This study sought to determine the value of a tiered model of staff development for five districts using Teacher Consultants (TCs) drawn from a parent district with a long writing project history. In these outreach projects, these TCs actualized the National Writing Project (NWP) principle of regarding teachers as expert consultants to their colleagues. Stake’s Contingency and Congruence Evaluation Model was used to establish 1) the relationships among the preconditions necessary for successful implementation of the staff development program, 2) the processes by which the program was to be implemented, and 3) the outcomes which were intended. Data across five replication sites consisted of evaluations from 366 participants; self-reports of changes in skill levels by 191 participants; self-reports of classroom practices implemented by 216 participant and control teachers; and pretest and posttest scores on essays written by 3,927 students of participant and control teachers. It was determined that most required preconditions were observed; that all intended processes of the programs in the replication sites were successfully accomplished; and that the expected outcomes in replication sites were, in fact, achieved. In terms of outcome data, most of the analyses reflect significant differences in writing achievement between treatment and control subjects, favoring students of trained teachers. A survey of classroom practices indicated that trained teachers at all levels implemented more varied composition activities than did non-trained teachers. The results are viewed in light of the literature on effective staff development and, in particular, on the NWP

The purpose of this study was to examine the effects of teacher training in the NWP model on student writing. The sample consisted of 383 students, in junior and senior high school at the time of the study, with ten essays each gathered over three years. Teachers responded to a questionnaire of practices in teaching composition. Results favored the treatment group at the junior high level. The highest mean score was achieved by senior high students of trained teachers. Statistically significant differences were found between trained and nontrained teachers for four instructional practices and for the amount of interaction with other professionals.