7860-U1D1 - Describe the type of quantative research, the variables investigated, what is meant by scientific merit, ect. See details below.

12

Ethnographic Methods: Applications From Developmental

Cultural Psychology

Peggy J. Miller, Julie A. Hengst, and Su-hua Wang

Ethnographic modes of inquiry have had a long and distinguished history in the social sciences, especially in anthropology and sociology. Like all interpretive methods, ethnographic approaches are oriented to the study of meaning, but, in the case of ethnographic methods, meaning is understood to be structured by culture—that is, by collectively shared and transmitted symbols, under- standings, and ways of being. The word ethnography dates from the emergence of anthropology as a discipline in the late-19th century. Anthropologists coined the term to describe monograph-length descriptions of people who were ethnoi or “other” (Erickson, 1986). Intrigued by distant cultures, many of which were European colonies, they traveled to far-off outposts to see them first-hand. Ethnographic methods evolved out of these cross-cultural encounters. The goal was to understand a particular culture on its own terms, to represent the meaning of actions and institutions from “the native’s point of view” (Malinow- ski, 1922). In his ground-breaking study of the Trobriand Islanders, Malinowski combined long-term participant-observation with in-depth interviewing, the two hallmarks of modern ethnography (Erickson, 1986). Ethnographic methods remain the privileged mode of inquiry in cultural anthropology and have become increasingly important in the fields of education and communication.

In psychology, where the prevailing orientation has been positivist, propos- als for a “second” or “cultural” psychology were part of the intellectual landscape from its inception as a discipline (Cahan & White, 1992; Jahoda, 1989). Wilhelm Wundt wrote extensively on cultural psychology and was “captivated by the ethnographic material he pursued so tirelessly” (Jahoda, 1993, p. 181). Despite this early history, modern psychology has excluded ethnographic approaches from its methodological repertoire. Even community psychology, with its com- mitments to contextual understandings and to collaborative models of research, has marginalized ethnographic methods (Stewart, 2000).

However, the recent renewal of interest in cultural psychology makes it timely to consider the nature of ethnographic methods, given the affinity of ethnography for problems in cultural psychology. Although psychologists from

219

http://dx.doi.org/10.1037/10595-012 Qualitative Research in Psychology: Expanding Perspectives in Methodology and Design, edited by P. M. Camic, J. E. Rhodes, and L. Yardley Copyright © 2003 American Psychological Association. All rights reserved.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

220 MILLER ET AL.

many corners of the discipline have contributed to recreating a cultural psychol- ogy (e.g., Bruner, 1990; Cohen, Nisbett, Bowdle, & Schwarz, 1996; Gergen, 1985; Markus & Kitayama, 1991), scholars of child development have played a particularly important role, and some have written extensively about ethno- graphic methods. In addition, there are several traditions of interdisciplinary study of child development in which ethnographic methods have been privi- leged. For these reasons, this chapter will draw heavily on developmental questions to illustrate the assumptions and aims of ethnographic methods.

But before we turn to specific instantiations, it is necessary to provide additional background about the nature of ethnographic methods. First, it is important to stress that ethnographic modes of inquiry do not constitute a single, unified perspective or set of methods. Rather, here, as in qualitative inquiry in general, diversity reigns. This is amply illustrated in Denzin and Lincoln’s (1994) Handbook of Qualitative Research. Denzin and Lincoln’s intro- duction to their volume provides an excellent survey of the diversity of interpre- tive paradigms. They see this diversity as anchored in positivism, on the one extreme, and postmodernism, on the other. The naive realist position—there is a reality out there that can be studied objectively and understood—is coun- tered by the postmodern, poststructuralist position of radical doubt. Articulat- ing the latter position, Denzin (1996) wrote, “There can never be a final, accu- rate representation of what was meant or said, only different textual representations of different experiences” (p. 132).

Between these two extremes are the middle-ground positions of postpositiv- ism and constructivism. According to Denzin and Lincoln (1994), postpositivism rests on the assumption that reality can never be fully apprehended, only approximated. Postpositivists use multiple methods to capture as much of reality as possible; emphasize the discovery and verification of theories; and apply traditional evaluative criteria, such as validity. Denzin and Lincoln (1994) defined constructivism as involving “a relativist ontology (there are multiple realities), a subjectivist epistemology (knower and subject create un- derstandings), and a naturalistic (in the natural world) set of methodological procedures” (p. 13). Evaluative criteria include trustworthiness, credibility, and confirmability.

Some Examples of Problems for Study

Denzin and Lincoln emphasized that these various positions are realized within particular disciplinary traditions that inflect them in distinctive ways and that each researcher enters the research process from the vantage point of his or her particular interpretive community, with its unique history of research practices. The interpretive community to which we belong is an interdisciplin- ary community that has tried to bring together culture and children into a culture-sensitive understanding of child development. For the most part, this community has drawn on the middle-ground positions of postpositivism and constructivism (e.g., Gaskins, 1994; Goncu, 1999; Grau & Walsh, 1998; Jessor, Colby, & Shweder, 1996; Rizzo, Corsaro, & Bates, 1992; Shweder et al., 1998).

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 221

Consider, for example, Gaskins, Miller, and Corsaro’s (1992) framing of a set of papers pertaining to children’s socialization, one of the fundamental problems in developmental cultural psychology. Gaskins et al. advocated an interpretive approach that views reality as socially constructed; recognizes that the complex relationship between the researcher and the participants is part of the research question; and defines knowledge as understanding that makes sense to the actors themselves in terms of collectively shared interpretive frameworks, a criterion that privileges the actor’s point of view. This approach makes sense given the kinds of problems that this community of scholars has identified as central to their interests, problems that rest on the premise that all children grow up to be cultural beings. This characteristic is unique to our species and is perhaps the most important reason why human beings experience a prolonged period of immaturity (Bruner, 1972). The process of human develop- ment is thus inextricably bound to the process of enculturation, of orienting oneself within systems of meaning.

But, as Gaskins et al. (1992) pointed out, no child orients him- or herself within culture in general. Rather, each child navigates a specific culture, with a specific set of beliefs, practices, and interpretive frameworks. The process of becoming a participant in a culture is therefore enabling and limiting at the same time. Socialization, the universal process of becoming a participant in a culture, cannot be understood except by studying enculturation, the process of meaning creation in particular cultures (Mead, 1963).

Thus, the fundamental developmental question from this perspective is how do children come to invest cultural resources with meaning? Born into a world of already existing traditions and semiotic systems, children use their growing interpretive abilities to participate in cultural practices. This process is constructive and it is necessarily individual and collective. It is individual in that each child creates personal meaning out of the particular, necessarily limited set of resources to which he or she is exposed. It is collective in that these resources were created by previous generations and are made available to the child by other people. By participating with caregivers and peers in day- by-day encounters with cultural resources, children shape their own develop- mental experiences while at the same time contributing to the production of social order (Cook-Gumperz & Corsaro, 1986).

No one has probed a child’s meaning-making process more profoundly than Jean Briggs in her book, Inuit Morality Play (1998). Offspring of Never in Anger (J. Briggs, 1970), a classic of psychological anthropology, and informed by three decades of work with the Inuit, this study focuses on a single three-year-old child, Chubby Maata, as she engages a distinctive kind of emotional drama that is common to many Inuit families. J. Briggs sees culture as a “ ‘bag of ingredients’ actively used by individuals in creating and maintaining their social-cognitive worlds” (p. 14). This view allows her to realize that she cannot provide a full interpretation of the meanings that Chubby Maata is making because every fragment of data “explodes with potential meanings” (p. 20). But it is not only the witnessing ethnographer but the child herself who has to live with this ambiguity. Chubby Maata is making educated guesses, based on her past and present apprehension of the patterns in her own and other people’s words and actions. The ethnographer’s task is to follow the child. She is making

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

222 MILLER ET AL.

educated guesses about Chubby Maata’s educated guesses. The resulting eth- nography is “a cloth full of holes, the very sort of cloth that Chubby Maata herself was weaving” (p. 20).

