This study tracked the reading progress of seven grade level cohorts of former Reading Recovery (RR) students through the end of the 2004-2005 school-year. The purpose of this longitudinal study was to investigate the reading achievement outcomes of 214 students who received RR tutoring during first grade. The subjects were elementary students (grades 2-8) enrolled in one school district in southern Illinois.

This study used a cross-sectional, retrospective longitudinal design to track the reading performance of students through the elementary grades. A quantitative methodology was utilized in the collection and analysis of data for the study. The reading performance of Reading Recovery subjects was measured using four dependent variables: 1^st grade pre/post reading text levels (RTLs), Iowa Test of Basic Skills (ITBS) reading scores, Illinois Standards Achievement Test (ISAT) reading scores, and the Standardized Test for the Assessment of Reading (STAR) test results.

Descriptive statistics including means, standard deviations, and percentiles were calculated to analyze the data. Mean scores obtained were then described, along with cut scores established to indicate expected levels of proficiency. In addition, descriptive data was recorded and analyzed to examine the impact of gender, ethnicity and socio-economic status on differences in reading outcomes, for the elementary students involved in this study.

The data analysis revealed that in the short-term the RR subjects' reading text level gains exceeded expected levels, based on pre/post reading text level gains, and RTL cut scores. However, the long-term results indicated that the RR subjects' performance on the ISAT, ITBS, and STAR did not meet or exceed cut scores. These findings suggest that the RR subjects made reading progress in the short-term at a rate faster than would be expected of average achieving students.

As it relates to the short-term and long-term reading performance of the subjects in this study, the findings reveal persistent disparities across instruments, and grade levels, and across gender, ethnicity and socio-economic status lines. Although such disparities were not the major focus of this study, this was by far the most significant finding and warrants further investigation of the nature and causes of disparities among student groups.