Interviewing Joshua: on the importance of leaving room for serendipity.

Joshua, a second grader reading at the preprimer level, resists answering the author's questions about his conceptions of reading until she agrees to spend equal time drawing and looking through Waldo books with him. Surprisingly, it is while "doing Waldos," rather than during classroom observations or in answering her carefully planned interview questions, that Joshua shows the author his developing sense of narrative, his earliest attempts at phonetic decoding, and the importance of minimal-text books like Waldo books as a nonthreatening gateway into literate experience for him and other struggling readers. The author comes to see such shared agenda setting as not just the most ethical way to interview people but also the most effective, because it allows for the serendipity of discovering allows to questions the author had not even thought to ask.

In this article, I want to share something I am learning about interviewing in qualitative research: how "leaving room" in interviews for talk about the interests of each participant can increase the benefits of the relationship for both. I also want to share some things I learned during my first experience with this sort of interviewing about the process of learning to read, and finally, to introduce you to Joshua(1) who was my teacher in both of these learnings.

I interviewed Joshua and observed him in his second-grade classroom throughout the 1993-1994 school year as part of my larger dissertation study (Knapp, 1994; see also Knapp & Grattan, 1994). Joshua's teacher, Kathrin Grant, taught at Spruce Elementary, a school serving a primarily Caucasian, socially and economically diverse population in a small rural/suburban midwestern community that is a county seat and is also 10 miles outside a state capital. Ms. Grant was particularly experienced and thoughtful in using a whole language, literature-based approach to reading instruction. At the time of the study, she had more than 9 years' teaching experience and had recently completed her master's in literacy at Michigan State University. Her teaching related to reading was varied, rich, and flexible. She used a variety of materials, including poems, chapter books, picture books, student texts, scholastic newspapers, textbooks, and the required basal. Sometimes students chose the texts they would read; at other times she assigned texts to the whole class. She encouraged children to access these texts in many different ways, through individual silent reading, partner reading, oral reading in heterogeneous groups, watching films, listening to tapes or to the teacher reading aloud, and choral readings of poetry. Students read aloud texts they brought from home; gave "commercials" for favorite texts; compared different versions of texts; found different types of words in texts; responded to texts in literature logs, through art, and through class discussions; used texts as references or as jumping off points for composing texts of their own; and simply enjoyed texts. Reading was also an inherent part of instruction in science and social studies and was seen as a resource and model for work in the writer's workshop.

Ms. Grant wanted all her students to become self-directed thinkers, readers, questioners, and problem solvers. Her reading instruction focused on helping students to value and enjoy literature, to use it to "open their horizons" beyond their own thought and experiences. She also wanted them to value reading for learning and to be able to access written information. She helped them become aware of the conventions of various literary genres and encouraged them to use these to expand their own writing patterns. She wanted them to be able to recognize narrative structure and retell basic tales from their own and others' cultures (e.g., Grimm's tales from Europe, the Anansi stories from Africa). Through all these activities, she worked to help them develop sufficient fluency in decoding to be able to access quality children's literature on their own.

I was particularly interested in the diverse conceptions about reading and learning to read that students experiencing this sort of instruction, as opposed to traditional basal instruction, might construct. In looking at these issues, I was in part drawing from phenomenological (e.g., Thomas 1928/1966; Weiner, 1984) and constructivist (e.g., Piaget, 1973; von Glasersfeld, 1987) learning theories that emphasize the importance of the learner's conception of the learning situation and subject, and in part responding to calls such as the following from Erickson and Schultz (1992):

Virtually no research has been done that places student experience at the

center of attention.... If the student is visible at all in a research study

he is usually viewed from the perspective of adult educators' interests and

ways of seeing, that is, as failing, succeeding, motivated, mastering,

unmotivated, responding, or having a misconception. Rarely is the perspective

of the student herself explored. (p. 467)

Because I agree with McGill-Franzen, Lanford, and Killian (1993) that it is particular "children's day-by-day experiences within particular classroom communities [that] shape their cumulative understandings of the uses, purposes, and possibilities of written language" (p. 3; see also Florio-Ruane, 1989), I decided to do case studies of seven focus children in Ms. Grant's classroom, chosen through a process of purposeful sampling (Patton, 1990). These case studies were based on triangulated data from a number of sources. Main data sources were classroom and playground observations, individual and group student interviews, interviews with Ms. Grant, and copies of classroom materials and student products. Secondary sources included interviews with other adult informants such as the school principal and the Chapter 1 reading teacher, copies of testing reports and other student-related documents, and informal conversations with students and adults in the classroom throughout the school year. I wanted to use all these data to build "thick descriptions" (Geertz, 1973) of each student's world of reading and to try to see each child's experience from his or her own point of view as well as my own.

Given this focus, I knew from the inception of my project that the student interviews would be a particularly key data source for me. Perhaps because I came from a background in educational psychology, which, emphasizes comparability, control, and codability in interview studies, my first interviews with students were based on carefully arranged protocols of questions that I asked each of the students to answer. These protocols did include initial, more general questions about the students' activities in and out of school, designed to "break the ice" and "establish rapport," but soon shifted to focus on areas that I believed important to young children's ideas about reading. Like most novice interviewers, I tended to plan …