The metaphor of a cloth full of holes is compatible with Howard Becker’s (1996) understanding of a key interpretive challenge. Operating out of the Chicago school of sociology, Becker focuses not on geographically distant cul- tures but on poor urban neighborhoods, medical schools, the art world, and other contexts that are nearer at hand. He says that people—he is speaking of adults, not children—are “not sure what things do mean: they make vague and woolly interpretations of events and people” (p. 60). The implication for ethnographers is that we should respect people’s confusion and indecision and not represent their meanings as more coherent or stable than they are.

The general problem of how children make meaning out of cultural resources implies several questions: What exactly is happening here? That is, what kinds of activities are these children and their companions engaging in? What are the folk theories—informal, local belief systems about children, child-rearing, and development—that inform and rationalize their activities? What are the larger contexts and activities in which these activities are embedded? To some social scientists, these will seem like uninteresting questions, inviting “mere” descrip- tion. But as Becker (1996) stressed, it is all too easy to think we know what people are up to. He cautioned, “Don’t make up what you could find out” (p. 59).

Play provides an excellent example of how these general questions have been applied in a specific research arena. In the past decade, play has inspired several substantial ethnographic studies in different parts of the world. Schol- ars have asked questions about the types of play that occur under everyday conditions (e.g., pretend play, exploratory play, teasing), about the folk theories that parents hold about the nature of children, of development, and of play itself (e.g., play develops naturally to children vs. play must be taught), and about the larger contexts and activities in which play is embedded (e.g., do children contribute to the family’s livelihood, and if so, how and from what age, and how much time does this leave for play? Gaskins, 1996; Goldman, 1998; Goncu, 1999; Lancy, 1996; Taylor & Carlson, 2000).

These studies have produced findings that challenge fundamental assump- tions about the nature of play (Miller, 2001). They show that play is constituted differently within and across cultures: communities vary in the types of play; the time, space, and personnel available for play; whether play is valued by adults; what role, if any, play is seen to have in children’s development; and the kinds of imaginative resources that are drawn on for play. These findings challenge developmentalists to revise our assumption that pretend play belongs to a single ontological category. When viewed from the perspective of this or that local meaning system, pretend play emerges as a blessed spiritual encoun- ter, demon possession, deceit, or self-indulgent idleness.

Ethnographic Methods: An Overview

Ethnographic research involves taking up a rigorous program of scientific in- quiry marked by repeated and varied observations and data collection; detailed

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 223

recordings of, and reactions to, such observations; a skeptical stance by the researcher that forces as many questions from the continuous interpretation of the data as it provides answers; and the presentation of ongoing interpreta- tions to the larger scientific community. Despite the diversity, common issues and practices cut across ethnographic research, whether conducted in the origi- nating discipline of anthropology for the purposes of documenting whole cul- tures or conducted by researchers addressing a diversity of questions across multiple disciplines. In this section we first address characteristics common to ethnographic methods, then briefly outline four key phases in ethnographic research. For more detailed discussions of how to conduct ethnographic re- search see Agar (1980), Erickson (1986), Hymes (1982), and Wolcott (1995).

Characteristics of Ethnographic Inquiry

One important characteristic of ethnographic methods is the sustained and engaged nature of data collection. “Classic” ethnographic studies within anthro- pology focus on cultures “foreign” to the researcher and, as a consequence, fieldwork necessarily includes time for the researcher to become familiar with, and learn to navigate within, unfamiliar physical, social, and communicative environments (e.g., Basso, 1996; C. L. Briggs, 1986; J. Briggs, 1970; Schieffelin, 1990). As ethnographic methods have been taken up by researchers in other disciplines, such as education and psychology, and applied to problems closer to “home,” the researcher often enters a research site where he or she has already spent time and is acquainted with “local” linguistic, social, or institu- tional histories and practices (e.g., Baym, 2000; Denzin, 1993; Giorgio, 1999; Heath, 1983; Prior, 1998; Wolf & Heath, 1992). In such cases, time “in the field” may be shorter as the researcher is able to draw more heavily on personal experiences and communicative practices in customizing data collection. In either case, to penetrate participants’ meaning systems, ethnographers must familiarize themselves with the participants’ community—the physical and institutional settings in which they live, the daily routines that they and their companions follow, the beliefs that guide their actions, and the linguistic and other semiotic systems that mediate all of these contexts and activities.

Through such sustained community contact, researchers necessarily be- come deeply engaged in the lives, practices, celebrations, and problems of their participants. In remote and isolated sites, the very survival of the researcher may depend on the strengths of the relationships the researcher has been able to forge and the goodwill of the community members under study (e.g., J. Briggs, 1970; Gottlieb & Graham, 1993). Even when life and limb are not at stake, the research itself is shaped and strengthened by the willingness of individuals to participate in the researcher’s project. Much has been written about the complexities of researcher–participant relationships in the interpre- tive process (see Behar, 1993; Scheper-Hughes, 1992; Wolcott, 1995; Wolf, 1992). In fact, Engstrom (1996) argued that one way to measure the validity and generalizability of research findings is to look for successful collaborations between the researchers and the participants. In ethnographic work, research- ers often find that because of their relationships with participants and their

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

224 MILLER ET AL.

developing emic understandings, they are in a unique position to help speak across cultures on behalf of the group being studied (e.g., Basso, 1996; Philips, 1983) and to help identify avenues of change that support community goals (e.g., Engstrom, 1996). These opportunities for personal, social, and political intervention make ethnographic research an attractive choice for action re- search traditions in education (e.g., Cochran-Smith & Lytle, 1993) and commu- nity psychology (e.g., Stewart, 2000).

Ethnographic methods also carry with them an implicit multicultural per- spective, a perspective that is often made explicit within particular research programs (e.g., Miller, Fung, & Mintz, 1996; Rogoff, Mistry, Goncu, & Mosier, 1993). In attempting to apprehend local meanings, ethnographers try not to mistake their own deeply taken-for-granted, culturally saturated understand- ings for those of the study participants—a challenge that is never fully met. To anticipate an example that will come later in the chapter, if the goal is to appreciate the interpretive frameworks of parents from a particular American community, and the ethnographer is Taiwanese, then the process of bringing these parents’ (American) frameworks into focus will also expose the ethnogra- pher’s own (Taiwanese) frameworks. Thus, even when ethnographers study a single cultural case, they aim for double vision at least. In fact, American parents and Taiwanese ethnographers belong to multiple communities and are likely to live and breathe meanings that flow within and across multiple cul- tures. This does not mean that cultural boundaries have no reality, but it does make a mockery of the idea that cultural boundaries can be neatly drawn in this increasingly globalized world.

Another characteristic of ethnographic inquiry is that data collection and analyses are both microscopic and holistic (Gaskins et al., 1992). Focusing on the details of particular participants and practices, ethnographic methods capture unanticipated nuances and variations of human interaction. However, Geertz (1973) argued that detailed description of behavior alone, what he calls “thin description,” is not sufficient to recoup meaning. Instead, ethnographers engage in what Geertz (1973) termed “thick description.” To ensure that their understandings are culturally valid, ethnographers ground their interpreta- tions of cultural events in an accumulation of specific details from the events of everyday life and from the participants’ reflections on those events. It is in this way that ethnographers approach broad interpretations “from the direction of exceedingly extended acquaintances with extremely small matters” (Geertz, 1973, p. 21). By way of illustrating the distinction between “thin” and “thick” description, Geertz (1973) borrowed Ryle’s example of two boys who are “rapidly contracting the eyelids of their right eyes” (p. 6). Are they blinking, winking, parodying a wink, faking a wink, practicing a wink? It is impossible to say without understanding the multiple embedded contexts in which these actions took place and the socially established communicative code that renders them intelligible. Thus, it is necessary not only to examine actions microscopically but also to contextualize them in a more holistic sense to successfully describe an event as it was understood by the actors themselves.

