Regularities, Rules and Consumer Behavior: Tangencies Between Positivist and Interpretive Approaches to Research

Eileen Fischer, York University
ABSTRACT - Positivist approaches to consumption behavior have as a goal the identification of regularities which have law-like, causal effect on the actions of individuals. Many interpretive approaches to studying consumption behavior have an analogous goal, the identification of rules, which are the tacitly understood norms and conventions underlying the behavior of individuals. Despite the profound disparities between positivist and interpretive science, the regularities/ rules which each seeks to identify constitute a point of tangency. This paper explores the relationship between positivist regularities and interpretive rules in consumer behavior and traces their implications for reconciling alternative routes to seeking knowledge.
[ to cite ]:
Eileen Fischer (1990) ,"Regularities, Rules and Consumer Behavior: Tangencies Between Positivist and Interpretive Approaches to Research", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 19-24.

Advances in Consumer Research Volume 17, 1990      Pages 19-24


Eileen Fischer, York University


Positivist approaches to consumption behavior have as a goal the identification of regularities which have law-like, causal effect on the actions of individuals. Many interpretive approaches to studying consumption behavior have an analogous goal, the identification of rules, which are the tacitly understood norms and conventions underlying the behavior of individuals. Despite the profound disparities between positivist and interpretive science, the regularities/ rules which each seeks to identify constitute a point of tangency. This paper explores the relationship between positivist regularities and interpretive rules in consumer behavior and traces their implications for reconciling alternative routes to seeking knowledge.

Compelling arguments have been mounted regarding the need to consider interpretive approaches as alternatives or supplements to positivist approaches to studying consumption behavior (for instance Anderson 1986; Hudson and Ozanne 1988; Peter and Olson 1983). The challenge now facing consumer researchers is how to employ one or more research approaches in a manner of practical value for the development of the field. Hudson and Ozanne (1988) have identified four alternative strategies for coping with the diversity of research approaches. Specifically, they suggest: (1) a supremacy alternative, in which researchers attempt to resolve which scientific approach is best; (2) a synthesis alternative, in which researchers merge interpretive and positivist approaches, ignoring differing underlying assumptions; (3) a dialectic alternative, in which multiple research approaches are applied with an appreciation for their divergent philosophical foundations, but with an aim to utilize competing insights within a single analysis in the hope that an entirely new form of understanding may be generated; or (4) a relativistic alternative, in which the differing approaches are all considered acceptable so long as they are evaluated by the standards appropriate to the method of knowledge production.

Those who opt for pursuit of the latter two of these alternatives are faced with the question of how research premised on one philosophical foundation may usefully and appropriately be integrated with research from a different philosophical perspective. It is clear that those who follow the dialectic route inherently require some means for integrating the findings of divergent research streams, but it may be less evident why those who adopt the relativistic approach should share this concern. Unless relativistic researchers operating within one research tradition are content to ignore findings relevant to their topic but generated within an alternative tradition, however, they must consider how the products generated by discrepant research approaches can be, in some sense, reconciled.

The objective of this paper is to explore one important way in which the foci of interpretive and positivist research are tangent, thereby suggesting a point of departure for researchers who attempt to draw on knowledge generated by divergent traditions. Specifically, the paper will examine the tangency between the causal regularities sought in positivist research and the hermeneutic rules sought in interpretive research. The benefits of recognizing these tangencies will be discussed. As well, limitations to the tangency between rules and regularities will be outlined. It is important to recognize that interpretive approaches are no more monolithic than positivist approaches: not all interpretive research can be said to seek interpretive rules, just as not all positivist research seeks causal regularities. Moreover, there are likely important and irreducible differences in the perspectives an interpretivist and a positivist will take on a seemingly identical phenomena: for instance, a fundamental difference in perspective occurs because interpretive researchers place much greater emphasis on the agency of actors whose behavior is studied. Given, however, that many forms of interpretive research (such as hermeneutic analysis, symbolic interactionism, and certain forms of phenomenological sociology and ethnomethodology) do generate insight into social rules, it is useful to seek tangencies between these research findings and the causal regularities which are sought in much positivist research. Even if the similarities between the perspectives taken on a phenomenon of interest is limited, it is useful to consider the nature and value of tangencies in alternative perspectives.


Although the research traditions of positivism and interpretivism are incommensurable, they are not necessarily incomparable or contradictory. Tracing the implications of incommensurability, and showing that they do not include incomparability or inconsistency, will help to clarify the extent to which a rapprochement between positivism and interpretivism is possible.

