A Primer For Ethnographic Research With a Focus on Social Policy Issues Involving Consumer Behavior

Ronald Paul Hill, Villanova University
ABSTRACT - The purpose of this paper is to depict the process involved in the conduct of ethnographic research as well as the problems faced by consumer behavior ethnographers. The paper opens with an introduction to the topic. Then, this qualitative method is defined and described, data collection and approaches to analysis are explicated, and recommendations for writing ethnographies are provided. Finally, troublesome issues involving objectivity, the ethnographer's presence, scientific integrity, and ethical considerations are delineated. Each facet of the ethnographic research process as well as these areas of concern are discussed within the context of research on the social issue of homelessness.
[ to cite ]:
Ronald Paul Hill (1993) ,"A Primer For Ethnographic Research With a Focus on Social Policy Issues Involving Consumer Behavior", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 59-62.

Advances in Consumer Research Volume 20, 1993      Pages 59-62


Ronald Paul Hill, Villanova University


The purpose of this paper is to depict the process involved in the conduct of ethnographic research as well as the problems faced by consumer behavior ethnographers. The paper opens with an introduction to the topic. Then, this qualitative method is defined and described, data collection and approaches to analysis are explicated, and recommendations for writing ethnographies are provided. Finally, troublesome issues involving objectivity, the ethnographer's presence, scientific integrity, and ethical considerations are delineated. Each facet of the ethnographic research process as well as these areas of concern are discussed within the context of research on the social issue of homelessness.


Recent research in the consumer behavior field has utilized ethnographic methods to investigate market settings, purchase decisions, and consumption behaviors (see Belk, Sherry, and Wallendorf 1988; Belk, Wallendorf, and Sherry 1989; Heisley, McGrath, and Sherry 1991; Sherry 1990; and Wallendorf and Arnould 1991 for some excellent examples). While each of these studies makes an important methodological contribution, none has been published with the expressed intent of focussing on such issues. Therefore, the purpose of this paper is to describe the process involved in conducting ethnographic research as well as the problems faced by ethnographers in our discipline using research on homelessness as the focus (Hill 1992; 1991; Hill and Stamey 1990).


Historically, ethnography is interdisciplinary, arising out of the qualitative research traditions in sociology and anthropology (Adler and Adler 1987). This method is defined "by its attempt to generate participant insight into aspects of group life," and data typically is collected through "participation in settings, observations, interviews, and other discussions" with informants (Prus 1987, p. 254-255). The researcher's understanding of the phenomenon "emerges" over the course of this data collection, and moves from an item level of analysis to a search for identifiable patterns among these objects to the development of themes that show how these patterns are related to one another (Borman, LeCompte, and Goetz 1986). The final step involves an integration with existing literature and theory within the social sciences.

The following subsections discuss specific facets of ethnography and are organized according to the recommendations of Fetterman (1989).

Guiding Principles

As Fetterman (1989, p. 11) notes, "The ethnographer enters the field with an open mind, not an empty head." Thus, researchers usually take a holistic perspective that incorporates both emic and etic points-of-view. A holistic outlook attempts to gain a comprehensive picture of the group under study by framing their activities within the larger context in which they occur. For example, in my work with the hidden homeless (Hill and Stamey 1990), our investigation began with a focus on how individuals cope with a lack of consumer products from adequate shelter to health care and hygiene to food and clothing. However, data revealed that homeless persons' survival strategies are often tied to larger communities that reject the traditional welfare system as well as many of its supporting institutions. This broader context helped us understand the sources of self-esteem for the hidden homeless as well as their preferences for and avoidance of certain sources of products.

Central to this approach is the emic perspective which concentrates on natives' understandings of their world, and recognizes and accepts multiple realities among informants in a nonjudgmental fashion. For example, the hidden homeless (Hill and Stamey 1990) viewed themselves as living by their own resources, separate and apart from the welfare system while the sheltered homeless (Hill 1991) often were dependent both physically and emotionally on this same system for survival. While both subpopulations faced some of the same difficulties, they held very different views of supporting institutions. Thus, through extensive research, very distinctive realities emerged.

The etic perspective, on the other hand, represents an external, "scientific" viewpoint of this same reality. This outlook requires that the researcher make sense of the data that have been collected through scientific analysis. Occasionally, this task is performed after the ethnographer leaves the field and creates a certain amount of distance between him/herself and informants. The end result is interpretive themes that illuminate the informants' world through placement of findings within the context of theory and research in the social sciences.