Finally, ethnographic inquiry is a dynamic process marked by generative and self-corrective methodologies (Gaskins et al., 1992). Successful researchers

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 225

need to be flexible from the beginning, prepared to revise or discard initial research questions and adjust data collection procedures as they position them- selves physically and socially in the research site. J. Briggs (1970), for example, set out to study shamans among the Inuit, only to discover that shamans no longer existed in the community she had entered. When Miller, Sandel, Liang, and Fung (2001) formulated their research questions about the role of personal storytelling in Longwood, hell-raising stories were not on their list; the parents in this community brought such stories to their attention. In addition, research- ers must be open to learning locally appropriate ways to ask questions and hold interviews (C. L. Briggs, 1986); they must develop effective ways to present their research project and their role as researcher to the participants, a problem that is especially complex when the participants are children (Corsaro, 1985, 1988); and they must learn to situate themselves physically and socially in ways that allow them to observe the phenomena of interest (Ochs, 1988). Often, such negotiations include a willingness on the part of the researcher to accept the interactions that are offered and to look for new ways to augment data collection (Prior, 1998).

The generative and self-correcting nature of ethnographic inquiry is also evident during data analysis and writing. The interpretive process, guided by the notion of cultural validity, is theory-generating. The goal is to provide a deeper understanding of the multiple perspectives that are operating in all human interactions. Therefore, categories used in analysis are not predeter- mined but are developed through a continual process of iterative division, classification, and evaluation (Bloom, 1974; Strauss, 1987). The researcher begins with a tentative descriptive framework—what Pike (1967) called an etic classification—often gleaned from other data sets or theoretical positions, and proceeds to test that framework through successive passes through the data. The outcome of this self-corrective process of constant comparison is an emic classification (Pike, 1967) that captures the patterns in the participants’ meanings. In addition, deepening interpretations of the data emerge when researchers revisit earlier work. For example, accounts written early in a research program are necessarily expanded in later accounts as researchers combine existing data with new data. (This process will be described more in the latter half of this chapter.) In other cases, researchers apply their evolving perspectives to a reinterpretation of earlier work. When Wolf (1992) reexamined her 30-year-old field notes concerning the case of a young Taiwanese mother who suddenly began behaving in a decidedly aberrant manner, she was dissati- sfied with her earlier account. In an effort to better display the multiple perspec- tives of participants and researcher, she ended up producing three separate accounts of the same incident. (See J. Briggs, 1998, for an excellent example in the same vein.)

Phases of Ethnographic Research

Despite the flexibility inherent in ethnographic research practices, the research process generally unfolds in a series of phases.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

226 MILLER ET AL.

DEVELOPING QUESTIONS AND GAINING ACCESS. The ethnographer begins the research process by formulating a problem for study, drawing on previous scholarship, and learning as much as possible about the particular community or institution in which the study will be conducted. Any previous ethnographic work and other formal or informal sources of information about the same community are invaluable in allowing the researcher to hone the initial re- search questions, anticipate field conditions, and design an approach that will best address the research questions. It is in this phase that the researcher makes initial decisions about what in psychology is referred to as subject sampling. Ethnographers, however, are less concerned about random sampling than they are about specifying the social positioning of the participants who agree to work with them, thereby delimiting their interpretations. The re- searcher must make preliminary contacts to obtain initial institutional permis- sion to conduct research and work to establish relationships with possible participants. In the classic case, the ethnographer enters the research site as an outsider, and the task of gaining access to particular groups or institutions may take a great deal of patience and interpersonal skill. In our own work, which spans several working-class and middle-class communities in the United States and Taiwan, we have found that doors open much more rapidly if the ethnographer has a trusted associate in the community.

The importance of the process of negotiating access to a research site cannot be overstated. The physical and social positioning the researcher is able to establish and maintain within the community of study critically shapes the entire research enterprise. Nor is this a task that applies only to the initial phase of fieldwork. Ethnographer–participant relationships must be renegoti- ated throughout the course of study, and this requires ongoing documentation and reflection. In other words, this relationship becomes an object of study in its own right, adding to the broader research questions. The epistemological assumptions outlined earlier for constructivist ethnographies imply that the knowledge that is gained through ethnographic inquiry will be conditioned by the ethnographer’s positioning in the local scene and by the nature of the relationships that he or she is able to create with participants. For example, a female ethnographer will have access to certain kinds of contexts and infor- mants, a male ethnographer to others. An ethnographer who has connections to cultural elites will have access to different perspectives than an ethnographer who has connections to the poor. Each ethnographer will come to an understand- ing that is inevitably partial. The rigor of this approach lies partly in delineat- ing that partiality, which itself contains clues as to how local meanings are constructed.

COLLECTING AND MANAGING DATA. Ethnographic research is known for pro- ducing copious amounts of data. Learning to direct data collection and organize data for ongoing interpretation are daunting tasks for novice ethnographers. The bulk of the data collection occurs during fieldwork as the researcher care- fully compiles detailed records of research-related activities and his or her initial reactions and interpretations (Wolcott, 1995). Such documentation takes many forms, including field notes, interviews, indirect observations, and artifacts.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 227

Field notes (Emerson, Fretz, & Shaw, 1995) are written descriptions and reflections about the participant–observation. Most researchers make notes in their logs as frequently as possible, jotting down short notes “on the fly” and more detailed notes later. Field notes may contain physical descriptions of the site (augmented by photographs, maps, sketches, etc.), descriptions of daily routines of the participants (augmented by work schedules, seasonal activities, etc.), and detailed descriptions of observed interactions and participant inter- views. To facilitate such detailed record keeping, researchers routinely make use of any technologies appropriate to the site (e.g., audiorecording, video- recording, etc.).

Interviews may be conducted with individuals or groups, and the general organization of the interview is usually planned in advance. However, specific interview techniques depend on the nature of the community and research questions (see C. L. Briggs, 1986; Mishler, 1986), as will be illustrated in the final section of this chapter. Whenever possible, interviews are audiorecorded and transcribed for analysis. In addition to more formal interviews, ethnogra- phers find opportunities to insert their questions into casual conversation.

The researcher may also collect indirect observations by working with participant–collaborators, especially in cases where the phenomenon of interest occurs infrequently or only with limited audiences. In such cases, research assistants are taught to take notes or make recordings, ask questions, and make specific observations to address the research questions.

Finally, collecting artifacts about the community, the participants, the physical setting, the institution, and the practices may also be a critical form of data collection. Which artifacts are appropriate to collect will depend on the goals of the research project but may include maps, newspapers, legal documents, popular texts, diaries, letters, tools, and so forth. In addition, the researcher will need to make notes about the circumstances and reasons for obtaining each artifact.

INTERPRETING AND ANALYZING DATA. Data analysis begins early in the re- search process and continues throughout what is often a long program of ethnographic inquiry, with new research projects building on previous ones. In fact, effective fieldwork requires the direction such ongoing interpretation provides (e.g., who to interview next, what questions to ask, what activities to observe, etc.). Fitting with ethnography’s general goal of developing under- standings consistent with the meaning-making practices of the community being studied, the interpretive process is primarily inductive in nature, and coding systems and categories evolve from a continual comparison of the grow- ing data set (see Strauss, 1987; Wolcott, 1994). Novice ethnographers who are familiar with preset coding systems applied intact to complete data sets often find the evolving and inductive coding practices of ethnographic work difficult to manage.

The specific nature of the coding systems and types of analyses vary widely, depending on the goals of the specific research project and the disciplinary training and theoretical inclinations of the specific researchers. Early analysis is often focused on developing categories that account for the diversity and breadth of the data being collected. As the analysis progresses, categories are

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

228 MILLER ET AL.

filled in with more depth, and interconnections within and across categories are analyzed. Particular examples may be extracted for in-depth analysis, as we illustrate in the next section of this chapter.

The credibility of the findings is in part a result of a well-documented and systematic analysis of the data. Although not focused on reliability in a traditional sense, ethnographic researchers are very concerned about present- ing “accurate” or “valid” representations of the phenomena in question from the participants’ perspective—that is, getting the “story” right. One way trust- worthiness of interpretations is achieved is through comparing and integrating data from different sources, a process often referred to as triangulation (Denzin, 1978; Marshall & Rossman, 1995; Rizzo et al., 1992). In addition, researchers will discuss their interpretations with participant–collaborators, seeking both contesting and supportive responses, which will allow them to thicken their analysis or reinterpret their data.