The incommensurability of positivism and interpretivism stems from the differences in axiology, ontology and epistemology which characterize the two approaches. A positivist axiology entails a commitment to explanation and is based on an ontological belief that a single, multifaceted reality exists, coupled with an epistemological belief that distinct, separable facets of the one reality can be known. An interpretive axiology entails a commitment to understanding and is based on an ontological belief that realities are mediated by the perception and comprehension of individuals, coupled with an epistemological belief that these realities can only be grasped partially, in both senses of the term (i.e. all understanding is incomplete and is conditioned by the researcher's own perspective (Packer 1985)).

Incommensurability of approaches means that "two groups of scientists see different things when they look at the same point and in the same direction" (Kuhn 1970, p. 150). That is, positivists and interpretivists may look upon the same phenomenon, but see different aspects of it. Further, they may construe the problem to be investigated in differing ways, and apply different standards to the evaluating research efforts (Bernstein 1985).

Incommensurability described in this way does not imply incompatibility. For positivist and interpretive research projects to be incompatible, the two projects would have to entail some logical contradiction. Such contradictions presuppose a shared framework of logic, a possibility upon which the incommensurability of positivism and interpretivism places severe constraints. While positivist and interpretive research programs may contain contradictions in some aspects, they are not totally contradictory simply because they are incommensurable.

In addition to being (at least partially) logically compatible, positivist and interpretive research projects may also be comparable. In fact as Anderson (1986) demonstrates, incommensurable research programs can be compared on numerous bases. Although Anderson is contrasting incommensurable research programs that share a common cognitive aim and can all be labelled positivist, his approach is equally valid for comparing interpretive and positivist approaches. Such comparisons, it must be recognized, do not require a common, fixed grid of characteristics which is universally applicable: some comparison is possible even if one research program has characteristics without any analogue in another research program.

Comparisons between certain facets of positivist and interpretive research projects are not merely appropriate: they may also be useful. In particular, it is valuable to consider comparable -though incommensurable -- components of positivist and interpretive research which have the potential to afford a practical linkage between projects grounded in the two research streams: the regularities sought in positivist research and the rules sought in interpretive research. Regularities and rules are analogous facets of positivism and interpretivism in that each represents the kind of knowledge sought under specific axiological commitments, as is explained below.


Chief among the axiological commitments of positivism is a determination to explain causal regularities which are believed to underlie human behavior. As described by contemporary philosophers of science, the form of argument used to assert causal regularities is characterized by the deductive-nomological model: if a number of conditions (C1, C2, etc.) are met, then any object (such as a person or system) that meets those conditions will have a certain property. That property is the outcome to be explained. For instance, consider the causal regularity posited in the relationship between a gift giver's ideal self concept and the type of gift chosen (Belk 1979). Given the task of choosing gifts for other people (the condition), gift-givers (the objects), choose gifts reflecting their own ideal self image more than their actual self image or their perceptions of the recipient (the outcome).

In positivist social science, perfect regularity is rarely if ever considered necessary for causal generalizations to be made. Probabilistic statements suffice as approximations of universal law-like statements in most instances. That is, causal regularities are often considered to exist in instances where the regularities do not hold in every instance, but do hold considerably more often than-not. Causal regularities, then, may be loose regularities, but nonetheless be considered both causal and regular (Braybrooke 1987).

In some instances, casual regularities in social science may also be transitory. That is, regularities may not extend over time, or between cultures. For instance, while much research has documented as a regularity that women are more "communally" oriented and men are more "agentically" oriented (Been 1974; Spence and Helmreich 1978), it is anticipated that it will become increasingly common for women and men to exhibit more equal degrees of communal and agentic orientations -- that is, for women and men to be relatively androgynous (see, for example, Rossi 1984). One way of explaining why a regularity may be transitory is to argue that the regularity was premised on conditions including the specific times and/or places in which the regularity was detected; the other option is to treat causal regularities as genuine even if transitory (Braybrooke 1987). The point is that positivist regularities in social science need not be immutable to a qualify as causal explanations of practical value.

One other important characteristic of causal regularities is that they may express laws that apply at the individual or group level. The examples cited above refer to regularities which apply at the individual level. Consider, for example, studies of discord or unity in household purchasing decisions (such as Qualls 1987). In such studies, regularities detected in decision making patterns may be postulated to depend partially on individual differences in gender role attitudes, but they apply at the level of the couple: they cannot be construed merely as a conjunction of regularities governing individuals.


