a logical consequence. Fundamental assumptions about how consumers «are» (as an aspect of the theoretical domain) will typically condition our approach to describe them (as an aspect of the methodological domain).

As shown in Figure 1, we suggest that each domain influences and, in turn, is influenced by each of the other domains. For example, we suggest that advances in the methodological domain affect developments in the theoretic domain and, conversely, that theory development informs and affects our methods. Thus, each of the 6 numbered paths in Figure 1 represents 2 relationships (1 each way between pairs of constructs, giving a true total of 12). In addition, we acknowledge the impact of environmental changes on the substantive domain, giving a total of 13 «paths» for development and change in the marketing discipline. It is these interrelationships (not the existence of the domains themselves) that are key. We will briefly discuss each in turn and attempt to provide examples.

Attentional adequacy and decision relevanceOur attention should allow us to address and register events, states and processes entailed in our theories.
Our attention should allow us to address and register events and states and pertain to marketing phenomena even when they are not implied by our theories.
Our attention should cover phenomena that are considered important by decision makers or that might change their decisions, tactics and strategies.
Theoretical adequacyDo theories allow explanation and prediction of the phenomena they address?
Theories should allow us to attend to and understand phenomena worth explaining and predicting.
Theories should improve decision making.
Theories should lend themselves to empirical tests.
Methodological adequacyThe methods we apply should allow us to test complex explanatory models.
Methods should allow us to address and describe new phenomena worth addressing.
Methods should provide managerially relevant information.

Table 1 Criteria for judging the adequacy of marketing research

Relationship 1

We can view Relationship 1 from an idealist (or relativist) perspective (in which theories constitute the phenomena) and from a realist perspective where phenomena cause theories to emerge. Manfred Bruhn (2003), in line with a realist perspective on the link between theories and phenomena, suggests that the development of relationship marketing occurred in response to changes in the marketplace in the 1970s and 1980s: «Changes in the economic and competitive framework in the past few years have led to a new focus in marketing. For instance, today's marketers are faced with a marketplace which is enormously more complex, while in earlier stages marketing focused rather on mass marketing principles and 'making the sale'» (p. 1). Similarly, «quality», «attitudes» and «satisfaction» probably existed in a real sense even before the theories addressing these concepts were developed. From a realist perspective, the observation of a robust positive association between market share and profit-ability also called for explanation (see Varadarajan and Jayachandran, 1999, p. 134) and gave rise to four different types of explanation: the quality explanation, the market power explanation, the efficiency explanation and the third-factor explanation.

Types of criticismReference/Example

Attentional inadequacyDoes the phenomena addressed coincide with what should be within the boundaries of the discipline's subject matter?

Managerial irrelevanceDoes the research provide support and direction to decisions that affect marketing phenomena?

Theoretical inadequacyDoes the theoretical framework adequately describe and predict the phenomena it purports to account for?

Methodological inadequacyDoes the methodological approach allow a proper test of the theories and an adequate description of the phenomena that are investigated?

