Spradley describes ethnography as different from deductive types of social research in that the five steps of ethnographic research: selecting a problem, collecting data, analyzing data, formulating hypotheses, and writing all happen simultaneously (p. 93-94).

In The Ethnographic Interview, Spradley describes four types of ethnographic analysis that basically build on each other. The first type of analysis is domain analysis, which is “a search for the larger units of cultural knowledge” (p. 94). The other kinds of analysis are taxonomic analysis, componential analysis, and theme analysis.

All of Spradley’s theories about ethnographic analysis hinge on his belief that researchers should be searching for the meaning that participants make of their lives. These meanings are expressed through symbols, which can be words, but can also be nonverbal cues. However, because this book is about analyzing interviews, Spradley focuses on analyzing the spoken words of the participants. He explains that words are symbols that represent some kind of meaning for an individual, and each symbol has three parts: the symbol itself, what the symbol refers to, and the relationship between the symbol and the referent. Thus, the word computer is a symbol for me. It refers to many things, including my own personal computer, which is an Apple laptop. Thus, my laptop is a kind of computer in my mind, and this shows the relationship I am making between the symbol (computer) and the referent (the particular computer I am typing on now).

Spradley defines a domain as the “symbolic category that includes other categories” (p. 100). A domain, then, is a collection of categories that share a certain kind of relationship. Computers is a domain that includes not only my laptop, but all the Dell, Toshibas, iMacs, and IBMs of the world. These all share the same relationship because they are all kinds of computers. Spradley explains that there are three elements of a domain. First, the cover term, which in my example is the word “computer”. Second, there are included terms, which are all the types of computers I just listed. Finally, there is the single, unifying semantic relationship, which is the idea that “X, Y, and Z are all kinds of A”.

When doing domain analysis, Spradley suggests first doing a practice run, which he calls preliminary searches. To do this, you select a portion of your data and search for names that participants give to things. You then identify whether any of these listed nouns might possibly be cover terms for domains. Finally, you can then search through your data for possible included terms that might fit under this domain you have identified.

Remember, this was just the warm-up. To actually do domain analysis, you look for relationships in the data, not names. Spradley is famous for his very useful list of possible relationships that may exist in your data:

1. Strict inclusion (X is a kind of Y)
2. Spatial (X is a place in Y, X is a part of Y)
3. Cause-effect (X is a result of Y, X is a cause of Y)
4. Rationale (X is a reason for doing Y)
5. Location for action (X is a place for doing Y)
6. Function (X is used for Y)
7. Means-end (X is a way to do Y)
8. Sequence (X is a step or stage in Y)
9. Attribution (X is an attribute, or characteristic, of Y)

To do domain analysis, you first pick one semantic relationship. Spradley suggests strict inclusion or means-end as good ones for starters. Second, you select a portion of your data and begin reading it, and while doing so you fill out a domain analysis worksheet where you list all the terms that fit the semantic relationship you chose. Third (if you follow along in Spradley’s book, you’ll notice I’m crunching his steps together for brevity) you formulate questions for each domain. So to revert to my example, if you identified from your interview with me that I feel that Macs are kinds of computers, you could test this hypothesis by making a question out of this semantic statement, “Are there different kinds of computers?” You could ask me, or another participant, and based on their answer, you would know if the cover term, included terms, and semantic relationship that you identified were correct. You could then probe with more questions like, “Why are Macs a kind of computer?” or “In what way are Macs a kind of computer?” In this way, your analysis feeds into your next round of data collection.

The final step in domain analysis is to make a list of all the hypothetical domains you have identified, the relationships in these domains, and the structural questions that follow your analysis.

• 1970 You Owe Yourself a Drunk: Adaptive Strategies of Urban Nomads. Boston: Little Brown. (Reissued Prospect Heights, IL: Waveland Press, 2000.)
• 1972 Culture and Cognition: Rules, Maps and Plans. San Francisco: Chandler.
• 1975 The Cocktail Waitress: Woman's Work in a Man's World. New York: Wiley.
• 1979 The Ethnographic Interview. United States, Wadsworth Group/Thomas Learning.
• 1980 Participant Observation. New York: Holt, Rinehart and Winston.

Spradley and David McCurdy

• 1989 Anthropology: The Cultural Perspective. Prospect Heights, IL: Waveland Press. (First published in 1980.)

James Spradley and Thomas Spradley

• 1979 Deaf Like Me. New York City, NY: Random House (Reissued Gallaudet University Press: Washington, DC, 1987)