In contrast with the axiological commitment of positivism to explaining causal regularities is the axiological commitment of interpretivism to understanding settled social rules. Social rules may be tacitly understood or explicitly stated, but are generally amenable to verbal expressions containing a descriptive "is" or a prescriptive "ought" implication. For instance, a settled social rule detected by Caplow (1984) amongst residents of Middletown in the late 1970's was that Christmas gifts between spouses should be more valuable than any other gifts given by husband or wife, but that the husband's gift to the wife may be more expensive than the wife's to the husband (p. 1313). While Caplow explicitly uses the term "rules" to refer to his findings, it should be noted that many interpretive researchers do not. Other labels, such as "norms," "patterns," or "conventions," are common in the methodological discussions of interpretive approaches. Nonetheless, many types of interpretive research seek similar "rule-like" findings.

Ryan and Bristor (1987), following von Wright's (1971) arguments, stress that there is a symbiotic relationship between causal regularities and interpretive rules because the forms of arguments offered in support of regularities and of rules are essentially similar. They characterize the form of reasoning followed in interpretive research as a "practical syllogism," which consists of a major premise that entails some end or goal the actor has, a minor premise that is the rule or relationship comprehended by the actor as a means to that end, and a conclusion which suggests that since the actor apprehends the relationship between certain actions and specific goals or ends, the individual forms the intention to perform the action. For instance, the major premise in the example above might be that the actor wishes not to threaten the integrity of the spousal relationship. The minor premise would be the rule understood regarding the scaling of monetary expenditure to the closeness of the relationship. The conclusion would be that the actor spends more on the spouse's gift than on that for any other recipient.

The similarity between the practical syllogism of interpretive research and the deductive-nomological model of positivist research is so striking that Ryan and Bristor assert "the practical syllogism makes contact with the reasoning procedures already employed by the scientific model and is methodologically continuous with it" (p. 193). Indeed, the similarity between the logic underlying positivist regularities and the reasoning underlying interpretive rules is striking. The tangencies between rules and regularities, however, extend well beyond the logic which informs the two approaches. Other important ways in which causal regularities are tangent to interpretive rules can be gleaned from further comparison between the two.

Interpretive rules, like causal regularities, may be argued to exist despite the fact that there is not perfect regularity or consistency observed in behaviors posited to be affected. One condition ascribing a settled social rule to a given social unit is that, for the most part, individuals in the unit conform to the rule. However, rules may exist without widespread evidence of conformity since agents may attempt unsuccessfully to conform to social rules (Giddens 1984): for instance, the gift giving rule cited above would not be invalidated observations that gifts were not exchanged between spouses in some families because of income constraints. Further, a rule may exist even when there are intentional lapses in conformity: evident of the existence and persistence of a rule arises w instances of non-conformity are decried and instances of conformity rewarded. So, for instance if an individual gave a friend of the opposite sex gift more expensive than that given to a spouse, persistence of the gift giving rule would be revealed by reactions of indignation on the part of the spouse.

As is true of causal regularities, interpretive rules may also be transitory over time or across geographic regions. It is a basic tenet of- the philosophy underlying interpretivism that rules exist primarily as they are comprehended by acto (though settled social rules may be codified into societal laws), and that these rules may or may n be shared by other individuals or groups. Rules a expected to vary across time and across groups, j as the laws reflecting settled social rules of one society at one point in time differ from the laws the same society at other points in time, or from laws of other societies. One reason rules might l transitory is that actors may come to conscious awareness of rules which underlie their behavior X choose to reject the rules or replace them. While interpretive philosophers argue about whether (or how much) conscious intention must be associate with actions presumed to be conditioned by rules they recognize that actors may be conscious of rules, and have considerable potential for voluntary control over their rule-following behavior (Giddens 1984).

A final important similarity between positivist regularities and interpretive rules is that both may be posited to exist as either "person fa or "group facts." While an interpretive finding apply to the actions of individuals, it has the status of a group fact if the individual is likely to be sanctioned by other social actors for failure to follow a rule or rewarded by some reciprocal action for conformity in following the rule. Many of the rules with which interpretivists are concerned (such as the gift giving rule cited above) pertain to group facts since scientific interest often centres on understanding behaviors which are characteristic more than one individual. Interpretive studies in marketing are particularly likely to focus on group facts because exchanges are focal behaviors, and exchanges characteristically involve understandings shared by more than one social actor.

Although settled social rules which illuminate group facts are arguably of greatest interest to interpretivists, person facts may also be important. Biographies of important historical figures or case studies of individuals considered aberrant by society can offer insights into a particular individual's unique understanding of social rules. These types of interpretive research can illustrate how unusual or unacceptable behaviors, premised on rules not shared by a broader society, may affect or reflect on more widely shared societal rules. So, for instance, a case study of a consumer with unique rules regarding the ownership of goods offers novel insights into the nature of materialism in modern society (McCracken 1988.)

