The purpose of the study was to describe the influence of conferences on revision knowledge and revision activity for an initially knowledgeable first-grade reviser and a relatively naive one. Using both quantitative and descriptive analyses, data from three baseline points and seven conference points were summarized. At conference information collection points, the children wrote, groups conferenced, children were interviewed about potential revisions, and students revised. At baseline points, there were no conferences. For analyses, we (a) read all of the writing and transcripts of interviews and conferences; (b) determined quality of first and last draft of each composition; (c) coded conference talk; (d) traced comments between conferences, interviews, and revisions carried out; (e) calculated counts of revisions the children talked about in interviews and revisions they carried out; and (f) reviewed the teacher's observations about the two children's behavior in the classroom. Conference influence was variable, both within and across children. Further, dramatic differences were noted between the two children in conference influence, with only the initially naive writer profiting significantly from conferences and evidencing clear developmental progress in revision.The purpose of this study was to describe the influence of conferences on revision knowledge and revision activity for an initially knowledgeable first-grade reviser and for a relatively naive one. The present study was a follow-up to prior research (Fitzgerald & Stamm, 1990) in which we examined the influence of conferences on 16 first-graders' revision. Earlier, we concluded that conferences did influence revision for many children, but that the influence was mediated by entrylevel revision knowledge and activity and writing level. For the present study, we tried to further describe how conference influence differed for different kinds of writers by looking in-depth at the revision development of 2 of the 16 children
22Journal of Reading Behavior who participated in the earlier study-one whose initial revision knowledge and activity were relatively low and one whose knowledge and activity were relatively high.Revision was defined as making changes at any point of writing (before, during, and/or after pen meets paper). It involves (1) detecting mismatches between intended and instantiated text, (b) deciding how changes could or should be made, and (c) actually making changes (cf. Beach, 1984;Fitzgerald, 1988;Fitzgerald & Markham, 1987;Flower & Hayes, 1981). The first two aspects of revision (detecting mismatches and deciding how to make changes) are knowledge components of revision, distinguished from the third aspect (making changes) which is actual revision activity. This definition of revision is highly consistent with the problemsolving view of writing as delineated by Flower and Hayes (1981), in which authors set goals for their pieces, check their work against those goals to detect problems, make decisions about where and how change...