Attentional inadequacyDoes the phenomena addressed coincide with what should be within the boundaries of the discipline's subject matter?
  • Too much attention on free-standing domestic companies (Kinnear, 1999).
  • Shortage of studies addressing cross-functional processes, decision-making mechanisms, organizational structures and cross-functional business teams (Deshpande, 1999).
  • Over-focus on dealers in dyadic relationships (Malhotra et al., 1999).
  • Marketing research has left it to other disciplines, such as organizational ecology and evolutionary economics, to pursue questions of market dynamics and diversity (Montgomery, 1995; Day and Montgomery, 1999).
  • Fixation with brand as the unit of analysis in marketing strategy (Varadarajan and Jayachandran, 1999).
  • Most choice models in marketing assume that the fundamental unit of analysis is the brand. In practice, however, many more of the decisions occur at the level of the stock-keeping unit (SKU) (Malhotra et al., 1999, p. 165).
  • Lack of competitive focus in marketing strategy (Varadarajan and Jayachandran, 1999).
  • Lack of international orientation in marketing strategy (Varadarajan and Jayachandran, 1999).
  • Lack of focus on tacit collusion in marketing strategy (Varadarajan and Jayachandran, 1999).
  • Limited focus on strategy-making processes and what mental maps guide strategy formulation processes (Varadarajan and Jayachandran, 1999).
  • Little focus on forces driving product failures (Malhotra et al., 1999, p. 168).
  • Lack of focus on consumers' participation in production processes and artificial separation of production and consumption (Firat and Venkatesh, 1995).
  • Consumer behaviour viewed as response behaviour (not goal-directed behaviour) in which purchase is only one element (e.g., Bagozzi and Dholakia, 1999).
  • Little interest in poverty, misery and violence (Firat and Venkatesh, 1995).
  • Lack of focus on symbolic aspects of consumption.
  • Application of Bourdieu's theory of cultural capital has been applied to explain consumption of objects and not (as it should) to understand consumption practices (Holt, 1998).
  • Too much focus on individual salespeople at the expense of sales-teams (Malhotra et al., 1999, p. 170).
Managerial irrelevanceDoes the research provide support and direction to decisions that affect marketing phenomena?
  • Marketing research deals too much with tactical, rather than strategic problems (Deshpande, 1999).
  • Executives need guidance that is not provided by marketing researchers with respect to cross-border, cross-disciplinary issues (Kinnear, 1999).
Theoretical inadequacyDoes the theoretical framework adequately describe and predict the phenomena it purports to account for?
  • Lack of empirical evidence that satisfaction accounts for repurchase behaviour (Mittal and Kamakura, 2001).Lack of theoretical base for marketing strategy (Varadarajan and Jayachandran, 1999).
  • Researchers erroneously assume that alliances are formed by individual firms that evaluate alternative courses of action and fail to acknowledge that actions of firms are embedded in social contexts (Varadarajan and Jayachandran, 1999).
  • Marketing research must be grounded in theory. Theory enables us to meaningfully interpret and integrate the findings with previous research. Plausible theories of how consumers actively and passively search for information have not received clue attention and, hence, our understanding of this phenomenon is lacking (Malhotra et al., 1999, p. 177).
  • Attitude research.
  • Personality research.
  • We have theories that account for the behaviour of humans, not of products and markets.
Methodological inadequacyDoes the methodological approach allow a proper test of the theories and an adequate description of the phenomena that are investigated?
  • Use of student samples.
  • Use of homogeneous samples (e.g., Oliver and Sarbo, 1988).
  • Lack of use of qualitative research (Gummesson, 2001).
  • Scarcity of empirical studies in marketing strategy (Varadarajan and Jayachandran, 1999).
  • Techniques that marketing researchers apply are too adaptive and not sufficiently proactive for evaluating investments in future technologies (Achrol and Kotler, 1999).
  • Marketing will need stronger support by anecdotes, rather than fully developed studies (Kinnear, 1999).
  • «The New World» views are supported by anecdotes, rather than fully developed studies (Kinnear, 1999).
  • Malhotra (1988) cautions researchers as to LISREL's analytical appropriateness (Malhotra et al., 1999, p. 168).
  • Unjustified use of estimation methods in regression analysis (Malhotra et al., 1999, p. 17S).
  • Few longitudinal studies (Malhotra et al., 1999, p. 168).
  • Few examples of using multiple samples for assessing external validity (Malhotra et al., 1999, p. 168).
  • Regrettably, integrating technology-based advancements in measurements to natural settings remains elusive (Malhotra et al., 1999, p. 162).
  • Over-reliance on laboratory experiments to study effects of brand extensions (Malhotra et at., 1999, p. 164).
  • Over-reliance on cross-sectional data (Malhotra, 1988; Malhotra et al., 1999, p. 165).
  • Over-reliance on non-experimental field data to examine causal relationships (Malhotra et al., 1999, p. 170).
  • Due to underutilization of existing theory, our understanding of several substantive areas is limited, despite numerous studies (Malhotra et al., 1999, p. 177).
  • Need for development of procedures for visual presentation of data, so that marketing research may be more clearly communicated, especially to managers (Holbrook, 1997; Novak, 1995; Malhotra et al., 1999, p. 175).