The foregoing list of similarities between rules and regularities suggests that, notwithstanding their incommensurability, there may be greater comparability and consistency between positivism and interpretivism than is typically imagined. An even stronger claim, moreover, may be advanced regarding this relationship.


Braybrooke (1987) suggests it is not merely possible, but typical, that causal regularities provide a basis for inquiry into interpretive rules and that interpretive rules provide a basis for inquiry into causal regularities. He insists, in fact, that positivist and interpretive inquiry are not merely mutually supportive, but are often presupposing of one another.

Settled social rules are counterparts of causal regularities in that the rules lead to regularities for which the rule itself counts as a "cause." So, for instance, the rule about purchasing a more expensive gift for a spouse than for any other person "causes" this behavior to be an observable regularity. Note that this does not negate the fact that rules are brought into existence through the tacit understandings and interpretation of social actors, but merely asserts that while the rules exist, they act as causes of behavioral regularities. Thus it is characteristic of interpretive rules (at least settled social rules) that they will lead to regularities amenable to positivist inquiry.

At the same time, some regularities in social phenomena imply the existence of settled social rules. A causal regularity is likely to have a counterpart interpretive rule when the causal regularity applies to behaviors which an individual may voluntarily perform or not perform, and when some penalty for deviance from the regularity or reward for conforming to the regularity can be detected (Braybrooke 1987). Consider the regularity, cited by Belk (1979), that gift givers choose gifts which reflect their ideal self image more than their actual self image or their perceptions of the recipient. The degree to which gift givers choose gifts which reflect their ideal self image is, in general, under voluntary control. The reward individuals may experience as a result of conforming to the practice of choosing gifts reflecting their ideal self image is the recipient's acceptance and approval of the gift and, by extension, of the gift giver. The deterrent associated with not choosing a gift reflective of ideal self image might be the recipient's tacit or explicit rejection of the gift. It seems reasonable to suggest that settled social rules relating to the "appropriate" symbolism for gifts are likely to be associated with this causal regularity. Moreover, it is likely that investigation of these rules could offer opportunities for further positivist inquiry into causal regularities pertaining to gift giving.


While the foregoing discussion develops an understanding of the nature and significance of the tangency-between rules and regularities, it is also important to recognize disjunctions between these facets of interpretivism and positivism. In order to proceed with due caution and sensitivity in either the dialectic or the relativistic approach to utilizing alternative means of seeking knowledge, it is critical that these disjunctions be understood and respected.

While many forms of interpretivism would focus primarily on rules which are held by a society or social group, some of the more solipsistic versions of interpretivism focus on interpreting rules so singular as to apply to only one or a few individuals (Burrell and Morgan 1979). Though they are viable forms of interpretive inquiry, such studies are not likely to yield knowledge claims about rules which are sufficiently consistent in effect to form a basis for study of positivist regularities. Thus some interpretive person facts may not be conducive to positivist inquiry regarding person facts or group facts.

A further disjunction between rules and regularities stems from the different ways in which regularities-and rules are presumed to affect behavior. It appears that positivist regularities are assumed to operate in a rather mechanistic fashion. Interpretive rules are considered to affect behavior through the tacit apprehension and interpretation of social actors. Individuals are presumed to act according to rules because of their implicit assumptions about the purposes or interests an action serves within a given context (Packer 1985). An actor's assumptions regarding rules are typically tacitly held -- that is, the actor is not consciously aware of the social rules underlying a specific behavior. Thus gift-givers need not ever reflect on the "rules" affecting the amount spent on a gift to a spouse as opposed to a friend. Nonetheless, behavior occurring in conformity to rules is voluntary, since the actor has the potential to come to awareness of rules and to modify behavior or continue to follow rules.

Since interpretive inquiry is premised on the belief that individuals can modify their behavior because they become aware of and choose to modify rules, this approach has the potential to focus on situations where rules are in flux or in conflict. For instance, at a time when tacit societal rules are increasingly being seen to operate to the seeming disadvantage of women, rules relating to appropriate behaviors for men and women are changing considerably (see, for example, Hoover-Dempsey, Plas and Strudler Wallston 1986). Interpretive inquiry into the emergence of new rules, then, is possible whereas positivist inquiry into the emergence of causes is less feasible.

