Eureka! and Other Tests of Signifcance: a New Look At Evaluating Interpretive Research

ABSTRACT - The present paper discusses two broad approaches to evaluating interpretive research. The first, foundationalism, is an epistemological position motivated by the subject-object distinction. It is proposed that a foundationalist logic underlies the evaluative criteria offered by both positivists and humanists. Advantages and disadvantages of evaluating interpretive research in terms of foundationalist criteria are discussed. An evaluative approach not premised on the subject-object distinction, anti-foundationalism, is then described. The critical relativist program is noted as the major example of anti-foundationalist thought in consumer research. It is proposed that critical relativism is a hermeneutic enterprise which can not completely obviate the need for epistemic foundations. Perceptual criteria are suggested as providing nondualistic, flexible bases for evaluating interpretive research and a description is given of one perceptually based criterion.

Citation:

Craig J. Thompson (1990) ,"Eureka! and Other Tests of Signifcance: a New Look At Evaluating Interpretive 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: 25-30.

Advances in Consumer Research Volume 17, 1990      Pages 25-30

EUREKA! AND OTHER TESTS OF SIGNIFCANCE: A NEW LOOK AT EVALUATING INTERPRETIVE RESEARCH

Craig J. Thompson, University of Tennessee

ABSTRACT -

The present paper discusses two broad approaches to evaluating interpretive research. The first, foundationalism, is an epistemological position motivated by the subject-object distinction. It is proposed that a foundationalist logic underlies the evaluative criteria offered by both positivists and humanists. Advantages and disadvantages of evaluating interpretive research in terms of foundationalist criteria are discussed. An evaluative approach not premised on the subject-object distinction, anti-foundationalism, is then described. The critical relativist program is noted as the major example of anti-foundationalist thought in consumer research. It is proposed that critical relativism is a hermeneutic enterprise which can not completely obviate the need for epistemic foundations. Perceptual criteria are suggested as providing nondualistic, flexible bases for evaluating interpretive research and a description is given of one perceptually based criterion.

INTRODUCTION

Within the social sciences, there exists a vast literature discussing various methodologies and analytic procedures for conducting research. The field of consumer research is no exception as numerous books and articles have been dedicated to this topic. While discussions of applying quantitative or qualitative research methodologies are widely available, much less work has explicitly discussed the evaluation of research. When research evaluation has been discussed, the focus is often on the negative aspects of the journal review process. As examples, Gardner (1985) proposes that psychology's progress has been hindered by reviewers turning lose their "methodological zeal" on any new theory and Holbrook (1986) has proposed that marketing's own review process is often a sadomasochistic endeavor. The specifics of these two critiques aside, they do give some indication that research evaluation is not a straightforward, reflexive process.

With the case of qualitative research, it seems that the evaluation process (either performed by an interested reader or a journal reviewer) could become even more difficult because: 1) qualitative techniques are not staple methodologies in consumer researcher and, thus, evaluators are less likely to be familiar with relevant procedures and theoretical frameworks; 2) qualitative methodologies are not monolithic (Jacob 1987). Important methodological and philosophical differences underlie different qualitative approaches such as protocol analysis (Ericcson & Simon 1984), naturalistic inquiry (Lincoln & Guba 1985), or existential-phenomenology (Giorgi 1983; Thompson, Locander and Pollio 1989). An evaluator applying assumptions from one approach to another will, in many cases, systematically misunderstand the research; and 3) in comparison to quantitative research, the evaluative criteria of qualitative methods often have less formalization and/or specification (Hudson & Ozanne 1988).

The present paper will discuss two broad approaches which can be taken in evaluating interpretive research: foundationalist and antifoundationalist. This discussion has descriptive rather than critical aim. The basic tenets of the two evaluative approaches will be delineated and their respective advantages and limitations described. Finally, an evaluative approach based on perceptual criteria will be offered and a description given of what some of its properties might be.

Foundationalist Approaches

Foundationalist thought is characterized by a desire to ground human knowledge. on a firm and indubitable archimedean point (Rorty 1979). The evaluative concern is primarily epistemological in seeking to identify procedures that can demarcate truth from non-truth and scientific from nonscientific knowledge. A common theme of foundationalism is an adherence to the subject-object distinction (Hekman 1986). With respect to interpretive research, foundationalism is embodied in both positivistic approaches, which emphasize "objectivity," and those versions of humanism emphasizing "subjectivity."