Data Collection

If the emic perspective is the "heart" and the etic perspective is the "brain" of ethnographic inquiry, then field work is the "soul." Through long-term participant observation, the ethnographer is able to "internalize the basic beliefs, fears, hopes, and expectations of the people under study" (Fetterman 1989, p. 45). Using this approach, the researcher acts as a human instrument, relying on all of his or her senses for data collection.

A typical ethnographic investigation may begin using a "big net" approach where the researcher interacts with a wide variety of informants in as many different settings as possible. The ethnographer often talks with whomever is willing and able in order to gain initial access to informants' private worlds. Over time, these contacts may act as key informants, providing additional persons from their social networks for the researcher to contact. For instance, in my work with the hidden homeless (Hill and Stamey 1990), the research team visited shanty towns, abandoned buildings and cars, bridge abutments, tunnels, and wooded areas known to be inhabited by homeless persons in an attempt to understand the heterogeneity that existed within this segment of the homeless population. In each of these environments, key informants were identified and long-term relationships were established. As trust and respect developed between the research team and these homeless persons, they helped identify other individuals within their communities who faced similar struggles and provided needed introductions.

While many forms of data collection are employed by ethnographers, the interview is often the most useful. From casual conversations with implicit agendas to formal interviews with more directed purposes, researchers attempt to discover how informants think and feel compared to one another. On occasion, discussions of songs, television shows, places, or events can be used as projective techniques to explore particularly troublesome experiences. Such interactions help the researcher establish a healthy rapport, especially if informants are allowed to control the flow of the discussion much of the time. Also, the occasional use of the tape recorder captures long verbatim quotes without the disruption of note taking by the ethnographer.

My experiences at the Sisters of Mercy shelter support this position (Hill 1992; 1991). During this investigation, I had literally hundreds of conversations, discussions, and interviews during collection of the data. Such repeated exposures to and long term interactions with informants built trust and resulted in their provision of intimate details regarding their lives. Further, the tape recorder proved invaluable, particularly in my interviews with children who were delighted by the prospect of hearing their voices from the machine after the sessions were completed. Also, one peer reviewer (a social worker) suggested that I use fantasy as a projective technique to examine underlying fears among these children, and the resulting fantasies involving "rich" kids and future homes were particularly enlightening.

Data Analysis

Data analysis is probably the most misunderstood aspect of ethnography, due, in part, to the fact that there is no single form of evaluation or phase of the process where it is performed. In truth, ethnographic analyses are iterative, starting at the first stages of data collection and continuing until the final stages of writing. Further, analyses are often multiple, may take a variety of forms within a single investigation, and typically cycle through and back to different issues as distinct data sources are collected or reviewed.

Occasionally, key events that represent shared experiences among the group members under investigation are used as the focal point of analyses. The ethnographer searches through the data involving these events for underlying patterns of thought or behavior. "Crystallization" occurs when there is "a convergence of similarities that spontaneously strike the ethnographer as relevant or important to the study" (Fetterman 1989, p. 101). However, the researcher continues to seek additional sources of the same patterns (and groups of interrelated patterns often referred to as themes) as well as limiting exceptions to what appears to be important commonalities. This process, known as triangulation, is considered a test of validity, and continues until all reasonable alternative explanations have been eliminated.

My field experience with homeless women at the shelter (Hill 1991) may shed some light on this facet of ethnography. Multiple data sources were used during this investigation including field notes, audio recordings of interviews, and photographs. These data were analyzed weekly over the course of the study, and helped to revise and modify different emerging interpretations at different points in time. The key event that remained the focus of these data was their decision to seek support from the shelter. As time passed and themes crystallized, interviews with the women and conversations with the staff were directed towards negative cases in an attempt to consciously seek disconfirming evidence. After all possibilities had been explored and final analyses were conducted, the writing up phase began.

Writing Ethnographies

Writing ethnographies represents a significant challenge for researchers, particularly in a discipline such as consumer behavior that has traditionally embraced quantitative methods. Nonetheless, good ethnographic writing typically is composed of two interrelated parts: "thick" description and "thick" interpretation (see Geertz 1973). Thick description provides the context in which behaviors take place, incorporating cultural meaning into the written text. The sources of this description are the field notes recorded during and after each interaction with informants, audiotapes of all interviews as well as their transcriptions, and photographs that capture important moments, possessions, etc. Whenever possible, the voice of informants through the use of verbatim quotes should be employed to provide details regarding their thoughts and actions.