Table 2 Types of criticism raised against marketing research (Continued)


Figur 1 Conceptualization of a discipline as consisting of four interrelated domains.

These are examples of theory building in marketing which starts with observation. However, we view the substantive phenomena that we investigate through the lens of extant theories and conceptualizations. Wilkie and Moore (1999, p. 123), citing Sirgy et al. (1982), suggest, «economists defined issues to fit their research forms, as did psychologists, sociologists, ecologists, and political scientists.» According to Gigerenzer (1991), cogniti!e phenomena were sup-pressed as research topics in psychology in the early 20th century due to the allied forces of behaviourism and operationism. In their criticism of the literature that addresses the «groupthink phenomenon», Aldag and Fuller (1993) claim that the groupthink model suggested by Janis (1971) has functioned as a heuristic device that conditions researchers to restrict their attention to negative decision outcomes (failures), to look for a limited set of explanatory variables and to serve as a biased tool for interpreting research findings.

Marketers are not immune from this type of criticism (see Wilkie and Moore, 1999, p. 139). More than 20 years ago, Oliver (1980, p. 207) made the observation that models borrowed from communications theory, social comparison theory and facet research had been «enlightening but epistemologically inefficient explanations of satisfaction perceptions.» The substantive areas we choose to investigate are often the ones where theories exist, definitions have been agreed on and research approaches have been established. Speaking to this difficulty of doing research in a «new» domain, Kinnear (1999, p. 113) states, «At the most elementary level, it is almost impossible to do high quality research that builds the state of knowledge without a set of agreed upon definitions. The emperor indeed may have no clothes, but because there is no definition of what clothes are, no one will know.» Until we have the vocabulary and a guiding theory, new phenomena in the substantive domain may not receive adequate attention (while more developed areas of research may be investigated beyond the point of diminishing returns).

Similarly, much research in marketing has been «predictor-oriented» -- a theory or construct (often borrowed from another discipline) investigated to see where in marketing it might apply or what marketing phenomena it might explain (cf., James et al., 1982). The starting point is an interesting theory or construct, not a phenomenon to be explained. We have all seen articles of the form, «An interesting theory X has been developed in discipline Y. This study describes theory X in the context of marketing phenomenon Z.» The theory drives the choice of subject matter.

While the availability of theory may affect our choice of which phenomena to investigate, the «normal science» view is that phenomena are observed and theories are developed to explain them. «We have observed phenomenon Z. How can we predict it?» As new phenomena are observed in the substantive domain, calls for new research (and new theories) are issued. As noted previously, the marketing discipline is often criticized on its choice of phenomena for investigation: «Marketing research does not reflect the top research problems facing the marketing discipline» (Deshpande, 1999, p. 95). «In a world with few boundaries among functions, channels, cultures, and countries; with networks of companies and e-commerce facilitating consumer and distribution relationships, and with new modes of supply chain and distribution channel dynamics, the challenges for excellence in scholarship abound. The old mode of research that focuses on a functional marketing issue in a free-standing domestic company simply does not deal meaningfully with the critical issues of the new world of marketing» (Kinnear, 1999, p. 113). For a further discussion of phenomena in need of investigation see Day and Montgomery (1999).

Relationship 2

We suggest that research tools restrict and condition the phenomena to which we pay attention and, conversely, that we develop tools and methods in response to the need to investigate particular phenomena; that is, theory drives and influences developments in the methodological domain, and, perhaps not as obviously, developments in the methodological domain allow for new theorizing in the domain of theory. These relationships are discussed below.

Ideally, the substantive phenomena we need to address would drive the development of tools and methods needed to address them. Indeed, the need to determine advertising effects leads to the development of split cable consumer scanner panels, the need to model choice behaviour leads to the development of the multinomial logit model, the need to assess brand loyalty/switching behaviour leads to the collection and analysis of scanner data, etc. However, it can be argued that the availability of research tools and methods affects the choice of substantive issues to which we devote our attention, both positively (we investigate areas where we have appropriate tools and methods) and negatively (we avoid addressing otherwise important phenomena when available tools and methods are inadequate).