An explicit application of positivist criteria to qualitative research is given by Miles & Huberman's (1984) self-termed "soft-nosed positivism." In adopting a positivistic world-view, Miles & Huberman contend that knowledge must be firmly grounded on the objective side of the subject-object distinction. Their evaluative criteria are premised on the assumptions that: 1) there is an extant reality waiting to be discovered; 2) since "reality" is independent of human experience, people may have different subjective perceptions but they are perceiving the same objective entity; 3) "truth" is a correspondence between an interpretive (theoretical) term and a state of the world. This is a "soft-nosed" positivism because there is no requirement that the knowledge claims of qualitative research be of a hypothetico-deductive nature or validated by posthoc quantitative measures.

Miles and Huberman contend that qualitative research must address issues of verification (validity of conclusions), reliability, and generalizability. With this approach, an evaluator of qualitative research would want to see: 1) what procedures were used to assess the reliability and validity of an informant's statements. That is, did the informant express the same views over time and have sufficient knowledge to give a valid account of the phenomenon; 2) evidence of triangulation in which different sources of information are employed to converge on the phenomenon being investigated; 3) a description of procedures used to minimize researcher biases in both data collection and interpretation; and 4) assessment of the study's generalizability, including a description of the studies sampling procedures for both informants and critical events.

Miles and Huberman's "objective" approach to evaluating qualitative research has an undeniable foundationalist logic. The need for epistemic foundations, however, is no less for researchers taking a more overtly "subjective" approach. As a case in point, Lincoln & Guba (1985) describe a naturalistic research methodology based on assumptions diametrically opposed to those of positivism. Operating on the "subjective" side of the subject-object distinction, they propose that: 1) reality is a "multiple set of mental constructions" for which there is no ultimate benchmark for justifying truth value; 2) these multiple mental constructions will inevitably diverge; 3) the knower and that which is known are inseparable; and 4) truth is defined by community consensus. (It should be noted- that social constructionists, such as BerBer and Luckman (1966) and Gergen (1985), avoid the subject-object distinction by viewing knowledge as product of an ongoing interpersonal discourse grounded in a socio-cultural context rather than as particular state of a private mental world).

In arguing for the efficacy of this subjectivist world-view, a lengthy and detailed critique is given of classic positivist assumptions and research criteria. In implementing naturalistic inquiry, however, opposition turns to analogy as Lincoln and Guba's evaluative criteria (credibility, dependability, transferability, and, confirmability) bear a strong conceptual parallel to positivist criteria of internal validity, reliability, external validity, and construct validity (Hudson & Ozanne 1988; Thompson 1989). The major technique for establishing the confirmability of a naturalistic inquiry is a confirmability audit in which an independent auditor assesses the research process and the resulting interpretive results. The auditor looks for methodological procedures such as member checks, peer debriefings, triangulation, the keeping of reflexive journals (in which the reachers describe their own subjective mental processes) and then makes an evaluative judgment as to whether conclusions are logical, useful, adequate and firmly grounded in the data.

Despite their antithetical philosophical posturing, Lincoln and Guba's evaluative criteria are quite similar to those offered by Miles and Huberman. Both advocate triangulation, warn about potential sources of distortion and share a primary objective of providing foundations on which to base knowledge claims. For objectivists, the existence of epistemic foundations are simply a necessary background condition or core assumption. For subjectivists, epistemic foundations are needed to escape the philosophical quagmires of solipsism and skepticism. That is, if each person is in their own private subjective world (the solipsist assumption), then there is no guarantee that a person can have any knowledge outside of his/her idiosyncratic realm (the skeptical conclusion). For those operating on subjective side of dualism, conventional recourses have been to seek epistemic foundations from either a theological (a supreme being guarantees consistency among private worlds), nativist (common genetic structures guarantee consistency), or methodological standpoint (adapting "objective" methodologies to "subjective" inquiries). Lincoln and Guba have taken the latter recourse.

Other, more pragmatic, benefits are gained by evaluating interpretive in terms of foundationalist criteria. These criteria are easily explicated, are based on fairly standardized methodological procedures, and provide a widely shared frame of reference for evaluating interpretive research. In sum, foundationalist criteria allow those not highly familiar with qualitative methodologies to act as competent evaluators and can be readily offered as conventionally acceptable evidence of an interpretive study's epistemological merit.