On the other hand, positivist inquiry has considerably more to offer than interpretive inquiry when regularities occur which are the unintended consequences of human action or inaction. Consider, for instance, the regularity with which it happens that aggregate consumer demand for a product increases as the price of that product decreases. While an interpretive rule may underlie the behavior of independent actors who are separately seeking to make "wise" purchase decisions, it is not an intended consequence of any individual's behavior that the aggregate quantity of goods purchased should increase. In such situations, positivist science may offer more insight than interpretive inquiry.


This discussion of points of tangency and divergence between positivist and interpretive science suggests that there is a significant relationship between the two. It demonstrates that for researchers who seek either a dialectic or a relativistic approach to dealing with divergent research streams, there is considerable potential for judicious interchange of research ideas between the streams .

The foregoing discussion also helps to clarify instances in which it is likeliest that tangencies will exist and that interpretive and positivist research streams have the greatest potential to complement one another. In cases where rules are extremely settled, are shared by groups of considerable size, and are not consciously being rejected by social agents, causal regularities are most likely to be associated with rules. In cases where regularities apply to behavior under the voluntary control of actors and where rewards or penalties are attached to conformity or nonconformity to the behavioral pattern in question, interpretive rules are likely to be associated with causal regularities.

This discussion also helps to explain the relationship between the regularities and rules which generate either person facts or group facts. It leads to an appreciation of how, for instance, positivist group facts such as might be derived from an socioeconomic study of spending habits, and a phenomenological case study of one consumer's "deviant" overspending might appropriately and fruitfully be conjoined. Greater insight into an issue noted by Anderson (1986), the examination of "the relationship between psychological/sociological explanations of behavior at an individual level and sociological/anthropological explanations at the group or cultural level" (p. 169), is also generated.

In summary, this paper lends momentum to the effort to move beyond a frustrating "either/or" debate on the merits of one research stream versus another, and lays the foundation for a less fractious, more fruitful, search for knowledge.


Anderson, Paul (1986), "On Method in Consumer Research: A Critical Relativist Perspective," Journal of Consumer Research, 13 (September), 155-173.

Belk, Russell (1979), "Gift-Giving Behavior," Research in Marketing, 2, 95-126.

Bern, Sandra (1974), "The Measurement of Psychological Androgyny," Journal of Consulting and Clinical Psychology, 42, 155162.

Bernstein, Richard (1985), Beyond Objectivism and Relativism: Science, Hermeneutics and Praxis, Philadelphia: University of Pennsylvania Press.

Braybrooke, David (1987), Philosophy of Social Science, Englewood Cliffs, N.J.: Prentice Hall Inc.

Burrell, Gibson and Gareth Morgan (1979), Sociological Paradigms and Organizational Analysis, London: Heinemann.

Caplow, Theodore (1984), "Rule Enforcement Without Visible Means," American Journal of Sociology, 89 (6) 1306-1323.

Giddens, Anthony (1984), The Constitution of Society, Berkeley: University of California Press.

Hoover-Dempsey, Kathleen, Jeanne Plas and Barbara Strudler Wallston (1986), "Tears and Weeping Among Professional Women: In Search of a New Understanding," Psychology of Women Quarterly, 10, 19-34.

Hudson, Laurel Ann and Julie Ozanne (1988), "Alternative Ways of Seeking Knowledge in Consumer Research," Journal of Consumer Research, 14 (March), 500-521.

Kuhn, Thomas (1970), The Structure of Scientific Revolutions, second edition, Chicago: University of Chicago Press.

McCracken, Grant (1988), "Lois Roget: Curatorial Consumer in a Modem World," in Culture and Consumption: New Approaches to the Symbolic Character of Consumer Goods and Activities, Bloomington: Indiana University Press.

Packer, Martin (1985), "Hermeneutic Inquiry in the Study of Human Conduct," American Psychologist, 40 (10), 1081-1093.

Peter, J. Paul and Jerry Olson (1983), "Is Science Marketing," Journal of Marketing, 47 (Fall), 111-125.

Qualls, William (1987), "Household Decision Behavior: The Impact of Husbands' and Wives' Sex Role Orientation," Journal of Consumer Research, 14 (2), 264-279.

Rossi, Alice (1984), "Gender and Parenthood," American Sociological Review, 49 (February), 119.

Ryan, Michael and Julia Bristor (1987), "The Symbiotic Nature of Hermeneutical vs. Classically Generated Knowledge," in Marketing Educator's Conference on Marketing Theory eds. Russell Belk et al, Chicago: A.M.A.

Spence, Janet and Robert Helmreich (1978), Masculinity and Femininity: Their Psychological Dimensions, Correlates and Antecedents, Austin: University of Texas Press.

von Wright, George Henrik (1971), Explanation and Understanding, London: Routledge and Kegan.