Empirical tools, such as exploratory techniques like cluster analyses and common factor analysis, can help us see «phenomena» and patterns we were not aware of -- with or without theories. Cross-national, cross-cultural research has been given too little attention because of the difficulties associated with such research. Research in the «network» and «relationship» marketing areas has been sparse, given the importance of the topic, perhaps because of the lack of easily employed methods for their study. Many of the criticisms of marketing's «attentional» inadequacies in Table 12.1 can be traced to difficulties in conducting research in some areas, often due to underdeveloped, difficult or non-existent methodologies rather than a lack of awareness or a sense of their importance. The other side of the coin is that we tend to investigate areas where we have well-developed methods and tools. Like the drunk under a street lamp searching for his missing keys, we look in well-lit areas.

Relationship 3

We posit an interplay between the theoretical domain and the methodological domain. The «traditional» sequence would have it that tools and methods are developed in response to the demands of the theories we need to investigate. Theories suggest data necessary for their testing, along with methods for their assessment. Examples are abundant and will be familiar to most readers. For example, Kelly's repertory grid seems to be a method derived directly from the associated theory, and the IDB (information display board) as a research tool follows directly from theories of consumer information processing. Moreover, theoretical models can also serve as interpretive schemata or tools that condition the way we extract information from data, blurring the distinction between theory on one side and analysis on the other (see Aldag and Fuller, 1993).

A less obvious relationship seems to be the extent to which our tools and methods affect our theories. Gigerenzer (1991) documents and provides examples of this relationship, the most compelling being how theories of the brain and cognitive processing are inspired by analytical methods (such as ANOVA, factor analysis, hypothesis testing, etc.) and the computer (inputs, outputs, encoding, decoding, memory allocation, bits and bytes, multiprocessing, CPU, short-term storage, long-term storage, signal-to-noise ratio, etc.) resulting in the conception of mind as an intuitive statistician or as a «human computer». Indeed, «brain-as-computer» has become such a well-established metaphor that it is difficult to imagine how the brain was conceived of prior to the advent of the computer and its associated concepts and terminology. Similarly, as researchers became trained in statistics, theories (in psychology, in particular) came to incorporate statistical elements (the «brain-as-intuitive-statistician» -- e.g., the work of Kahneman and Tversky, 1979).

The field of structural equation modelling (SEM) was developed out of the need to test complex theories containing latent constructs, yet it seems apparent that the SEM approach to theory testing has affected theorizing in marketing. Prior to the 1980s few marketing theories were multivariate, with multiple endogenous variables. In recent years many marketing theories have been expressed in the form of a «boxes-and-arrows» diagram. Mediated relationships are explicitly theorized, and relatively complex theories have become the norm. However, until very recently, moderated relationships (interaction terms) have been difficult to model in traditional SEM software (e.g., LISREL). As a result, few theories have postulated interactions and moderated relationships.

This is in direct contrast with theories in areas where the experimental tradition, using ANOVA as the primary analysis tool, has held sway. In those areas, interaction hypotheses (easily dealt with in ANOVA) 2 are abundant and theories containing multiple dependent variables and mediated relationships are scarce.

In marketing, it can be argued that our theories of market segmentation are informed by the structure and logic of cluster analysis, that our theories of «product» both inform and are informed by the structure of conjoint analysis and that our theories of positioning are closely interwoven with the maps produced by popular multidimensional scaling algorithms.

Relationship 4

This relationship deals with the interplay between the substantive domain and the domain of managerial decisions. It is easy to see that changes in the substantive domain affect the decisions managers need to make. The rapid growth of the Internet calls for many decisions on the part of the manager that would not have been necessary without this technology. The availability of CRM (customer relationship management) systems likewise necessitates new decisions. Should an investment be made? Flow should the system be implemented? In general, changes in the substantive domain are the primary driving force behind the scope and configuration of the managerial decision process.