The negative consequence is that the implicit assumptions motivating foundational evaluative criteria are often inconsistent with the overt assumptions motivating interpretive research. As one example, Lincoln & Guba strongly advocate triangulation with multiple methods, sources, and, (in some cases) investigators, to "validate" information. Giving the strong argument that there exist multiple and divergent realities, it is difficult to see what purpose triangulation can serve. Why should different sources (i.e. different constructions), as a methodological principle, have to agree? In proposing that triangulated information reveals a central tendency or "consensus reality," the logic is strikingly similar to the much critiqued positivist distinction between true and error variances (Thompson 1989). As one other example, it seems inconsistent, given Lincoln and Guba's detailed arguments that understanding requires a prolonged engagement with the context being investigated, that the ultimate arbitrator of a naturalistic inquiry's trustworthiness is an "independent auditor;" a disinterested party who ideally has no prior involvement in the study.

In sum, the evaluative criteria offered by Miles and Huberman (1984) and Lincoln and Guba (1985) are premised on dualistic assumptions, and therefore- have an implicit foundationalist logic. While operating on different sides of the subject-object distinction, both seek to ground knowledge on methodological procedures and third party (objective?) verification. As increasing numbers of philosophers and social scientists are questioning the assumptions of dualism, the necessity of epistemic foundations is also being challenged. With respect to interpretive research, a very different set of evaluative criteria are offered by antifoundationalist approaches (Rorty 1979).

Anti-foundational Approaches

Anti-foundational thought refers to philosophies rejecting the subject-object distinction, notions of "absolute truth" and attempts to find an Archimedean point on which to ground knowledge claims (Hekman 1986; Hudson & Ozanne 1988). A common theme among anti-foundational philosophy is a movement away from epistemological concerns and toward ontological ones. That is, knowledge and understanding are conceived as ways of being-in-the-world and the relevant task of philosophy and/or social science is seen as describing the socio-cultural context in which knowledge is constructed (Gadamer 1976; Gergen 1985; Heidegger 1962; Wittgenstein 1953). Anti-foundationalist thought has been at the center of the "linguistic turn" occurring in philosophy, the social sciences, and the sociology of knowledge (Hekman 1986). For the major anti-foundationalist philosophers, to inquire into the nature of being is to inquire into the nature of language (Bleicher 1980). Thus, Gadamer (1976) states that "language speaks us." Heidegger (1962) describes that human beings are "thrown" into a language. Similarly, Wittgenstein (1953) discusses that language is not like a tool, which can be put down when not being used, because human forms-of-life are inherently linguistic.

For purposes of identifying evaluative criteria for interpretive research, Wittgenstein provides a sound starting point because 1) Wittgensteinian thought is central to the critical relativist program proposed for consumer research (Anderson 1986); and 2) Wittgenstein's philosophy has an explicit methodological focus in seeking to use linguistic analysis to clarify "pseudo-problems" passing as philosophical dilemmas. A prominent Wittgensteinian concept is that of language game. By postulating that words attain meaning from their relation in a given language game, Wittgenstein contextualized meaning. For this view, the "meaning of a word is it's use in a particular situation" and it is due to a misguided "craving for generality" that philosophers seek out the one absolute meaning of a concept (Gier 1981; Wittgenstein 1953). Wittgenstein redefined philosophical problems as "grammatical fictions" arising from the assumption that a word used in different language games must have the same meaning.

With respect to consumer research, the critical relativist program is amenable to Wittgenstein's account of linguistic meaning. With language games, meaning is not simply a correspondence between word and object. With critical relativism, scientific knowledge is not simply a correspondence between a theoretical term and a state of the world but emerges from a system of metaphysical beliefs, value commitments, and cognitive aims (Anderson 1986). Thus, it can be said that what constitutes scientific knowledge depends on which language game is being employed. Just as no one ostensive definition describes the meaning of a word, no one set of methodological procedures describes the constitution of scientific knowledge. Critical relativism proposes that the evaluative criteria must vary according to the assumptions and cognitive aims underlying the research. Researchers are required to explicitly state their assumptions, cognitive aims and also to provide appropriate evaluative criteria. The reader of the research is required to understand the study's aims and assumptions and faithfully seek to apply the relevant evaluative criteria.

Several advantages are readily seen with the critical relativist approach to evaluating interpretive research. First, and very pragmatically, the degree of explication required to specify the relevant paradigmatic assumptions, cognitive aims, and value commitments will improve the thoroughness and comprehensibility of an interpretive study. Second, the problem of applying inappropriate evaluative criteria is avoided because the researchers, themselves, provide the relevant criteria. Third, since critical relativism requires no specific set of methodological procedures, the approach facilitates methodological pluralism.

