“The many faces of research”

I just realized I never re-posted my October 2010 article summarizing James Spradley’s incredibly cool way of defining different types of research — by the role of the participant vis a vis the researcher.

Here’s the text:

Anyone who has ever commissioned, designed, conducted research will find these common but thorny questions all too familiar:

  • “What is this research going to give us that we can’t get from analytics and iterative design?”
  • “Don’t you need to ask all your interviewees the same set of questions so you can compare their answers?”
  • “Can you quantify these findings?”
  • And with qualitative research, the dreaded: “That’s an awfully small sample. Are these findings statistically significant?”

These questions can be difficult to answer clearly, succinctly and definitively. Wouldn’t it be helpful to have some kind of framework or model to help people understand how the various kinds of research (especially qualitative and quantitative fit together) to provide an organization what it needs to effectively engage and serve their customers?

James Spradley in The Ethnographic Interview provides such a framework. His approach is the identification of four different roles a research participant can play, each with a different relationship between researcher and participant and each producing a different kind of finding:

  • Informant – In ethnography, a participant is related to as an informant. Informants are “engaged by the ethnographer to speak in their own language or dialect”, providing “a model for the ethnographer to imitate” so that “the ethnographer can learn to use the native language in the way informants do.” The informant is related to as a teacher. What is learned is how the participant conceptualizes and verbalizes his experience. Informants give the researcher not answers to fixed predetermined questions, but the questions themselves. Informants help define what the researcher needs to learn in subsequent research. (Examples of research techniques with informants: unstructured and semi-structured interviews, diary studies, open card sorting, collaborative design exercises.)
  • SubjectSubjects are participants in social science research, upon whom hypotheses are tested. “Investigators are not primarily interested in discovering the cultural knowledge of the subjects; they seek to confirm or disconfirm a specific hypothesis by studying the subject’s responses. Work with subjects begins with preconceived ideas; work with informants begins with a naive ignorance. Subjects do not define what it is important for the investigator to find out; informants do.” (Examples of research techniques with subjects: usability testing, split testing, concept testing.)
  • Respondent – A respondent is any person who responds to a survey questionnaire or to queries presented by an investigator. “Survey research with respondents almost always employs the language of the social scientist. The questions arise out of the social scientist’s culture. Ethnographic research, on the other hand, depends more fully on the language of the informant. The questions arise out of the informant’s culture.” (Examples of research techniques with respondents: surveys, questionaires, structured interviews, closed card sorting.)
  • Actor – “An actor is someone who becomes the object of observation in a natural setting.” As with subjects and respondents, when participants are related to as actors, the terms of the description of the actor’s behaviors are those of the researcher, not of the participant. It should be noted, however, that in ethnographic research (and also in contextual inquiry, participants are interviewed as they are observed, which means the participant is still understood  primarily as an informant. The actor-informant teaches the researcher through showing and explaining in his own terms the significance of his actions, which allows the researcher to give (to use Clifford Geertz’s term) “thickness” to his descriptions of what he observes. (Examples of research techniques with actors: site analytics, business intelligence analysis, silent observation.)

Over the course of a research program, research participants may at various times be regarded as subjects, actors or respondents — but if the goal is to know what really motivates the participants, to understand how to engage them at an emotional level, and to cultivate an enduring relationship with them, it makes a lot of sense to begin by relating to research participants as informants, beginning with unstructured or semi-structured interviews.

By starting with an informant relationship with research participants researchers can develop a better idea of what matters to the participants, how they conceptualize and speak about these things, and most importantly how this motivates observable behavior. These insights (that is, findings that illuminate the inner life of participants) can focus subsequent research on the most relevant and impactful questions. It also improves the execution of the research by helping researchers use language that’s natural and understandable to participants, earning greater trust and cooperation, and minimizing misunderstandings. And in analysis researchers and planners will mine more valid insights from the data, since they understand the motives, thought process and language behind the responses and behaviors of the respondents, actors and subjects. And the insights will be accurate because they rely far more on fact than (often unconscious) assumptions.

The other types of research can then report in more quantifiable terms, using much larger samples, how many subjects or actors perform certain behaviors or how many respondents give one answer or another to certain questions on a survey or questionnaire — and these actions and responses will now carry much more meaning because now the researchers have subjective insights to complement the objective data.

Two more points worth making: 1) I haven’t mentioned segmentation in this article, but anywhere where I mention learning about research participants, I am talking about learning about segments of participants (defined by goals, needs, attitudes and behaviors), and understanding the similarities and differences among them. 2) Generally, it is in the role of informant that research participants provide findings that drive design and creative. Informants inspire empathy and creative approaches. Subjects, respondents and actors tend to yield information useful in making strategy decisions. Using the full range of qualitative and quantitative research methods together intelligently can enable strategists and designers to work together more effectively to harness the full power of experience design.

By understanding research better — recognizing the difference between research that produces subjective insights and research that produces objective data, by not mistaking them for rival methods for producing the same kinds of findings, and by understanding how they can be used together to gain a holistic picture of one’s customers that is far more than the sum of the facts — an organization becomes more capable of understanding its customers without sacrificing their individuality to empty statistics.

Leave a Reply