Analyzing Learning Equity Research Trends (ALERT)
|
Font Size : 
Analyzing Learning Equity Research Trends (ALERT) graphics are designed to provide information on state, district, and school level achievement gaps between African American/Caucasian, Male/Female, Mobile/Non-Mobile/ and Participants/Non-Participants in the Federal Free and Reduced Lunch Program (FRLP).
Goal of ALERT
Race Group ComparisonsThe achievement gap between African American and Caucasian students receives the most attention in Arkansas, as well as nationally. Jencks and Phillips (2002) identified national achievement gaps of .67 standard deviations in reading and .90 standard deviations in mathematics between African American and Caucasian students. Mulvenon, Ganley, McKenzie and Stegman (2003) examined the impact of desegregation in the Little Rock School District and determined that the achievement gaps in reading and mathematics were in some cases greater than expected given the national trends. Given that 40 percent of the Arkansas population is African American, it is important that achievement gaps between African American and Caucasian students are identified and targeted for reduction.Mobile/non-Mobile and Participant/non-Participant FRLP ComparisonsMobile/non-Mobile, and Participant/non-Participant in Federal Free and Reduced Lunch Programs (FRLP) were selected because of the prevalence of requests regarding the effectiveness of the Arkansas and national public school systems in educating members of these various groups. Given the readily available evidence of achievement gaps between Mobile/non-Mobile and Participants/non-Participants in FRLP, it is imperative we do all we can to better understand the size of the achievement gaps so we can begin developing methods to shrink the gaps.Male/Female ComparisonsIt is important to examine gender differences in the context of student achievement in mathematics and literacy and to determine if those differences are significant or meaningful. Use of Effect Sizes and Interpretation of ResultsDirect comparison of mean differences between the groups does not provide a 'meaningful' context of what is an important difference. To help with this interpretation, we have computed effect sizes for each comparison, which is a measure used in research to provide a 'meaningful' context. However, given many people accessing this page may not be familiar with effect sizes, we have generated interactive interpretations that are provided with a graph. For example, you would obtain the following text:The effect size of .367 indicates a moderate difference between groups. It is representative of a small educational achievement gap. It is our hope this type of interactive interpretation of the boxplots will grow or increase understanding of how to make scientifically based determinations of the 'differences' in performance.SummaryThe future holds the possibility for continuing to develop achievement gap comparisons on the student population both in Arkansas and nationally. We are open to suggestions and recommendations. A driving philosophy we adhere to is the belief that we must KNOW the differences before we can develop the remedies. The ALERT and GAP series are both attempts to help us better understand what the opportunities for improvement are and how we can begin to develop methods or strategies to improve education in Arkansas. The following is an excerpt from a U.S. Department of Education document:A number of recent groups have described current education practices as not resting on a solid research base (Coalition for Evidence-Based Policy, 2002; NRC, 1999; NRC, 2000; NRC, 2002). Instead, policy decisions are often guided by personal experience, folk wisdom, and ideology. We hope through NORMES, the GAPS series, and ALERT Series, we are beginning to address the educational problems in Arkansas using empirical evidence and scientifically based research methods. The next step is to generate effective educational policy using this information.Technical Notes:To ensure that no violations of FERPA occur, we have eliminated the development of boxplots for any district that has fewer than 10 students within any of the comparison groups. Some Comparisons, due to sample size issues, are not possible at this time including:
|