The one issue which anti-foundationalism has not fully accounted for, however, is how evaluative judgments can be made solely within the confines of a given conceptual system. Evaluation would seem to require that some criteria and beliefs transcend a specific context or conceptual system. For example, in arguing that a concept's meaning is given by its function in a language game, Wittgenstein's metaphor of language is a game transcended any specific linguistic usage. Similar transcendental assumptions can be seen in the works of other antifoundationalist philosophers. Heidegger postulates the existence of "existential" categories that transcend all forms-of-being (1962). Gadamer proposes that critical reason can be used to distinguish between "legitimate" and "illegitimate" prejudices (1976). While anti-foundationalists made a move toward ontology (seeking only to describe the nature of being), the need for some form of an epistemological foundation was not completely obviated.

The critical relativist program also requires some foundational assumptions, albeit very broad ones. As with Gadamer, there seems to be some requirement for "critical reason." It is assumed that the evaluator, once provided with a set of criteria, will reasonably and fairly apply the criteria. Since critical relativism is devoutly non-nihilistic, the evaluator must also have some means by which to distinguish between legitimate and illegitimate cognitive aims, assumptions, and criteria. Third, it is assumed that incommensurability is not an issue in evaluation. That is, to implement the critical relativist program, the reader must be able to comprehend the intended meanings of the researcher and use the provided evaluative criteria in the intended fashion. The critical relativist contends that research is evaluated on the basis of the researcher's guiding theoretical program rather than on a "fusion of horizons" (to use Gadamer's terminology) between theoretical perspectives of reader and researcher. While arguing for the inherent theory ladenness of scientific observation, critical relativism grants an epistemic exemption to its own evaluative endeavors. The critical relativist, like the scientists who conduct research, must adopt a perspective in evaluating research. For example, when assessing a research program's "methodological, ontological, metaphysical, axiological commitments" and its "realizable cognitive and social aims" (Anderson 1986), a hermeneutic process is being implemented. Rather than being monolithic, it would seem that critical relativism would be multi-perspectival so that disagreements might arise over the "realizability" of a given research program's cognitive aims or in specifying just what the cognitive aims are in the first place.

No charges of "self-refutation" are being levied against critical relativism or the antifoundationalist philosophy on which it is based. The point is that the concerns of foundationalist philosophers are not so easily laid to rest. Antifoundationalists also need some ground on which to base to their claims. The insight of critical relativism is that no one set of methodological procedures guarantees knowledge. The application of positivist criteria can be both limiting and philosophically inconsistent to interpretive research. -Yet, the foundationalist insight that knowledge claims must be based on something which is, at least, assumed to transcend contextual variations also has descriptive merit. Out of this dialectic tension, a need can be seen for nondualistic evaluative criteria that are not dependent on any one set of methodological procedures but that still allow for some grounding of knowledge claims.

Toward A Perceptual Approach

One foundation that might be used to ground the knowledge claims of interpretive studies, not premised on inherently dualistic notions, is perception. Perceptual experience often serves a root metaphor for anti-foundationalist, such as in Wittgenstein's contention that understanding is a process of "seeing as." The French phenomenologist Merleau-Ponty also suggests that conceptual knowledge has as its base pre-reflective perceptual experience (1962). As the following quote from the Phenomenology of Perception (1962) illustrates, Merleau-Ponty's view of perception is highly congruent with some of the more typical aims of interpretive research.

Our first task will be to re-discover phenomena, the level of living experience through which things and other people are given to us. We shall no longer hold that perception is an incipient science but conversely that classical science is a form of perception which loses sight of its origins and believes itself complete (p.57).

A common theme of interpretive research is to provide a thick description of lived experience, free from natural science prejudices (Belk, Sherry, & Wallendorf 1988; Hirschman 1986; Hudson & Ozanne 1988; Thompson, Locander & Pollio 1989). In this sense, evaluating interpretive research from perceptual criterion is philosophically consistent with the aims underlying consumer research's use of interpretive methods.

