The ability to read fluently and comprehend the reading is a fundamental skill increasingly necessary to be academically and socially successful (Topping & Paul, 1999, p. 214). Melton et al. (2004, p. 18) claimed the ability to read will have the greatest influence on the success of Mississippi students and thus led the authors to investigate how well a computer based accelerated reader program (ARP) influenced the reading performance of the fifth grade class at East Side Elementary school in Clinton County Mississippi overall and by race.
The authors conducted an ex-post facto study to determine ARP effectiveness by comparing the Terra Nova standardized reading achievement test results from Eastside Elementary, the experimental group, to the non-ARP control group from Hinds County Gary Road Elementary school. The authors used the analysis of covariance (ACNOVA) statistical approach to assess the significant difference of reading comprehension between the experimental and control groups.
Melton et al. (2004) findings, which are inconsistent with previous research identified in their own literature review, suggest ARP students’ reading performance is below that of their control group counterparts. The study does not provided a hypothesis explaining why the ARP has detrimental effects on the students’ reading performance and provides limited statistical data to validate the results. The authors leave it to the readers to make sense of the results. Since previous studies suggest findings contrary to those presented by the authors, it may be the results are not valid. The assessment presented here will focus on the study weaknesses and how those weaknesses may have contributed to the authors’ contrary findings. This paper will first discuss the weaknesses in approach and statistical analysis, followed by a discussion of the implications of the findings and then conclude with an overall assessment.
Melton et al. (2004, p. 18) determined the effectiveness of the accelerated reading program based their results on ingle ARP implementation at Eastside. To assess the ARP’s influence on reading improvement, ACNOVA was used to normalize the experimental and control group using Terra Nova standardized reading achievement pre and post-test scores, the ARP was the treatment, the pretest was the covariant (CV), and the posttest was the dependent variable (DV). This assessment will first address a weakness associated what is being measured.
Assessment of the ARP implementation. Melton et al. (2004) may not be measuring the effectiveness of ARPs to improve reading comprehension as the authors suggest. It may be that the authors are measuring the quality of the implementation of the ARP rather than how well it improves reading performance. Melton et al. (p. 20) imply the quality of ARP implementations nationwide is suspect because only 279,000 educators have ARP training, but the program is being used in 55,000 schools. The authors,...