Cinnabar and Alabaster

"traits of ethnographic inquiry that make it appropriate to the study of writing instruction": "contextuality, participant observation, multiple perspectives, hypothesis-generating, and meaning-making (see also Kantor, Kirby, & Goetz, 1981)(72). contextuality = "thick description" of the "specific classroom situations" under observation; will assignment sheets or CFPs & writers' descriptions of their situation/position suffice? Xparticipant observationX (i don't want to watch them write, i want to hear how they talk about their writing. so i guess if i'm participating in conversations w/them about their writing, & recording the conversation that kinda counts, but it seems pretty thoroughly… compromised? as if there should be some level of remove/sterility there that such a one-on-one encounter can't possibly afford. "multiple perspectives" = "ethnographers also seek to confirm, question, or add to their judgments by eliciting the judgments of others (73). this corraboration may be achieved by having more than one researcher in the setting, by teaming investigator and participant (Smith & Geoffrey, 1968), or by using key informant interviews (Jackson, 1968)." do one-on-one interview conversations count as "teaming"? somehow i doubt it…"hypothesis generating" = "an open, eclectic stance on the part of researchers, who attempt to generate rather than test theories. they view emerging patterns with a skeptical eye until they have opportunities to reexamine and reevaluate these patterns. at the same time they may shape, alter, or refine their investigations as they proceed"; "ethnographers generally enter settings with assumptions and predispositions, but maintain their options until the weight of evidence determines particular directions." meaning making = "ultimately ethnographers are concerned with the ways in which individuals construct meanings for themselves, especially as members within a social or cultural group" (74-5).

examples of one researcher's open-ended interview questions (composing process study):
1. What did the instructor say about your previous paper?
2. What did you consider to be the strengths and weaknesses of this piece?
3. How does this paper compare in terms of quality with your previous work?
4. What major changes have you made in revising it? Why have you made these changes?
5. Are there other changes you want to make? What are they?
6. Could you tell me how you went about writing this paper? (133)

"Writing in a Nonacademic Setting" 225-258.
Lee Odell & Dixie Goswami's study's structure includes initial interviews & follow-up interviews a few months later; before the each "data-gathering interview" "we asked each worker to keep copies of all the writing he or she did during a two-week period" that was collected & examined by the researchers, & from which unique "interview sheets based on writing we had collected from [each] participant[]" were prepared; for each piece of writing "we would identify from four to six choices" writers had made, & for each write down what the writer had chosen followed by several alternative approaches to the particular writing situation, so that during interviews they could discuss with participants why each had made the choices they had as opposed to other possibilities, & whether they would be willing to switch w/in the context of their writing, & why (workplace scenarios & genre-appropriate expectations, etc.) (237-8).

reducing anxiety involves many repeated assurances (a) that different choices were all "correct," (b) that the interviewee was the expert on making choices appropriate to his/her writing situation, (c) that researchers' interest was solely in writers' reasoning regarding their choices, + lots of careful wording: "early in the interviewing process we became concerned that the form of our questions, even our tone of voice, might influence a worker's response. it is, for example, surprisingly easy to say 'why didn't you do y rather than x?' in a tone of voice which implies that y is the only acceptable alternative. there is also a danger of paraphrasing a worker's comments in such a way that they fit neatly with the interviewer's expectations: 'oh, so what you're saying is that you were very concerned with your audience at this point.' inevitably, of course, the very presence of an observer has some influence on the phenomena being observed. but we tried to establish procedures that did not imply our values and that let us make our interviews as nondirective as possible. for example, after a worker responded to our initial question, 'would you be willing…,' he or she often paused. we would remain silent for a few seconds and allow the worker to elaborate. if no elaboration was forthcoming, we would begin a sentence with the words 'so what you're saying here is…' and then pause, allowing the worker time to complete the sentence. if the worker did not do so, we would either say 'i'm not sure i understand,' ask the worker 'could you elaborate?' or paraphrase the worker's comment as closely as possible and give him or her a chance to modify our statement" (238-9).

data-coding & analysis: interview transcripts & categorizing responses ("guided in part by our reading and rereading of interview transcripts and in part by assumptions from rhetoric and current theories about the purposes or functions of discourse"—chosen categories were "audience-based reasons, writer-based reasons, and subject-based reasons," based on "distinctions made by Gibson (1969), Kinneavy (1971), and Halliday (1973)"—but not "the functions of complete pieces of discourse" a la Kinneavy; rather they're "types of reasons given to justify several specific choices in any given text. discussion of a single text—indeed even discussion of a single choice—might elicit more than one type of reason" identified audience-based reasons included status, personal knowledge or relationship w/audience, personal characteristics, & "anticipated or desired action on the part of the audience"; writer-based reasons inc. "writer's role or position in the organization," "ethos or attitude the writer wishes to project or avoid," "writers feelings about subject or task at hand"; subject-based reasons inc. importance of topic, "desire to provide an accurate, complete, nonredundant account," or desire to make a point (241-2).