Identifying an applicable set of perceptual evaluative procedures is no small task. Gestalt psychology, however, may serve as a fruitful starting point (Koffka 1935; Kohler 1947). Gestalt theory suggests that one important perceptual criterion is that the research affords insight into the phenomenon being explored. From this perspective, insight is not conceived as a mysterious, subjective experience but as a type of understanding that is-in the-world (Kohler 1947; Merleau-Ponty 1962). For example, when a problem is solved by insight, an understanding emerges from grasping a pattern or organization (Kohler 1947). The problem and the pattern yielding a solution are not private, subjective entities but are rather public phenomena.

A demonstration of insight is provided in the following example taken from Kohler (1969). The problem is to comprehend the relationship between following three lines of integers:

A) 0,  1,  2,  3,   4,   5,    6,   7,   8...

B) 0,  1,  4,  9, 16, 25, 36, 49, 64...

C)      1,  3 , 5,   7,    9, 11, 13, 15...

The solution pattern is that line A consists of a series of integers, line B consists of the squares of those integers, and line C is derived from subtracting each square from the preceding square in line B: a procedure which gives the series of odd numbers. When these relationships are grasped, one has experienced an insight.

There are at least three types of insight that interpretive research could afford:

1) Conceptual Gestalt- In this form of insight, the interpretation allows the evaluator to see a set of qualitative data as a coherent pattern or gestalt. What might have previously seemed a set of discrete and unrelated events becomes a good conceptual figure. As an example, Freud's interpretations allowed a wide range of neurotic behaviors to be seen as manifestations of the same general phenomena: unconscious repressions of sexual experiences and desires (van den Berg 1961).

In that conceptual gestalts are not dependent on empathy, the evaluator may disagree with the interpretation while still seeing how the interpretive pattern derives from the data. With a conceptual gestalt, the reader can grasp the proposed meaning of the phenomenon even if it does not directly relate to his/her own field of experience. As an example, one can comprehend Thomas and Znaniecki's (1958) ethnography of the Polish peasant without knowing what such an experience feels like. A well-known example of a conceptual gestalt is learning by insight, such as in the "Eureka" experience described in the myth of Archimedes (Pollio 1982).

2) Phenomenological Gestalt- In this form of insight, the interpretation resonates with the reader in such a way that a more personalized, first-person understanding of the phenomenon is gained. In this case, the interpretation is seen as describing a like pattern in the reader's own field of experience. The interpretation may even allow the reader to articulate some aspect of their own experience which had previously defied overt description. Phenomenological gestalts afford to the reader a heightened self-awareness and a personally relevant understanding of the phenomenon. Unlike the previous form of insight where the reader can see the interpretive pattern without necessarily agreeing with it, the reader feels there is something "right" about the interpretation in that it meaningfully relates to his/her own experience. A well-known example of a phenomenological gestalt is Dilthey's conception of "verstehen.'

3) Paradigmatic Gestalt- With this form of insight, the interpretation has such a profound effect on the reader that the world is seen in an entirely new way. That is, the reader's previous world-view is replaced by a new organizational scheme. This form of insight has been described as a "paradigm shift" (Kuhn 1970). As one Kuhnian example, Einstein's theory of relativity gave rise to an entirely new understanding of space, time, and motion.

Paradigmatic gestalts have a different nature than those of the conceptual or phenomenological form. With the latter two, the interpretation affords an awareness of a novel pattern but this understanding is still within the reader's existing horizon of conceivability. With paradigmatic insight, this horizon altered in such a way that what had previously been inconceivable is now seen as a potential or actual state of the world. This is perhaps the most that could be asked of an interpretive study because it changes the horizons of what can be.

CONCLUSION

Two major approaches to evaluating interpretive research have been discussed. Foundationalism is described as a dualistically motivated epistemology seeking to ground knowledge on a firm, indubitable base. In accepting the subject-object distinction, positivist and humanist methods reveal a foundationalist logic. Anti-foundationalist approaches have a more ontological focus in seeking to describe understanding without dualistic categories. The critical relativist program is the major example of this approach in consumer research. It was noted that even anti-foundationalists cannot completely obviate the need for some epistemic ground on which to base their knowledge claims. Perceptually based evaluative criteria were suggested as one means for a non-dualistic grounding of interpretive research's knowledge claims.

REFERENCES

Anderson, Paul F. (1986), "On Method in Consumer Research: A Critical Relativist Perspective," Journal of Consumer Research, 13 (September), 155-170.

Belk, Russell W., John F. Sherry, and Melanie Wallendorf(1988), "A Naturalistic Inquiry into Buyer and Seller Behavior at a Swap Meet," Journal of Consumer Research, 14 (March), 449-470.

