This study describes the long-term reading achievement of a group of children identified by teachers in 1994 as low-achieving students in reading. Four research questions guided the study, and the 165 participants came from 12 elementary schools in Montgomery County, Virginia. They were selected, because they had participated in Reading Recovery in first grade and had taken the reading portion of the Stanford 9 Achievement Test in the spring of third grade.

The literature review covers the history of reading instruction, outlines the Chapter I model, compares and contrasts the New Zealand and American Reading Recovery models, and profiles the impact of Reading Recovery on the research community.

Means and standard deviations were analyzed to compare the relative performance of four major populations: Reading Recovery, Reading Recovery/Chapter I, Chapter I, and Waitlisted.

The results of the analyses suggest that Reading Recovery students successfully discontinued in any number of lessons had means in the top half of the distribution of means for all populations in the study. With the addition of one to two years of Chapter I instruction after Reading Recovery, some students achieved the highest mean.

The scores of Chapter I students with one year of instruction nearly matched the Reading Recovery mean. Two to three years of Chapter I produced lower means.

The waitlisted students scored slightly lower than the other three populations.

The results of the study confirm the efficacy of Reading Recovery as an early intervention, but indicate that using Chapter I as a transitional step between Reading Recovery and successful independent reading sustains long-term reading achievement.