It may also be the case that the decision domain affects the substantive domain. Managers do not simply react to the environment but, instead, are often proactive in changing it, through their advertising messages, lobbying efforts in the legal and political environment or through the introduction of new technology. Similarly, the decisions managers make can have an impact on the phenomena in the substantive domain. Managers' collective decisions to, for example, use email as a direct marketing tool have brought about public policy concerns regarding the impact of «spam» and methods to control it. Managers' decisions to promote, for example, cigarettes to young people raise issues that need to be addressed in the substantive domain.

Relationship 5

In a fashion similar to Relationships 1 and 2, methods and managerial decisions also seem to be interrelated. The development of conjoint analysis was at least in part driven by product developers' need for a research tool to aid in the product design decision. Although applied in many fields, substantial work in the area of cluster analysis resulted from managers' needs to make segmentation decisions. Managers have to make decisions, and tools (along with theories) that inform those decisions have been and are being developed.

At the same time, we posit that the availability of research tools, such as ServQual, satisfaction scales, conjoint analysis and MDS, influence the way managers think and act. Managers are explicitly addressing some decision areas due to the availability of tools for their investigation. The availability (or absence) of tools affects the form (if not the content) of managerial decisions. We suggest that managers' information needs drive the development of tools and methods, while, simultaneously, managers' perceptions of their decisions and, thus, their information needs are affected by the tools and methods at their disposal.

Relationship 6

Theories offer managers a mechanism for construing their environment. Attitude theory, attribution theory, satisfaction theory, relationship marketing, transaction cost theory, etc. have shaped the mental maps of managers and their theories-inuse. At the same time, managers' questions have guided theorizing in marketing. Much theory in marketing has been developed to answer managers' questions. When managers ask «Why?» theories are developed or amended to answer the question.

The willingness to serve practitioners' theoretical needs has not however necessarily been beneficial for marketing as an academic discipline with respect to theory construction, methodological developments and the content of the substantive domain. The economic interests of the firm can easily explain marketers' natural obsession with those factors and functions that lead to and facilitate a transaction. Marketing scholars on their side appear for a long time to have accepted this rather myopic view which has led to what Vargo and Lusch (forthcoming) label the dominant «goods-centred logic» in marketing where value or utility is viewed as embedded in a tangible good as opposed to an emerging «service-centred logic» that acknowledges the role of the customer in the value creation process.

In service-centred logic, offerings simply facilitate value creation and the ultimate values from a customer perspective are co-created in concert with the customer. Service-centred-logic meanwhile calls for conceptualization of the buyer as an actor rather than a responder and suggests a more phenomenological, experiential view of consumption (Holbrook and Hirschman, 1982). This shift in world view has obvious implications for theory construction and methodological developments.

Relationship 6 may well warrant further investigation. For example, what do we know about the theories underlying marketing practitioners' decisions? It would be beneficial to make systematic comparisons of Argyris and SchOn's (1974) «theories in use» (i.e., those theories that appear to guide marketing practice), «espoused theories» (i.e., theories that practitioners may accept, but not necessarily use) and «scientific theories» (i.e., theories developed by marketing academics).

Since research-based journals may not be fully comprehensible for non-academic audiences it is hoped that textbooks could play an important role of disseminating research findings and theories to students and practitioners outside the research community. However, recent cases (Armstrong and Schultz, 1993; Rothfeld, 2000) suggest that we cannot be sure this is happening as textbooks have been demonstrated to «institutionalize theoretical approaches from previous years» (Rotfeld, 2000) which can lead to «(t)he dogmatic retention of unsupported and outdated theories, as well as the implicit endorsement and propagation of these in journal articles.» Armstrong and Schultz (1993) found 566 normative statements pertaining to pricing, communications, and product or distribution decisions that were not supported by empirical evidence. With an in-creasing amount of research and an accompanying need to summarize research findings, the risk of oversimplifying and misrepresenting research may become ever larger.


Gaps will most likely occur if the discipline gets too «compartmentalized» and specialized. In other words, if theoreticians only theorize, if market research institutes only carry out market research and market surveillance, if methodologists only develop methods and if practitioners only make decisions, then the desired «natdral flow» between the various domains will not take place.