Though uncommon, it is possible for ethnographic studies to blend quanti- tative coding systems with qualitative coding strategies (Gaskins, 1994; Rizzo et al., 1992). For example, qualitative analysis of interview data can be used to illuminate the meaning of survey data obtained via conventional quantitative methods. As well, emic descriptions derived from fieldwork can be used to construct interview protocols or questionnaires that yield quantifiable results. However, it is important to dispel the myth that qualitative analyses are valuable only insofar as they can be converted into quantitative analyses (Hymes, 1982).

WRITING. One way of stating a guiding principle for ethnographic writing is “write early, write often.” This process begins with the researcher’s log and field notes and continues through the construction of published accounts. However, it is in the culmination of writing up and disseminating ethnographic accounts that the fieldwork of specific research projects is connected with broader programs of scientific inquiry. Like all research projects, “Fieldwork is validated only through the requisite reporting that results from it” (Wolcott, 1995, p. 66). Written accounts of ethnographic work take many forms, but typically the ongoing analyses of the data obtained during an ethnographic research project yield multiple publications.

In recent decades, the textual practices of ethnographers, as well as the appropriateness of various types of ethnographic accounts, have been at the center of intense debates (see Behar & Gordon, 1995; Clifford & Marcus, 1986; Geertz, 1988; Richardson, 1997; Van Maanan, 1988; Wolf, 1992). As ethnogra- phers have grappled with poststructuralist views of culture, issues of represen- tation have become a critical consideration. These issues include how to repre- sent “others” or let “others” represent themselves; how to represent the researcher’s roles, limitations, and biases within the research site; how to appropriately blend multiple, often contesting, perspectives; and how to respect the diversity and complexity of cultural practices. This has led to a diversifica- tion in published research accounts as ethnographers have experimented with issues of representation in ethnographic writing (see Behar, 1993; Sereme- takis, 1991).

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 229

The Nonnarration of Children’s Transgressions: An Interpretive Puzzle

In this section we address an interpretive puzzle from our own work by way of illustrating how ethnographers proceed in analyzing and interpreting data. The puzzle arose from a program of ethnographic research that is comparative in design, involving middle-class Taiwanese families in Taipei, Taiwan, and middle-class European American families in Longwood (a pseudonym), a neigh- borhood in Chicago (Fung, 1999; Miller, Hengst, Alexander, & Sperry, 2000; Miller, Wiley, Fung, & Liang, 1997; Miller et al., 2001).

How the Puzzle Arose

To set the stage for this puzzle, it is necessary to present some background about earlier phases of this research. The initial goal of the project was to investigate how personal storytelling is used to socialize young children within the family context. Do these families engage in personal storytelling—telling oral stories about one’s past experiences—in ways that involve young children? If so, how is personal storytelling defined and practiced with young children? We were particularly interested in the modes of participation and interpretive strategies that families used in narrating young children’s past experiences. Note that all of these questions are versions of the “What exactly is happening here?” question.

Through participant–observation and video recording of ordinary family interaction, we discovered that stories involving the focal child (2 years, 6 months, of age) as protagonist occurred at remarkably similar rates (about four per hour on average) in the Taipei and Longwood families. In addition, in both cases, stories were conarrated with young children, and stories were told about the child in the child’s presence.

These similarities coexisted with a striking difference in the content and manner of narration. The Taipei mothers were much more likely than their Longwood counterparts to treat children’s past transgressions as a didactic resource, as opportunities to teach young children the difference between right and wrong. Transgressions were talked of openly in front of siblings, research- ers, and guests; explicitly, often in strong language; and at length. Rarely was the language mitigated, although subtle nonverbal cues were used to signal humor. The ethnographer was treated as a judging witness to the child’s mis- deeds. By contrast, the Longwood families operated with a distinct self-favor- ability bias in narrating young children’s experiences. They rarely told stories about the child’s past transgressions. When they did so, they managed to portray the child in a positive light despite his or her misdeed, casting the researcher as an appreciative audience to the child’s exploits.

Defining the Puzzle

The puzzle, then, is this: How can we make sense of the Longwood practice of not narrating children’s transgressions? (The complementary puzzle from the

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

230 MILLER ET AL.

Taipei data is: How can we make sense of the narrative practice of foreground- ing children’s past transgressions? See Fung, 1999; Miller et al., 1996; for discussions of this puzzle.) From a Taiwanese perspective, this is baffling. This looks irresponsible. What are these Americans up to when they downplay or mitigate young children’s transgressions or strike them entirely from the narrative record?

Notice that this puzzle has been defined, in part, by the Taiwanese compari- son, which casts the American practices in relief. Although many ethnographic studies focus on a single cultural case, there is usually an implicit comparative perspective that informs what the ethnographer is able to identify as interesting problems. The inclination to emphasize children’s strengths is so common among middle-class Americans that it is next to invisible. We might have overlooked this puzzle were it not for the contrast with the Taiwanese findings. In other words, having a comparative vantage point on one’s own cultural ways is often crucial in rendering the familiar strange (Erickson, 1986; Ochs & Schieffelin, 1984).

Notice too that this puzzle arose out of careful documentation of a pattern that occurred in everyday family interaction, a pattern that emerged in re- sponse to our initial questions. In observation after observation we witnessed and participated in a baseline of personal storytelling activity in which chil- dren’s transgressions were rarely narrated. It is important to emphasize that to document that baseline we followed a complex set of analytical coding steps that, for lack of space, can only be briefly mentioned: We devised a descriptive code for defining personal storytelling in the two cultural cases, applied the code to the video-recorded observations for each family, transcribed the full set of personal storytelling events that were identified for each family, and devised and applied additional codes for describing the content and manner of narration (see Miller et al., 1997). Some ethnographers might have described the resulting baseline pattern as “routine,” without attaching any numbers; our preference was to count the stories that occurred and to calculate the proportion that involved child transgressions. In both cases, whether or not numbers are attached, there is a recognition that specific examples of interac- tion are interpretable only against a documented baseline of ordinary activity.

Microanalysis of a Strategically Chosen Example

Ethnographers often use the interpretive strategy of lifting out an example for microlevel analysis as a way of deepening their understanding of the phenome- non in question. This strategy illustrates the “microscopic and holistic” feature of ethnographic research that we discussed earlier in which an event is de- scribed in minute detail as a way of illuminating the meaning of some larger pattern. In the following analysis we illustrate this strategy, borrowing from an analysis presented more fully in Miller et al. (1996). Although ethnographers often choose “typical” examples to work with, Miller et al. chose a story that was exceptional within the baseline distribution. They focused on a rare instance in which a Longwood family not only told a story about the focal child’s transgres- sion but structured the story so as to establish the child’s transgression as the

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 231

point of the story, thereby mimicking the Taiwanese practice. This exceptional story was important analytically because it allowed us to disentangle two possible interpretations. Perhaps Longwood families narrated child transgres- sions in the same didactic manner as their Taipei counterparts but did so far less frequently. Or perhaps they narrated child transgressions in a qualitatively different manner on those rare occasions when they narrated them at all.

The story in question actually involved two transgressions. As narrated by the mother—in collaboration with Mollie (2 years, 6 months), the researcher, and Mollie’s older sister—Mollie first wrote on the wall and then tried to evade responsibility for her misdeed by falsely accusing her sister.

Mother: [To child] Did you tell Judy [the researcher] what you wrote on the dining room wall with? Child: Ah . . . key. Researcher: [To child] You wrote on the dining room wall? Mother: With a key, not even a pencil. Researcher: [To mother] You must have loved that. Mother: A key, the front end of that key. Sister: And behind a living room chair. Mother: I was sort of napping in there and I saw this and I thought it was a pencil. And I woke up and said [whispering], “Mol, you didn’t write on Mommy’s wall with a pencil, did you?” Oh, she was so relieved, she said, “No! Me no use pencil, me use key!” and I was like, “OH GOD! Not a key!” And she said, “No, no, ME no use key, Mom. Kara [her sister] use key,” and then I was even more upset. Sister: I didn’t even see her do it! Mother: But it’s so funny. You look at her and she’s like, “I didn’t use pencil.” Researcher: So, I’m in the clear. Mother: Oh, yeah. Sister: I didn’t even see her do it. I was at school.