Thick interpretation provides the reader with a "road map" to help him or her understand the complex nature of the field. This part of ethnographic writing requires the explication of the patterns and themes that emerged during interpretation, embedded within existing literature in the social sciences. The end result should give the reader a gestalt of the focal environment that moves his or her understanding and consciousness of the field to a higher level.

One of the greatest difficulties faced by ethnographers during this stage is the decision involving what information to include in the final write up. Interactions with the field may uncover a wide range of data and findings, some of which may be only tangentially related to the researcher's discipline. Thus, many ethnographers choose to ignore certain data and emphasize others, or write multiple reports, each directed to a different audience.

My work with the hidden homeless (Hill and Stamey 1990) provides an example of some of the dilemmas faced by ethnographers when writing from their data. The data collection phase of this project took place over several years, resulting in a "mountain" of field notes, audio tapes, and photographs. Since the primary investigators are from different disciplines (marketing/consumer behavior and sociology), we tended to emphasize different findings and search for different meanings within the same data bases. However, because the written paper was to be submitted to the Journal of Consumer Research, possessions and consumption behaviors became the focus of the final report, and the social network and relationships among informants became a secondary concern.


While there are many issues that pose significant problems for consumer researchers employing ethnography, the following four have particular relevance given the positivist orientation of the discipline and the sensitive nature of social policy matters.

Objectivity Versus Subjectivity

One of the most critical debates among ethnographers is the balance between objectivity and subjectivity which indicts the researcher's level of involvement in or detachment from the field. As Adler and Adler (1987, p. 10) note:

Ideally, ethnographers were to get close to members, participate in some of their activities, gain their trust and confidence, and discover their subjective perspectives and interpretations. At the same time, they were to keep themselves firmly anchored in the scientific conceptual framework so that they could analyze the observations and accounts they were gathering from a detached, objective vantage.

This "balancing act" was the result of the widespread influence of positivism which recommended that ethnographers seek to improve the validity and reliability of their methods.

Existential sociologists, on the other hand, advocated that researchers get as close to the phenomenon as possible in order to gain intimacy and trust and eliminate evasive tactics by informants (see Douglas 1976). "Scientific analysis does not require objective detachment, they argue, but occurs within the theoretical self-reflection of the trained social scientist" (Adler and Adler 1987, p. 12). Thus, to gain a truly emic perspective, ethnographers must get as close to the phenomenon as possible and reflect upon their own subjective feelings and experiences with informants.

I have struggled with this issue within the context of my own ethnographic studies, particularly in my limited work with homeless children (Hill 1992). While I found no perfect position along the objective-subjective continuum, the recommendations of Borman et al. (1986) provided a reasonable guide. They suggest that ethnographers adopt a "disciplined subjectivity" that requires a rigorous search for possible biases in each interaction with informants as well as an almost psychotherapeutic introspection that demands a constant dialogue with the self. Also, periods of detachment from the field, working in teams, and seeking input from peer reviewers within and outside the discipline can broaden the researcher's perspective and supply fresh approaches to data collection and analysis.

The Ethnographer's Presence

A separate but related problem involves the physical presence of the ethnographer in the field. According to Stoddart (1986, p. 107-108), the interactional presence of researchers with the phenomenon of interest can erode "the possibility of achieving the goal that occasioned their presence in the first place." Thus, ethnographers view their attendance in the field as potentially changing its natural state, and often work to reduce the possibility of such contamination.

One potential solution to this dilemma is to "fade into the background" of the domain under investigation. The ideal ethnographer is the invisible researcher "who sees without being observed and, consequently, captures the natural field without tainting it" (Stoddart 1986, p. 108-109). However, this ideal is difficult if not impossible to attain. Instead, Stoddart (1986) suggests that ethnographers minimize their visibility through "disattending" which may involve choosing a locally appropriate style or role within the focal environment.

In my work on homelessness, this concern required attention. During the investigation of the hidden homeless (Hill and Stamey 1990), the research team was careful to avoid serious consequences of this problem by dressing informally, using informal language, and engaging in normal, communal behaviors such as eating or sharing drink. Further, in my ethnography of the shelter (Hill 1992; 1991), I assumed the role of volunteer so that I could blend into the background activities that regularly occurred. My job consisted of sweeping the floors, setting the dining room tables, preparing food, and transporting residents to and from the facility. Since the shelter employed many volunteers, my presence was a natural part of this environment.