Berger, Peter and Thomas Luckmann (1966), The Social Construction of Reality, Garden City, NY: Doubleday.

Bleicher, Josef (1980), Contemporary Hermeneutics, London: Routledge & Kegan Paul.

Ericsson, Anders K. and Herbert A. Simon (1984), Protocol Analysis: Verbal Reports As Data, Cambridge, MA: MIT Press.

Gadamer, Hans:Georg (1975), Truth and Method, New York: Seabury Press.

Gardner, Howard (1985), The Mind's New Science, New York: Basic Books.

Gergen, Kenneth J.98-S), "The Social Constructionist Movement in Modern Psychology, American Psychologists, 40, 266-275.

Gier, Nicholas F. (1981), Wittgenstein and Phenomenology, Albany, NY: State University of New York Press.

Girogi, Amedeo (1983), "Concerning the Possibility of Phenomenological Psychological Research," Journal of Phenomenological Psychology, 14 (Fall), 129-170.

Heckman, Susan J. (1986), Hermeneutics and the Sociology of Knowledge, Cambridge: Polity Press.

Heidegger, Martin (1962), Being and Time, New York: Harper & Row.

Hirschman, Elizabeth C. (1986), "Humanistic Inquiry in Marketing Research: Philosophy, Method, and Criteria," Journal of Marketing Research, 13 (August), 237-249.

Holbrook, Morris B. (1986), "A Note on Sadomasochism in the Review Process: I Hate When That Happens," Journal of Marketing Research, 50 (July), 104-108.

Hudson, Laurel Anderson and Julia L. Ozanne (1988), "Alternate Ways of Seeking Knowledge in Consumer Research," Journal of Consumer Research, 14 (March), 508-521.

Jacob, Evelyn (1987), "Qualitative Research Traditions," Review of Educational Research, 57 (Spring), 1-50.

Koffka, Kurt (1935), Principles of Gestalt Psychology, New York: Harcourt, Brace, & World.

Kohler, Wolfgang (1947), Gestalt Psychology, New York: New American Library.

Kohler, Wolfgang (1969), The Task Of Gestalt Psychology, Princeton: University of Princeton Press.

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

Lincoln Yvonna S. and Egon Cuba (1985), Naturalistic Inquiry, Beverly Hills: Sage.

Lombardo, Thomas J. (1987), The Reciprocity of Perceiver and Environment: The Evolution of James J. Gibson's Ecological Psychology, Hillsdale, NJ: Lawrence Erlbaum Press.

Merleau-Ponty, Maurice (1962), The Phenomenology of Perception, London: Routledge & Kegan Paul.

Mick, David Glen (1986), "Consumer Research and Semiotics: Exploring the Morphology of Signs, Symbols, and Significance," Journal of Consumer Research, 13 (September), 196-213.

Miles, Matthew B. and A. Michael Huberman (1984),Qualitative Data Analysis, Beverly Hills, CA.: Sage.

Pollio, Howard R. (1982), Behavior and Existence, Monterey, CA.: Brooks/Cole Publishers.

Ricoeur, Paul (1976), Interpretation Theory, Fort Worth, TX: Texas Christian University Press.

Rorty, Richard (1979), Philosophy and the Mirror of Nature, Princeton: Princeton University Press.

Thomas, William I. and Florian Znaniecki (1958), The Polish Peasant in Europe and America, New York: Dover Press.

Thompson, Craig J. (1989), "Machines and Melodies: Reliability and Validity in Qualitative Research," in Proceedings of the Society for Consumer Research, ed. David W. Schumann, Atlanta, GA: American Psychological Association.

Thompson, Craig J., William B. Locander, Howard R. Pollio (1989), "Putting the Consumer Back into Consumer Research: The Philosophy and Method of Existential-Phenomenology," Journal of Consumer Research, (forthcoming September).

van den Berg, J.H. (1961), The Changing Nature of Man, New York: W.W. Norton & Company.

Wertz, Frederick J. and Joan M. Greenhut (1985), "A Psychology of Buying: Demonstration of a Phenomenological Approach in Consumer Research," in Advances in Consumer Research, Vol. 12, eds. Elizabeth Hirschman and Morris Holbrook, Provo, Utah: Association for Consumer Research.

Wittgenstein, Ludwig (1953), Philosophical Investigations, Oxford: Basil Blackwell.

----------------------------------------