In this excerpt, Mollie’s mother prompts her to confess her wrongdoing to the researcher. Mollie complies, and the researcher invites additional response. Several turns ensue in which the mother emphasizes that Mollie used a key to write on the wall, the researcher aligns herself with the mother through an ironic expression (“You must have loved that!”), and Mollie’s older sister— whom Mollie falsely blamed—contributes further information about the inci- dent, emphasizing that she was not even there when the incident happened. Having established Mollie’s wrongdoing by eliciting supporting accounts from the parties involved, the mother then explains more fully to the researcher what happened. That is, she tells a story about Mollie, referring to her in the third-person, in which she situates the wall-writing incident within the events that preceded and followed it. She explains that she was napping when the misdeed occurred. Her dawning realization that Mollie wrote on the wall while she napped is recreated through the mounting suspense of parallel, but increas- ingly damaging, admissions by the child. The mother represents Mollie as trying to mitigate her responsibility for wrongdoing, first by explaining that she used a key and not a pencil and second by falsely accusing her sister. The humor lies in the fact that the child’s inept and increasingly transparent

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

232 MILLER ET AL.

attempts to explain away her misdeeds have exactly the opposite effect. Her mother’s subsequent comment, “But it’s so funny,” explicitly frames the narra- tion as nonserious. Note also that although the mother says that she was “even more upset” by the child’s lie than by the misdeed that occasioned it, there is no further mention of the more serious transgression. Also, the interaction that preceded the story about the child, including the elicited confession from Mollie, includes no mention of her false accusation.

Miller et al. (1996) compared this story with a Taiwanese story that is strikingly similar in content and structure: Angu, like Mollie, wrote on the wall and then tried to shift blame to someone else. Although it is beyond the scope of this chapter to present the parallel microanalysis of Angu’s story, it is important to summarize some of the key differences in how Angu’s misdeeds were narrated: Angu’s caregiver developed the story at far greater length; foregrounded the more serious transgression of falsely accusing another person; shamed the child for her misdeeds; and framed the story as serious. What, then, do these twin microanalyses tell us? Even in the rare instance in which an American family constructed a story around the child’s transgression, creat- ing a story that resembled a Taiwanese story in content and structure, close analysis revealed that it conveyed a qualitatively different interpretation of the child and of her experience. Instead of creating an opportunity for moral education and remediation, Mollie’s mother developed the amusing dimensions of the incident. She created a charming and naive mischief maker, not a trans- gressor.

The Puzzle Partially Unraveled

To summarize, we made several analytical moves in attempting to understand this interpretive puzzle. First, we established, through participant–observation and transcription of video-recorded home observations, that personal storytell- ing occurred routinely in Longwood and Taipei families. Second, we documented a contrasting pattern in the content and structure of personal storytelling such that Longwood families, compared with Taipei families, were far less likely to narrate young children’s transgressions. In other words, the initial analytical moves involved documenting an observed pattern in ordinary family interac- tion, drawing on comparative observations to aid in the identification of that pattern. The third analytical move involved microanalysis of a particular story that was strategically chosen because of its outlier status in the baseline distri- bution of storytelling. This microanalysis deepened our understanding of the meaning of the baseline pattern by zeroing in on a violation of that baseline. Although the story, in this exceptional instance, was “about” Mollie’s misdeeds, it was also “about” how funny those misdeeds were. This series of analyses, thus, supports the following rendering of Longwood parents’ perspective on young children’s misdeeds: best to leave them un-narrated; if one happens to slip through, background it, mitigate it, laugh about it, or in some way undercut its importance. These analyses suggest that young children’s wrongdoing has a qualitatively different meaning for Longwood parents, compared with Taipei parents. Apparently, young children’s wrongdoing is a somewhat delicate mat- ter for Longwood parents.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 233

Although this series of analyses allowed us to deepen our understanding of the nonnarration of children’s transgressions, it is important to emphasize that no analysis is the final analysis. In the constructivist ethnographic ap- proach that we advocate, each analysis leads seamlessly to a more pointed set of questions; thus, the boundary between one research report and the next is somewhat arbitrary. In the case at hand, our interpretations were based en- tirely on observations of the families’ enactments of personal storytelling, not on their expressed ideas about storytelling. To understand these practices more fully, we needed to examine parents’ reflections on child-rearing. What was at stake for Longwood parents when they engaged in these narrative practices? What kinds of ideas were informing their child-rearing?

Sequel: Toward Additional Unraveling

To pursue these questions, Mintz (1999) inquired into parents’ belief systems about child-rearing, drawing on interviews, the other stock-in-trade ethno- graphic tool. In-depth interviews with the Longwood mothers revealed that promoting their young children’s self-esteem was a matter of the first impor- tance to them. They believed that self-esteem provides the foundation for happi- ness, inner strength, and moral autonomy. They spoke of the devastating consequences of low self-esteem on children’s psychological functioning and success in the world. They tried to support children’s self-esteem by praising them, emphasizing their strengths, and avoiding invidious comparisons. When discussing discipline, they made a distinction between “being bad” and “doing bad things,” contrasting their child-rearing practices with those of their own parents. They believed that discipline had to be handled with care, lest it undermine children’s self-esteem. Like the mothers in Harwood, Miller, and Irizarry (1995), they sought a balance between cultivating self-esteem and respect for others.

These findings shed further light on the Longwood practice of downplaying, laughing about, or simply not narrating young children’s transgressions by suggesting that a collective commitment to the goal of supporting children’s self-esteem may underlie this practice. Longwood families’ reluctance to dwell on young children’s past misdeeds is intelligible within a folk theory that valorizes self-esteem, linking it to a host of psychological goods, just as Taipei families’ routine narration of child’s transgressions is intelligible within a folk theory that is distinctly Confucian, valorizing moral instruction and “opportu- nity education” (Fung, 1999; Miller et al., 1996; Miller et al., 1997).

Self-Esteem As Folk Theory

In this final section of the chapter, we present a research case by way of illustrating the process of conducting ethnographic research (see Miller, Wang, Sandel, & Cho, 2002, for a detailed report of the methods and results of this study). This case emerged directly out of the line of inquiry described in the preceding section. Our findings from Longwood led us to be interested in Ameri-

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

234 MILLER ET AL.

can folk theories of self-esteem. By parental folk theories we mean parents’ informal, culturally organized understandings about children, child-rearing, and development. These understandings vary within and across cultures, in- forming and rationalizing child-rearing practices (Bruner, 1990; Goodnow & Collins, 1990; Harkness & Super, 1996).

The Research Problem

The idea that the Longwood mothers articulated—that children’s self-esteem should be fostered because it lays the groundwork for a host of psychological strengths—is shared by many American parents, teachers, and psychologists. The ubiquity of reference to self-esteem in both scientific arenas and popular culture naturalizes self-esteem, promoting a kind of invisibleness. This invisi- bleness is supported, as well, by two striking omissions from the discourse of self-esteem. Rarely is self-esteem and its associated folk theory recognized to be a culture-specific, historically situated discourse. And rarely is the debate about self-esteem informed by the voices of parents as they reflect on these ideas in raising their children. Our study was intended to address these omissions. Its purpose was to examine the meanings and practices associated with self-esteem and the larger folk theory in which it is embedded.

Design and Research Sites

Because other cultures do not necessarily share Americans’ preoccupation with self-esteem (see Harwood et al., 1995; Heine, Lehman, Markus, & Kitaymama, 1999; Stevenson et al., 1990) and because perspectives from other cultures can help to expose the cultural specificity of self-esteem, we chose to study this problem comparatively, building on our earlier work with American and Tai- wanese families. We wanted to identify the variety of meanings American and Taiwanese caregivers associate with the idea of self-esteem and to delineate the local folk theories that contextualize this idea or that offer alternative understandings of child-rearing. Although we approached this question through participant–observation and interviewing, this question lent itself particularly well to interviewing, and thus we focus on interviewing in this brief sketch.

In choosing research sites, we considered two factors. In recognition of intracultural variability, we wanted to move beyond large urban areas. And because personal contacts facilitate fieldwork, we chose research sites where our research team had preexisting personal networks. We briefly describe some key features of the two research sites.