Scientific Integrity

Serious questions regarding the scientific integrity of ethnographic research have surfaced within and outside the discipline of consumer behavior. According to Agar (1983, p. 41), "a classic criticism of ethnographic reports [is] that they present general conclusions with a few supporting anecdotes." Another criticism involves concern that multiple studies of the same or similar groups can result in very different conclusions due to dissimilarities in the cultures and backgrounds of the researchers as well as the intended audiences of publications (Agar 1982). The end result is that "there is no way to be certain that what is portrayed is anything more than a researcher's etic imposition of meanings and constructs upon a setting rather than an authentic representation of the thoughts and beliefs of the people under study" (Borman et al. 1986, p. 49).

One approach to resolving these dilemmas calls for the analysis of "strips," defined as "any bounded phenomenon against which ethnographers test their understanding of the group" such as a particular encounter or event (Agar 1982, p. 789; also see Agar 1983). These strips are then subjected to sequential analysis consisting of breakdowns, resolutions, and coherence. Breakdowns occur when the researcher's preconceived notions based on his or her background or training are violated during the evaluation of strips. Resolution of these breakdowns requires that the ethnographer "tinker" with these inferences until a new pattern emerges that incorporates this new knowledge. The result should be coherence - a clear understanding of the cause(s) of the breakdown and its integration into the researcher's view of the field.

However, this emerging perspective must be considered tentative until it has been corroborated by other sources. An active search using additional methods and informants must be employed with an emphasis on negative cases and limiting exceptions. Further, when appropriate, peer reviewers from the same and other disciplines who are concerned with the focal population or issues can provide additional insights through their own analyses of these strips.

During my investigations of the homeless, apprehension regarding the scientific integrity of ethnography began to surface in the consumer behavior discipline. Fortunately, a timely and well-conceived piece by Wallendorf and Belk (1989) was circulated among my peers and I was able to use many of their methodological suggestions to guide my work. In both of my studies (Hill and Stamey 1990; and Hill 1992; 1991), findings from and perceptions of the field were crafted into interpretations through the use of triangulation across methods and informants as well as negative cases. Further, peer reviews from professionals in the disciplines of anthropology, psychology, sociology, and social work were employed throughout the conduct of these ethnographies to broaden interpretations beyond my own training in marketing/consumer behavior.

Ethical Considerations

While ethical issues should be considered by all ethnographers, they are particularly relevant for those who investigate social policy matters. The centerpiece of long-term relationships between ethnographers and informants is trust, which places certain demands on researchers. In order to establish and maintain this trust, the ethnographer must be honest in all dealings with informants, explaining his or her purpose for entering the field in a clear and nontechnical manner. The intended uses of all forms of data, especially audio recordings and photographs, should be explained and permission must be received prior to collection.

Most importantly, the researcher should do everything possible to disguise the identity of informants during the writing up phase. This tactic is particularly relevant in situations where the ethnographer has been privy to what Fetterman (1989, p. 135) calls "guilty knowledge," described as "confidential knowledge of illegal or illicit activities." In such cases, field notes should be devoid of identifying labels, pseudonyms should be used in written reports, and the names and locations of places or organizations should be concealed. Finally, the researcher investigating social issues should play an advocacy role by disseminating findings beyond the academic world so that policy makers whose actions are currently impacting the lives of informants are made privy to results.

Clearly, my work with the homeless raised important ethical issues. In my investigation of the hidden homeless (Hill and Stamey 1990), the research team regularly came in contact with persons who illegally established living quarters in settings like abandoned buildings. Further, many persons had connected appliances to municipal electrical outlets or scavenged from buildings/containers without permission from the owners. Also, in my study of sheltered women (Hill 1991), some of my informants were concerned that if anyone knew they were living at this facility it might inhibit their ability to be reunited with their children or embarrass their families. Thus, in both situations, several steps were taken to protect the homeless including the use of pseudonyms or generic descriptions (e.g., WM 30s) to identify informants, disguising the locations of living quarters and the shelter, and only publishing photographs that did not expose persons who were vulnerable to retribution. The process of dissemination of these findings to policy makers is now underway, especially to those interested in helping homeless children or improving health care delivery systems to the homeless.