Many of the criticisms of marketing scholarship fall under our Relationship 1: the linkage between the realms of theory and the substantive domain. Perhaps abductive research (where peculiarities and puzzles are observed and lead to the search for theories and explanatory models) presents a possible partial remedy. The work of Kahnemann and Tversky on decision making is perhaps an example of observed anomalies resulting in a rich behavioural theory. Theories in this paradigm become tools to extract information from reality.

Along with abductive research, marketing knowledge could similarly be advanced through greater incorporation oftheory triangulation -- bringing alternative theories to bear on the job of explaining a given puzzle. Testing «theory A versus theory B» can be much more productive and informative than testing «theory A versus the (usually weak) null hypothesis». Paul Meehl (1990) points out the lack of progress made in psychology and attributes many of the problems to testing against a weak null hypothesis (i.e., no relationship, no difference) wherein large sample sizes work «in favour» of the theory. He suggests the examination of competing theories as a preferred practice.

Another recommendation is to use an expanded set of methodologies to capture the richness and complexity of a given phenomenon, as suggested by Aldag and Fuller (1993) in their criticism of the groupthink tradition.

We believe marketing would benefitfrom multidisciplinary projects addressing different phenomena, using different theories, but using the same research context (to avoid narrow-minded «contextless» research). One example is a project on tourism at the Norwegian School of Economics and Business Administration sponsored by the Royal Ministry of Industries. In this project the topics pursued ranged from guest (customer) satisfaction, service quality, organizational culture and economic performance measures, like capacity utilization, profit margins and ROI. By using the same hotels and allowing different researchers to use a number of alternative theoretical perspectives, like satisfaction theory, quality management, organization behaviour and industrial economics, a comprehensive, mtiltidisciplinary, yet context-specific understanding of the interplay between these different areas was possible (see Troye and 4 gaard, 1999). .

A challenge is to establishbetter ways to align marketing practice and marketing research. This challenge raises two issues: first, how can marketing research gain from better insight into marketing practice and vice versa? and, second, how can practice benefit from marketing as an academic discipline? The first issue was addressed by Wilson and Ghingold (1980) who suggested that theories could be built from practice, using what they referred to as a «theory-in-use approach» in which an attempt is made to extract the theories that are being used by practitioners: «Thus, the theory-in-use approach calls for the study of the methods that the common man, the marketing practitioner, [employs] in attempting to construct reliable and valid statements about the world ... beyond the methods posited by scientific/academic investigators» (Wilson and Ghingold, 1980, p. 237).

The second issue -- how marketing practice can make better use of marketing research -- deals partly with how to popularize theories, research findings and methods in order to bridge many of the gaps involving the managerial domain. The findings reported by Rotfelcl (2000) and Armstrong and Schultz (1993), cited earlier, suggest that care should be taken to avoid the «textbook effect» -- distorting theories and findings while ignoring valid and potentially useful research contributions.Industry-academic research teams and projects is one option that can alleviate the practitioner-academic communication gap: by allowing such teams to be cross-disciplinary, knowledge transfer between industry and academic institutions and between various disciplines would be facilitated and should lead to cross-fertilization.