Chhan-chng (a pseudonym) is a small Taiwanese farming community that embodies a complex mix of old and new cultural practices. The residents speak Taiwanese and Mandarin Chinese and observe traditional religious practices, worshipping their ancestors before the family’s ancestral tablets and going to the local temples to ask for peace and prosperity. The grandparents’ generation continues to work in the fields, growing rice, sugar cane, fruits, and vegetables. Most families own a motor scooter or automobile and have access to American

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 235

and Japanese programming on cable television. Although many young families are choosing to have fewer children and some of the mothers work outside the home or even in a nearby city, the traditional three-generation household is still the norm. Like previous generations, children are not segregated from adult activity. They live in a community where homes, farms, shops, and businesses are often joined; children witness and participate in economic activ- ity and they are accustomed to seeing people come and go on a daily basis.

Centerville (a pseudonym) is a small city located in the rural midwest. Although soybeans and corn remain an important part of the county’s economic base, Centerville is best known as the home of a major university, which attracts a culturally diverse group of students. Centerville supports a remark- able number of places of worship, including two synagogues, two Buddhist sanghas, a mosque, and more than 100 Christian churches. Because Centerville is much more diverse than Chhan-chng, it is not possible to describe family life in the same sweeping terms. In some families both parents work and young children go to daycare; in others mothers are full-time housewives. Despite these differences, two-generation households are the norm. Contact with grand- parents varies widely. Some grandparents provide daily childcare; others live far away and keep in touch through telephone calls and occasional visits. Unlike their counterparts in Chhan-chng, young Centerville children do not have much access to parents’ work lives.

The Researchers and Field Entry

In contrast to many classic ethnographic studies, our research team included individuals with varying life experiences in the two cultures. All of us had lived in Centerville for extended periods of time (one to eight years). Miller has been studying American and Taiwanese families with Taiwanese collabora- tors for many years. Sandel, who speaks Mandarin and some Taiwanese, was born and raised in the United States, but his wife grew up in Chhan-chng and her parents and other relatives still reside there and treat Sandel as kin. Although Chhan-chng was unfamiliar to Wang, she was born and raised in Taiwan and is a native speaker of Taiwanese and Mandarin. Both Wang and Sandel had lived in Centerville for at least a year before we embarked on this study. Sandel has young children, which put him in contact with schools and churches.

These various personal contacts were crucial in allowing the researchers to recruit participants for the study and in easing relationships between re- searchers and participants. For example, Sandel’s mother- and father-in-law helped to explain the study to local families. Beyond the initial phase of field entry, the cultural variability within the research team enabled us to draw on multiple insider–outsider perspectives in conducting the study and interpre- ting the findings.

The Participants

At each site 16 families participated in the study. The families were chosen to be homogeneous on several demographic variables. Each family had a three-

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

236 MILLER ET AL.

year-old child, who was the focus of the questions about child-rearing. In addi- tion, the families were two-parent families who represented the more highly educated segment of their respective communities. Most of the Centerville mothers had a college degree; most of the Chhan-chng mothers had 14 years of education. The average number of children per family was two for Chhan- chng and three for Centerville families. In both research sites, mothers were either the primary caregivers or shared childcare with a grandmother or a childcare provider.

Conducting/Adapting the Interviews

The researchers talked with the mothers in their homes, using their native language (either English or Mandarin or Taiwanese). The interviews were open-ended, and content areas included child-rearing goals and values, disci- pline, strategies for promoting development, sources of child-rearing informa- tion, shame and pride, and self-esteem. The protocol was intended to provide a rough guideline for conversation. However, the researchers waited until late in the interview to ask questions about self-esteem.

Instead of thinking of interviewing as simply a matter of asking questions and listening to responses, we treated interviewing as an observable social practice that may be more or less familiar to the participants, more or less in need of adaptation to local norms. This perspective owes a great deal to Charles Briggs’s book, Learning How to Ask (1986). Building on insights from his own extensive ethnographic work, C. Briggs argued that interviews are not transparent windows into informants’ beliefs but rather communicative events, analyzable in terms of the metacommunicative features of the talk and nonver- bal action that interviewer and interviewee construct together. When these features are addressed, along with other data from local communicative rou- tines, it becomes possible to offer a more precise and well-grounded interpreta- tion than could be achieved through conventional “content” analyses. Like other ethnographers who have written about interviewing (Mishler, 1986; Wolcott, 1995), C. Briggs attends not only to what people say but to when and how they say it, what they convey nonverbally, how silence is patterned. As it applies to comparative research, this approach implies that it will often be necessary to devise different “interview” events, reflecting the different communicative norms of the communities being compared, to yield equivalently meaningful discourse.

In fact, in our study the interview event unfolded in strikingly different ways in the two communities. The Centerville participants seemed to share an interview script that included a particular kind of staging. The mother led the researcher to a table, where she and the researcher sat facing one another. When children were present, mothers would ask them to play with toys or computer games and not interrupt the interview. Although the ethnographers did not request that any special arrangements be made, the Centerville partici- pants established a self-contained time and space for the interview. In addition, although we intended that the interview protocol would be followed loosely, allowing the interviewer to pursue the mother’s interests, the researchers found

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 237

that little alteration was required. The interview script—the researcher asks a question, the interviewee responds—seemed to be taken for granted by these participants. Thus, in Centerville interviewing emerged as a familiar practice shared by both parties.

By contrast, few Chhan-chng participants set up a self-contained time and space for the interviews. Indeed, most interviews took place in the presence of more than one family member. Sometimes even a bypassing neighbor would join in the conversation. Moreover, the protocol had to be altered, as the local women were not in the habit of answering formal questions in the course of everyday life. The researchers tried to find more suitable initial topics that would put people at their ease, and they responded to the participants’ curiosity about their lives in the United States. Once the participants felt more comfort- able, the researchers retrieved other questions from memory and inserted them into the conversation in as natural a way as possible. They also discovered that talk flowed more freely when they participated in whatever domestic task was underway. In sum, the interview was not a familiar or comfortable speech event to the participants in Chhan-chng. To learn about mothers’ child-rearing beliefs, the interviewers had to adapt to local communicative norms, all but abandoning the interview format for a more conversational approach in which participants had significant control over the topics of talk, multiple speakers were accommodated, and everyone got on with the domestic work at hand.

Data Analysis

Our objective in this study was to examine the variety of meanings that Center- ville and Chhan-chng caregivers associated with self-esteem and to delineate the local folk theories that contextualize this idea or that offer alternative understandings of child-rearing. It is beyond the scope of this chapter to present the full set of data analyses. Instead, we recap the analytical moves involved in addressing a single subsidiary question: Did self-esteem figure importantly in the mothers’ folk theories of child-rearing in the two research sites?

It is necessary first to say a few words about terminology. In the American interviews, the researchers used the term “self-esteem,” but the mothers some- times used such terms as “self-confidence,” “self-respect,” and “feeling good about oneself,” and these were treated as synonyms for self-esteem. There is no term in Mandarin Chinese or Taiwanese that translates directly as “self- esteem.” However, there are two terms that approximate some of the meanings associated with self-esteem. One is zi zun xin in Mandarin or chu chun sim in Taiwanese; the literal English translation is “self-respect-heart/mind.” The second term is zi xin xin in Mandarin or chu sin sim in Taiwanese; the literal translation is “self-confidence-heart/mind.”

In ascertaining whether self-esteem figured into the mothers’ understand- ings of child-rearing, we first determined whether the mothers spontaneously mentioned self-esteem before the researcher introduced the term, coding from verbatim transcripts in the original language. We found that the majority of American mothers invoked these terms in response to a whole range of ques- tions. For example, some mothers mentioned self-esteem or self-confidence in

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

238 MILLER ET AL.

response to the initial question, “What are your goals as a parent? What do you hope for your children?” When the self-esteem questions were asked, most of these mothers talked easily and fluently about self-esteem, studding their responses with real-life examples.

By contrast, only a few of the Taiwanese mothers brought up self-respect- heart/mind or self-confidence heart/mind before the researcher’s explicit que- ries, and no one invoked these terms repeatedly. When asked directly about these terms, some of the Chhan-chng women did not seem to find these ques- tions intelligible or meaningful, and no one elaborated on her views in the detail that characterized many of the American responses. However, the Taiwanese mothers did talk at length about other child-rearing issues.