Concluding Remarks

While this paper has not addressed all of the dilemmas faced by consumer behavior ethnographers, many of the central issues have been raised and some possible solutions within a social policy context have been addressed. However, I would like to stress three additional points. First, ethnography is often very satisfying to its practitioners because it allows them to directly experience the world of informants and all of its variations. Living through the "highs" and the "lows" of their lives allows the researcher to know the phenomenon under investigation in a way that few other methodologies permit. Second, I have been asked by would-be ethnographers where to begin. My first reaction is to be "opportunistic." According to Reimer (1977, p. 469), this strategy recommends that researchers "rely upon their own unique biographies, life experiences, and situational familiarities in doing their research." For example, my eight years as a coach of a martial arts team took me into many lower class communities for competitions. In this environment, I learned how to mix with persons of different races and socioeconomic backgrounds which, later on, facilitated my research with the homeless. Third, as a discipline, consumer behavior is just beginning to embrace qualitative methods and must rely upon other fields for support and experience. Therefore, it is essential that we seek input from professionals in anthropology and sociology, share our findings with them, and seek long-term collaborative relationships if we are to build a coherent body of knowledge based on such explorations (see Prus 1987 for additional ways to accomplish this synergy).

In closing, the words of the writer Scott Sanders (1991; p. X) provide sage advice, especially for ethnographers:

I do not expect to arrive at the absolute center or circumference of things, at least not along a path of words. I will follow that path as far as it leads, then go on ahead in silence. The journey home is my effort to come fully awake, to understand where I actually live. If, on the way, I have discovered any secrets worth telling, they must be ones known to all of us in our clear moments. I seek a truth as common as dirt or laughter, and as rare.


Adler, Patricia A. and Peter Adler (1987), "The Past and the Future of Ethnography," Journal of Contemporary Ethnography, 16 (April), 4-24.

Agar, Michael H. (1983), "Ethnographic Evidence," Urban Life, 12 (April), 32-48.

Agar, Michael H. (1982), "Toward an Ethnographic Language," American Anthropologist, 84, 779-795.

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

Belk, Russell W., Melanie Wallendorf, and John F. Sherry, Jr. (1989), "The Sacred and the Profane in Consumer Behavior: Theodicy on the Odyssey," Journal of Consumer Research, 16 (June), 1-38.

Borman, Kathryn M., Margaret D. LeCompte, and Judith Preissle Goetz (1986), "Ethnographic and Qualitative Research Design and Why It Doesn't Work," American Behavioral Scientist, 30 (September/October), 42-57.

Douglas, Jack D. (1976), Investigative Social Research: Individual and Team Field Research, Beverly Hills, CA: Sage Publications, Inc.

Fetterman, David M. (1989), Ethnography Step by Step, Newbury Park, CA: Sage.

Geertz, Clifford (1973), The Interpretation of Culture, New York: Basic.

Heisley, Deborah D., Mary Ann McGrath, and John F. Sherry, Jr. (1991), "`To Everything There is a Season:' A Photoessay of a Farmers' Market," in Highways and Buyways: Naturalistic Research from the Consumer Behavior Odyssey, Russell W. Belk, ed. Provo, UT: Association for Consumer Research, 141-166.

Hill, Ronald Paul (1991), "Homeless Women, Special Possessions, and the Meaning of `Home': An Ethnographic Case Study," Journal of Consumer Research, 18 (December), 298-310.

Hill, Ronald Paul (1992) "Homeless Children: Coping with Material Losses," Journal of Consumer Affairs, 26 (Winter), forthcoming.

Hill, Ronald Paul and Mark Stamey (1990), "The Homeless in America: An Examination of Possessions and Consumption Behaviors," Journal of Consumer Research, 16 (December), 303-321.

Prus, Robert (1987), "Generic Social Processes: Maximizing Conceptual Development in Ethnographic Research," Journal of Contemporary Ethnography, 16 (October), 250-293.

Reimer, Jeffery W. (1977), "Varieties of Opportunistic Research," Urban Life, 5 (January), 467-477.

Sanders, Scott Russell (1991), Secrets of the Universe: Scenes from the Journey Home, Boston: Beacon Press.

Sherry, John F., Jr. (1990), "A Sociocultural Analysis of a Midwestern American Flea Market," Journal of Consumer Research, 17 (June), 13-30.

Stoddart, Kenneth (1986), "The Presentation of Self in Everyday Life: Some Textual Strategies for `Adequate Ethnography,'" Urban Life, 15 (April), 103-121.

Wallendorf, Melanie and Russell W. Belk (1989), "Assessing Trustworthiness in Naturalistic Consumer Research," in Interpretive Consumer Research, ed. Elizabeth C. Hirschman, Provo, UT: Association for Consumer Research, 69-84.