Problem and purpose: The purpose of this study was to investigate reading achievement levels of fourth, fifth, and sixth grade Central Valley students who participated in Reading Recovery in first grade and those students eligible that did not participate in Reading Recovery in first grade (1999, 2000, and 2001). The study also sought to understand what decision-making processes and data analysis administrators use to make decisions regarding interventions. Methodology: This study utilized a quantitative descriptive design. Ex post facto data utilizing a nonrandom sample group and telephone interviewing were the sources of data. Two cohorts over three years (1999, 2000, and 2001) were included in the study. Students at risk for reading failure were divided into two cohorts. This quantitative study examined matched ELA/CST scale scores from participants' fourth, fifth, and sixth grades for both cohorts. Interview responses from administrators responsible for decisions regarding the implementation and/or discontinuation of interventions for at-risk students were collected. Findings and conclusions: There was a significant difference in all three grades of the 1999 cohort, with the non-Reading-Recovery cohort performing at a higher scale score. In the 2000 cohort, the non-Reading-Recovery students performed higher in all three grade levels, as measured by the fourth, fifth, and sixth grade ELA/CST. However, the 2000 cohort's fifth grade year was the only year where there was a significant difference in the performance of the two cohorts. In all three grades, the 2001 Reading Recovery cohort performed at a higher average scale score than the non-Reading-Recovery cohort. The independent t-test revealed a significant difference in the 2001 cohort's performance on the fourth and fifth grade but not the sixth grade ELA/CST. Analysis of interview responses revealed that decisions around intervention programs are accomplished without the use of formal decision-making processes or data analysis. In many cases, the administrators claimed lack of knowledge and time were to blame. Recommendations: Replication of this study with a larger sample over a more extensive geographical area is recommended. It is also recommended that future studies include other interventions students participated in if they did not participate in Reading Recovery. Research studying the use of data analysis in decision-making processes is warranted. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]