After describing how key terms were used over the course of the interview, we examined all passages in each interview in which the participant talked about self-esteem or related terms. In keeping with the concept of folk theory, which implies that parents hold a set of ideas that are conceptually related, we coded these passages in terms of the ideas that the participant linked to self-esteem. For example, every Centerville mother said that self-esteem was important to children’s development and that she actively tries to build, culti- vate, or protect her child’s self-esteem. They said that self-esteem provides an essential foundation for a wide array of psychological strengths: Children who have high self-esteem are able to learn and grow with ease; they are not afraid to achieve; they interact well with others; and their mental health is good.

The few Taiwanese mothers who spoke about self-respect-heart/mind or self-confidence-heart/mind linked these terms to strikingly different ideas. For example, one mother said that it is best for children to have “normal” self- respect-heart/mind because they will become less frustrated than those whose self-respect-heart/mind is strong. This idea contradicts the American mothers’ belief that high self-esteem allows children to keep trying in the face of failure. In short, the mothers from Chhan-chng believed that high self-respect-heart/ mind creates psychological vulnerabilities, whereas American mothers believed that high self-esteem creates psychological strengths.

This brief sketch of one set of data analyses reveal that self-esteem loomed large in the Centerville mothers’ folk theory of child-rearing but not in the Chhan-chng mothers’ folk theory. For the Centerville mothers, self-esteem served as a central organizing concept, an idea that came readily to mind when child-rearing was discussed, whether or not the researcher mentioned the term. The Taiwanese mothers’ folk theories were just as complex, but they were organized around the ideas that children grow up naturally and that parents are responsible for their moral education. The few Taiwanese mothers who talked about self-respect-heart/mind did so in ways that contradicted the Amer- ican mothers.

The ethnographic methods used in this research case strengthen the credi- bility of these findings. First, the intelligibility of the mothers’ responses was enhanced by the care that was taken to familiarize the participants with the researcher and to create a communicative event that fit local norms. If we had imposed our conception of interviewing on the Taiwanese women and they had had little to say about self-esteem, we would not have been able to interpret their omission as revealing anything about self-esteem. Second, in analyzing

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 239

the mothers’ talk we sought patterns not only in the content of their talk but in how they expressed themselves. Whether self-esteem was introduced before the researcher mentioned it, how often the term was used, and in which con- texts, where there were moments of confusion or unintelligibility—these and other metacommunicative patterns helped us to determine the place of self- esteem in the two folk theories.

In sum, this research case is offered not as a model or a recipe but as one example of how ethnographic work proceeds when addressing a problem in developmental cultural psychology. This case departs from classic ethnographic research in which the ethnographer begins as an outsider, focuses on a single culture, and operates out of an implicitly comparative framework. By contrast, our study was comparative in design and involved a research team that occupied complex insider–outsider positions from the outset. However, like classic ethno- graphic studies, this case exemplifies the systematic but flexible deployment of method that lies at the heart of ethnographic practice, a flexibility that is disciplined by the goal of understanding meaning from the perspectives of local participants. Sometimes social scientists who have grown up in positivist traditions believe that research cannot be empirical and interpretive at the same time. This case, like much contemporary ethnographic research, demon- strates that this is a misconception.

References

Agar, M. (1980). The professional stranger: An informal introduction to ethnography. San Diego, CA: Academic Press.

Basso, K. H. (1996). Wisdom sits in places: Landscape and language among the Western Apache. Albuquerque: University of New Mexico Press.

Baym, N. K. (2000). Tune in, log on: Soaps, fandom, and online community. Thousand Oaks, CA: Sage.

Becker, H. S. (1996). The epistemology of qualitative research. In R. Jessor, A. Colby, & R. A. Shweder (Eds.), Ethnography and human development (pp. 53–71). Chicago: University of Chicago Press.

Behar, R. (1993). Translated woman. Boston: Beacon. Behar, R., & Gordon, D. (Eds.). (1995). Women writing culture. Berkeley: University of Califor-

nia Press. Bloom, L. (1974). Commentary. In F. F. Schachter, K. Kirshner, B. Klips, M. Friedricks, &

K. Sanders (Eds.), Everyday preschool interpersonal speech usage: Methodological, develop- mental, and sociolinguistic studies. Monographs of the Society for Research in Child Develop- ment, 39 (2, Serial No. 156).

Briggs, C. L. (1986). Learning how to ask: A sociolinguistic appraisal of the role of the interview in social science. London: Cambridge.

Briggs, J. (1970). Never in anger: Portrait of an Eskimo family. Cambridge: Cambridge Univer- sity Press.

Briggs, J. (1998). Inuit morality play: The emotional education of a three-year-old. New Haven, CT: Yale University Press.

Bruner, J. (1972). Nature and uses of immaturity. American Psychologist, 27, 687–708. Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press. Cahan, E. D., & White, S. H. (1992). Proposals for a second psychology. American Psychologist,

47, 224–235. Clifford, J., & Marcus, G. E. (1986). Writing culture: The poetics and politics of ethnography.

Berkeley: University of California Press.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

240 MILLER ET AL.

Cochran-Smith, M., & Lytle, S. L. (1993). Inside/outside: Teacher research and knowledge. New York: Teachers College Press.

Cohen, D., Nisbett, R. E., Bowdle, B. F., & Schwarz, N. (1996). Insult, aggression, and the southern culture of honor: An “experimental ethnography.” Journal of Personality & Social Psychology, 70, 945–960.

Cook-Gumperz, J., & Corsaro, W. A. (1986). Introduction. In J. Cook-Gumperz, W. A. Corsaro, & J. Streeck (Eds.), Children’s worlds and children’s language (pp. 1–11). Berlin: Mouton de Gruyter.

Corsaro, W. A. (1985). Friendship and peer culture in the early years. Norwood, NJ: Ablex. Corsaro, W. A. (1988). Routines in the peer culture of American and Italian nursery school children.

Sociology of Education, 61, 1–14. Denzin, N. K. (1978). The research act. Chicago: Aldine. Denzin, N. K. (1993). The alcoholic society: Addiction and recovery of the self. New Brunswick,

NJ: Transaction. Denzin, N. K. (1996). The epistemological crisis in the human disciplines: Letting the old do the

work of the new. In R. Jessor, A. Colby, & R. A. Shweder (Eds.), Ethnography and human development: Context and meaning in social inquiry (pp. 127–151). Chicago: University of Chicago Press.

Denzin, N. K., & Lincoln, Y. S. (Eds.). (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage.

Emerson, R. M., Fretz, R. I., & Shaw, L. L. (1995). Writing ethnographic fieldnotes. Chicago: University of Chicago Press.

Engstrom, Y. (1996). Developmental studies of work as a testbench of activity theory: The case of primary care medical practice. In S. Chaiklin & J. Lave (Eds.), Understanding practice: Perspectives on activity and context (pp. 64–103). Cambridge: Cambridge University Press.

Erickson, F. D. (1986). Qualitative methods in research on teaching. In M. C. Wittrock (Ed.), Handbook of research on teaching (3rd ed.). New York: Macmillan.

Fung, H. (1999). Becoming a moral child: The socialization of shame among young Chinese children. Ethos, 27, 180–209.

Gaskins, S. (1994). Integrating interpretive and quantitative methods in socialization research. Merrill-Palmer Quarterly, 40, 313–333.

Gaskins, S. (1996). How Mayan parental theories come into play. In S. Harkness & C. M. Super (Eds.), Parents’ cultural belief systems: Their origins, expressions, and consequences (pp. 345– 363). New York: Guilford Press.

Gaskins, S., Miller, P. J., & Corsaro, W. A. (1992). Theoretical and methodological perspectives in the interpretive study of children. In W. A. Corsaro & P. J. Miller (Eds.), Interpretive approaches to children’s socialization. New directions for child development (pp. 5–23). New York: Jossey-Bass.

