120 ACCESSING AND GENERATING DATA
interviewers in an interpretive mode long ago gave up on the idea that one could control for
“interviewer effects.” Even if one could train survey researchers to adhere strictly to the question-
naire protocol—something that is humanly very difficult—one cannot control for the “effects” of
nonverbal communication, including tone of voice and other paralinguistic elements, physical
characteristics such as hairstyle and posture, facial and hand gestures, and so on. These, too,
communicate meaning and are interpreted, whether in keeping with intended meanings or not.
The alternative to “control” is to be as reflectively aware as possible of the ways in which one
might be affecting the exchange.
Moreover, as Schaffer notes, “nonjudgmental” does not mean “blank” or “passive.” Some
methodologists argue that it is inhumane and unethical for researchers to withhold their opinions
when they feel strongly about something their interlocutors say (on this, see Holstein and Gubrium
1995). Without the constrained response possibilities of a survey, with the focus on understand-
ing meaning rather than on opinion or attitude, and with the format of a conversation, people are
far less likely to try to second-guess the interviewer’s hidden motives and to reply with a sur-
mised “right” answer. Their views of their own lives, in the end, are what interpretive researchers
are trying to get close to. And researchers are always contextualizing what they are told, with
other interviews and/or documents and/or their own observations.OBSERVINGAnother central mode of accessing and generating potential data is observing, with whatever
degree of participating (Gans 1976). This mode can be used as a stand-alone source of data, or it
can be combined with interviewing, in which each method is used to corroborate (or refute)
provisional interpretations derived from the other. In researching demographic changes in a neigh-
borhood, it can be as edifying to pay attention to the amount and kinds of laundry hanging on the
line, for example, as it is to interview the neighbors.
Ethnographic and participant-observational research rest on observation. (Both may include
varying degrees of participation, which itself entails observing, and they may also draw on docu-
mentary evidence; as both methods engage people in conversation, they typically do not desig-
nate “interviewing” as a separate method.) The two approaches were divided in the United States
with the demarcation between sociological ethnography and anthropological ethnography, sepa-
rating the territory between domestic studies and overseas ones. Within the United States, the
“unmarked” names of fields in sociology and other areas that imply “U.S.” or “domestic” as a
modifier—such as public administration, public policy, urban studies, labor studies, community
studies, social problems, organizational studies—tend to use “participant-observer” methods; those
fields whose research typically requires facility in a language other than English and an extended
sojourn in some “foreign” locale—such as international studies, comparative government, devel-
opment studies—tend to use “ethnographic” methods. As Katz (2004, 9) describes the difference,
anthropological ethnography “has been much more likely to require long term personal planning,
travel documents and funding”; whereas sociological study entails a “relatively full and tempo-
rally open ended participant observation in the social lives of subjects.”
In chapter 8, Ellen Pader develops the idea that it is possible to learn to observe with “an
ethnographic sensibility.” Ethnography is “the attempt to understand another life world using the
self—as much of it as possible—as the instrument of knowing” (Ortner 1996, 281). In such a
study, “the whole self physically and in every other way enters the space of the world the re-
searcher seeks to understand.” Contemporary ethnographers, as Ortner notes, commit “to produc-
ing understanding via richness, texture, and detail rather than parsimony, refinement, and (in the