  • 1: Similarly, Kuhn (1989, p. 357) briefly mentions five criteria for adequacy: (1) accuracy (i.e., consequences deducible from a theory should be in demonstrated agreement with the results of existing experiments and observations), (2) consistency, (3) scope, (4) simplicity and (5) fruitfulness (of new research findings).
  • 2: It is also conceivable that the ease with which interaction effects are tested in most ANOVA programs makes it tempting to formulate hypothesis involving the interaction term and to provide a theoretical rationale for it, even in situations where there may be no a priori ground to assume such effects.
  • Achrol, R.S. and Kotler, P. (1999). Marketing in the network economy. Journal of Marketing, 63(Special issue): 146-463.
  • Achrol, R.S. and Stern, L.W. (1988). Environmental determinants of decision making uncertainty in marketing channels. Journal of Marketing Research, 25(February): 36-50.
  • Aldag, R.J. and Fuller, S.R. (1993). Beyond fiasco: A reappraisal of the groupthink phenomenon and a new model of group decisions processes. Psychological Bulletin, 113(3): 533--552.
  • Amistrong, J.S. and Schultz, R. (1993). Principles involving marketing policies: An empirical assessment. Marketing Letters, 4(3): 253--265.
  • Argyris, C. and Schon, D. (1974). Theory in Practice. San Fransisco: Jossey-Bass. Bagozzi, R.P. and Dholakia, U. (1999). Goal setting and goal striving in consumer behavior. Journal of Marketing, 63(Special issue): 19--32.
  • Bass, F. and Wind, Y. (1995). Special issue on empirical generalizations in marketing. Marketing Science, 14(3): 1--236.
  • Brinberg, D. and McGrath, J.E. (1988). Validity and the Research Process. Newbury Park, CA: Sage Publications.
  • Bruhn, M. (2003). Relationship Marketing: Management of Customer Relationships. London: Prentice Hall.
  • Day, G.S. and Montgomery, D.B. (1999). Charting new directions for marketing. Journal of Marketing, 63(Special issue): 3-13.
  • Deshpande, R. (1999). Introduction: «Foreseeing» marketing. Journal of Marketing, 63(Special issue): 164--167.
  • Farley, J.U. and Ring, L.W. (1970). An empirical test of the Howard--Sheth model of buyer behavior. Journal of Marketing Research, 7(November): 427--438.
  • Firat, A.F. and Venkatesh, A. (1995). Liberatory postmodernism and the reenchantment of consumption. Journal of Consumer- Research, 22(December): 239--268.
  • Gigerenzer, G. (1991). From tools to theories: A heuristic of discovery in cognitive psychology. Psychological Review, 98(2): 1--14.
  • Gummesson, E. (2001). Are current research approaches leading us astray? Marketing Theory, 1(1): 27--48.
  • Hirschman, E.C. (1991). Secular mortality and the dark side of consumer behavior. In: R. Holman and M. Soloman (eds), Advances in Consumer Research (Vol. 18, pp. 145-156). Provo, UT: Association for Consumer Research.
  • Holbrook, M.B. (1997). Visual representation of data: Stereographic visual displays and the three-dimensional communication of findings in marketing research. Journal of Marketing Research, 34(November): 526-536.
  • Holbrook, M.B. and Hirschman, E.C. (1982). The experiential aspects of consumption: Consumer fantasies, feelings, and fun. Journal of Consumer- Research, 9(September): 132-140.
  • Holt, D.B. (1998). Does cultural *capital structure American consumption? Journal of Consumer Research, 25(June): 1--25.
  • Howard, J.A. and Sheth, J.N. (1969). The Theory of Buyer Behavior New York: John Wiley & Sons.
  • Hunt, S.D. (1991). Modern Marketing Theory: Critical Issues in the Philosophy of Marketing Science. Cincinnati, OH: South-Western Publishing.
  • James, L.R., Mulaik, S.A. and Brett, J.M. (1982). Carnal Analysis. Beverly Hills, CA, Sage Publications.
  • Janis, I. (1971). Groupthink. Psychology Today, 5(November): 43-46 and 74-76. Kahneman, D. and Tversky, A. (1979). Prospect theory: An analysis of decisions under risk. Econometrica, 47: 313-327.
  • Kinnear, T.C. (1999). Introduction: A perspective on how firms relate to their markets. Journal of Marketing, 63(Special issue): 112-114.