Geertz, C. (1973). The interpretation of cultures. New York: Basic Books. Geertz, C. (1988). Works and lives: The anthropologist as author. Stanford, CA: Stanford Univer-

sity Press. Gergen, K. J. (1985). The social constructionist movement in modern psychology. American Psychol-

ogist, 40, 266–275. Giorgio, G. (1999). Living lesbian battering. In N. K. Denzin (Ed.), Cultural studies: A research

annual (pp. 3–18). Stamford, CT: JAI Press. Goldman, L. R. (1998). Child’s play: Myth, mimesis, and make-believe. New York: Oxford. Goncu, A. (1999). Children’s engagement in the world: Sociocultural perspectives. Cambridge:

Cambridge University Press. Goodnow, J. J., & Collins, W. A. (1990). Development according to parents: The nature, sources,

and consequences of parents’ ideas. Hillsdale, NJ: Erlbaum. Gottlieb, A., & Graham, P. (1993). Parallel worlds. Chicago: University of Chicago Press. Grau, M. E., & Walsh, D. J. (1998). Studying children in context: Theories, methods, and ethics.

Thousand Oaks, CA: Sage. Harkness, S., & Super, C. M. (1996). Parents’ cultural belief systems: Their origins, expressions,

and consequences. New York: Guilford Press.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

ETHNOGRAPHIC METHODS 241

Harwood, R. L., Miller, J. G., & Irizarry, N. L. (1995). Culture and attachment: Perceptions of the child in context. New York: Guilford Press.

Heath, S. B. (1983). Ways with words. New York: Cambridge University Press. Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is there a universal need for

positive self-regard? Psychological Review, 106, 766–794. Hymes, D. (1982). What is ethnography? In P. Gilmore & A. A. Glatthorn (Eds.), Children in and

out of school: Ethnography and education (pp. 21–32). Washington DC: Center for Applied Lin- guistics.

Jahoda, G. (1989). Our forgotten ancestors. In J. J. Berman (Ed.), Cross-Cultural Perspectives: Nebraska Symposia on Motivation (pp. 1–40). Lincoln: University of Nebraska Press.

Jahoda, G. (1993). Crossroads between culture and mind: Continuities and change in theories of human development. Cambridge, MA: Harvard University Press.

Jessor, R., Colby, A., & Shweder, R. A. (Eds.). (1996). Ethnography and human development: Context and meaning in social inquiry. Chicago: University of Chicago Press.

Lancy, D. F. (1996). Playing on the mother ground: Cultural routines for children’s development. New York: Guilford Press.

Malinowski, B. (1922). Argonauts of the western pacific. New York: Dutton. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion,

and motivation. Psychological Review, 98, 224–253. Marshall, C., & Rossman, G. B. (1995). Designing qualitative research. Thousand Oaks, CA: Sage. Mead, M. (1963). Socialization and enculturation. Current Anthropology, 4, 184–188. Miller, P. J. (2001, April). New insights from developmental cultural psychology: What the study

of intra-cultural variation can contribute. Paper presented at the biennial meeting of the Society for Research in Child Development, Minneapolis, MN.

Miller, P. J., Fung, H., & Mintz. J. (1996). Self-construction through narrative practices: A Chinese and American comparison of early socialization. Ethos, 24, 1–44.

Miller, P. J., Hengst, J. A., Alexander, K., & Sperry, L. (2000). Versions of personal storytelling/ versions of experience: Genres as tools for creating alternate realties. In K. S. Rosengren, C. N. Johnson, & P. L. Harris (Eds.), Imagining the impossible: Magical, scientific, and religious thinking in children (pp. 212–246). Cambridge: Cambridge University Press.

Miller, P. J., Sandel, T., Liang, C-H., & Fung, H. (2001). Narrating transgressions in Longwood: The discourses, meanings, and paradoxes of an American socializing practice. Ethos, 29(2), 159– 186.

Miller, P. J., Wang, S., Sandle, T., & Cho, G. E. (2002). Self-esteem as folk theory: A comparison of ethnographic interviews. Parenting: Science and Practice, 2, 209–239.

Miller, P. J., Wiley, A. R., Fung, H., & Liang, C-H. (1997). Personal storytelling as a medium of socialization in Chinese and American families. Child Development, 68, 557–568.

Mintz, J. (1999). Self-esteem as ideology and practice: A study of narrative discourse practices among parents and preschool children in a middle-class, European-American community. Unpublished doctoral dissertation, University of Chicago.

Mishler, E. G. (1986). Research interviewing: Context and narrative. Cambridge, MA: Harvard University Press.

Ochs, E. (1988). Culture and language development: Language acquisition and language socializa- tion in a Samoan village. Cambridge: Cambridge University Press.

Ochs, E., & Schieffelin, B. B. (1984). Language acquisition and socialization: Three developmental stories and their implications. In R. Shweder & R. LeVine (Eds.), Culture theory: Essays on mind, self, and emotion (pp. 276–320). Cambridge: Cambridge University Press.

Philips, S. U. (1983). The invisible culture: Communication in classroom and community on the Warm Springs Indian reservation. New York: Longman.

Pike, K. L. (1967). Language in relation to a unified theory of the structure of human behavior. The Hague, Netherlands: Mouton.

Prior, P. A. (1998). Writing/disciplinarity: A sociohistoric account of literate activity in the academy. Mahwah, NJ: Erlbaum.

Richardson, L. (1997). Fields of play: Constructing an academic life. New Brunswick, NJ: Rutgers University Press.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .

242 MILLER ET AL.

Rizzo, T. A., Corsaro, W. A., & Bates, J. E. (1992). Ethnographic methods and interpretive analysis: Expanding the methodological options of psychologists. Developmental Review, 12, 101–123.

Rogoff, B., Mistry, J., Goncu, A., & Mosier, C. (1993). Guided participation in cultural activity by toddlers and caregiver. Monographs of the Society for Research in Child Development, 58 (Serial No. 8).

Rosaldo, R. (1988). Introduction: Grief and a headhunter’s rage. In R. Rosaldo (Ed.), Culture and truth: The remaking of social analysis (pp. 1–21). Boston: Beacon.

Scheper-Hughes, N. (1992). Death without weeping: The violence of everyday life in Brazil. Berkeley: University of California Press.

Schieffelin, B. B. (1990). The give and take of everyday life: Language socialization of Kaluli children. New York: Cambridge University Press.

Seremetakis, C. N. (1991). The last word: Women, death, and divination in inner Mani. Chicago: University of Chicago Press.

Shweder, R. A., Goodnow, J., Hatano, G., LeVine, R. A., Markus, H., et al. (1998). The cultural psychology of development: One mind, many mentalities. In W. Damon & R. M. Lerner (Eds.), Handbook of child psychology, Vol. 1 (5th ed., pp. 865–937). New York: Wiley.

Stevenson, H. W., Lee, S. Y., Chen, C., Stigler, J. W., Hsu, C. C., et al. (1990). Contexts of achievement: A study of American, Chinese, and Japanese children. Monographs of the Society for Research in Child Development, 55 (1–2, Serial No. 221).

Stewart, E. (2000). Thinking through others: Qualitative research and community psychology. In E. Seidman & J. Rappaport (Eds.), Handbook of community psychology (pp. 725–736). New York: Kluwer/Plenum Press.

Strauss, A. L. (1987). Qualitative analysis for social scientists. New York: Cambridge Univer- sity Press.

Taylor, M., & Carlson, S. M. (2000). The influence of religious beliefs on parental attitudes about children’s fantasy behavior. In K. S. Rosengren, C. N. Johnson, & P. L. Harris (Eds.), Imagining the impossible (pp. 247–268). New York: Cambridge University Press.

Van Maanan, J. (1988). Tales of the field: On writing ethnography. Chicago: University of Chi- cago Press.

Wolcott, H. F. (1994). Transforming qualitative data: Description, analysis, and interpretation. Thousand Oaks, CA: Sage.

Wolcott, H. F. (1995). The art of fieldwork. Walnut Creek, CA: Altamira. Wolf, M. (1992). A thrice told tale: Feminism, postmodernism, and ethnographic responsibility.

Stanford, CA: Stanford University Press. Wolf, S. N., & Heath, S. B. (1992). The braid of literature: Children’s worlds of reading. Cambridge,

MA: Harvard University Press.

Co py

ri gh

t Am

er ic

an P sy

ch ol og ic al A ss oc ia ti on . No t fo r fu

rt he

r di

st ri

bu ti

on .