  • Kuhn, T. (1989). Objectivity, value judgment, and theory choice. In: B.A. Brody and R.E. Grandy (eds), Readings in the Philosophy of Science. Englewood Cliffs, NJ: Prentice Hall.
  • Laudan, L. (1977). Progress and Its Problems. Berkeley, CA: University of California Press. Laudan, L. (1989). From theories to research traditions. In: B.A. Brody and R.E. Grandy (eds), Readings in the Philosophy of Science (pp. 368-378). Englewood Cliffs, NJ: Prentice Hall.
  • Lehman, D.R., O'Brien, T.V., Farley, J. and Howard, J. (1974). Some empirical contributions to buyer behavior theory. Journal of Consumer Research, 1(December): 48-55. Malhotra, N. (1988). Self-concept and product choice: An integrated perspective. Journal of Economic Psychology, 9(1): 1-28.
  • Malhotra, N., Peterson, M. and Kleiser, S.B. (1999). Marketing research: A state-of-the-art review and directions for the twenty-first century. Journal of the Academy of Marketing Science, 27(Spring): 160-183.
  • Meehl, P.E. (1990). Appraising and amending theories: The strategy of Lakatosian defense and two principles that warrant it. Psychological Inquiry, 1(2): 108-141.
  • Mittal, V. and Kamakura, W. (2001). Satisfaction, repurchase intent, and purchase behavior: Investigating the moderating role of customer characteristics. Journal of Marketing Research, 38(1), February: 131-142.
  • Montgomery, C.A. (1995). Resource-based and Evolutionary Theories of the Firm. Norwell, MA: Kluwer.
  • Novak, T.P. (1995) MANOVAMAP: Graphical representation of MANOVA in marketing research. Journal of Marketing Research, 32(August): 357--374.
  • Oliver, R.L. (1980). Theoretical bases of consumer satisfaction research: Review, critique, and future direction. In: C.W. Lamb and P.M. Dunne (eds), Theoretical Developments in Marketing (pp. 206-210). Chicago: American Marketing Association.
  • Oliver, R.L. (1999). Whence consumer loyalty. Journal of Marketing, 63(Special issue): 33-44.
  • Oliver, R.L. and DeSarbo, W.F. (1988). Response determinants in satisfaction judgements. Journal of Consumer Research, 14(March): 495-507.
  • Rotfeld, H.J. (2000). The textbook effect: Conventional wisdom, myth, and error in marketing. Journal of Marketing, 64(2).
  • Sirgy, M. Joseph, A., Samli, C. and Meadow, H.L. (1982). The interface between quality of life and marketing: -A theoretical framework. Journal of Marketing & Public Policy, 1(1): 69-84.
  • Troye, S.V. and 0gaard, T. (1999). The many effects of service: An empirical study of service and product quality in the hospitality industry. In: B. Edvardsson and A. Gustafsson (eds), Nordic School of Quality Management (pp. 362-392). Lund, Sweden: Studentlitteratur.
  • Varadarajan, P.R. (2003). From the editor: Musings on relevance and rigor of scholarly research in marketing. Journal of the Academy of Marketing Science, 31(Fall): 368-376.
  • Varadarajan, P.R. and Jayachandran, S. (1999). Marketing strategy: An assessment of the state of the field and outlook. Journal of the Academy of Marketing Science, 27(Spring): 120-143.
  • Vargo, S.L. and Lusch, R.F. (2004). Evolving the new dominant logic for marketing. Journal of Marketing, 680): 1-17.
  • Wells, W.D. (1993). Discovery-oriented consumer research. Journal of Consumer Research, 19(March): 489-504.
  • Wilkie, W.L. and Moore, E.S. (1999). Marketing's contribution to society. Journal of Marketing, 63(Special issue), 198-218.
  • Wilkie, W.L. and Moore, E.S. (2003). Scholarly research in marketing: Exploring the «four eras» of thought development. journal of Public Policy and Marketing, 22(2): 116-146.
  • Wilson, D.T and Ghingold, M. (1980). Building theory from practice: A theory-in-use approach. In: C.W. Lamb and P.M. Dunne ((ds), Theoretical Developments in Marketing (pp. 236-239). Chicago: American Marketing Association.

Econa er foreningen for høyt utdannede innen økonomi og administrasjon. Er du ikke medlem?
Sjekk medlemstilbudene og meld deg inn i dag.

© Econas Informasjonsservice AS, Rosenkrantz' gate 22 Postboks 1869 Vika N-0124 OSLO
E-post:  Telefon: 22 82 80 00.  Org. nr 937 747 187. ISSN 1500-0788.