 | Go directly to "Performance Graphs for Grades 3 – 8, 2001." |
| Go directly to "Performance Comparisons for Alabama School Systems, 2001." |
| Go directly to "High-Performing Schools in Grades 3 – 8, 2001." |
HOW
TO USE THESE DATA ON SCHOOL PERFORMANCE
Accountability in the public schools has become an important national issue. One of the key initiatives now before the Congress is to promote school accountability through reforms in the way federal education dollars are distributed, and it appears that there is broad agreement for some kind of change in this direction. Many states, including Alabama, already have accountability laws on the books. (If you are not familiar with Alabama’s accountability law, you may want to read a brief description; click on this link to the section titled "Academic Accountability in Alabama’s Public Schools.")
These accountability laws and proposals focus on student performance, and particularly on test scores. Typically they set minimum test-score standards for local schools and require state assistance or intervention for schools that fall below them. Such minimum standards serve a useful purpose by encouraging school staffs to focus on academic improvement issues; state assistance and intervention also can be helpful in isolated instances. However, Alabamians will be best served if as many schools as possible are able to improve on their own, without having to undergo a process directed by state officials. The best way for them to do this is to learn from each other.
If we expect the schools to improve on their own initiative, then we ought to provide them with performance data that allow similar schools to be compared with one another. The information in this report is organized to do just that.
Many studies of school performance have shown that test-score averages are affected by student characteristics. PARCA’s research on Alabama’s schools has shown that schools with higher percentages of students from poverty backgrounds tend to have lower scores on the Stanford Achievement Test (SAT). The poverty level for schools is commonly measured by the percentage of students who qualify for the federal program of free and reduced-price lunches (FRL). The relationship between SAT scores and FRL percentages shows up year after year; it reflects in part the educational disadvantages that such students face, and in part the failure of schools to overcome them. But the important fact about this pattern is that there are schools at every poverty level that score much better than average, as well as schools that unfortunately perform well below average.
High-performing schools obviously are doing something right – and the practices that make them successful ought to be copied, adapted, or at least considered by others. The key to school improvement in Alabama is for the successful practices of high-performing schools to be transferred to those schools that need to improve.
The first step in bringing about this transfer is to organize test-score information so that the high-performing schools at every level can be identified. This is what we have tried to do. Our purpose is not to label any school, but to assist those that want to improve and thereby to make a contribution toward raising the level of student performance throughout the state.
The school report cards published annually by the State Department of Education contain information on the FRL percentage in each school, as well as the overall ("total battery") SAT score for each grade within the school. The chart below shows the relationship of the third-grade SAT score to the FRL percentage[i] for every school in the state. Each of the blue dots in the chart represents a school. Schools with lower FRL percentages are on the left side of the chart; they generally have higher SAT scores, while schools on the right side of the chart have higher FRL percentages and generally lower SAT scores. The black line running through the middle of the dots shows the SAT score we would predict for the third grade class in any school, given its FRL percentage.
The predicted score for schools with low FRL percentages is well above the national average of 50, and it drops below 50 in schools with two-thirds or more of their students from poverty backgrounds. Some schools on the left side of this chart were not high-performers even though they had a third-grade SAT score that was above the national average of 50. Most of their peers did much better. On the other hand, some schools on the right side of this chart did far better than their peers even though they were only slightly above the national average. When performance is looked at in this way, we can see that there are many schools in Alabama, all across the chart, where performance is well above average. These schools are likely to have practices that are worth considering by others not yet as successful.
The chart contains two other lines that parallel the predicted SAT scores. The green line is ten points above the black line, and the red line is ten points below. Dots above the green line represent schools with third-grade SAT scores ten or more points above the predicted level. In seeking best practices that might be copied, the first place to look is the schools above the green line. From a school-improvement perspective, these were the highest-performing schools in the state. It is important to see that these schools can be found at all poverty levels, which means that any school with a score below expectation is likely to find similar schools that are high-performing.
The same kind of chart can be prepared for each school that has a fourth-grade class, a fifth-grade class, a sixth-grade class, a seventh-grade class, or an eighth-grade class. (To see all of the charts, click on this link to the appendix titled "Performance Graphs for Grades 3 – 8, 2001.") From these charts, we have developed predicted SAT scores for every Alabama public school with students in grades 3 – 8. SAT scores also are available for grades 9 – 11, but the FRL data are unreliable for many high schools, and so we have not calculated expected SAT scores at these levels.
The results of our analysis of SAT scores for Alabama’s public schools with students in grades 3 – 8 are illustrated in the chart below, which shows the Autauga County School System. Schools are listed individually in the table, and the first column of numbers shows the FRL percentage. The next six columns contain the predicted SAT score for grades 3 – 8 in each school. These predictions are based on the relationship between SAT scores and the FRL percentage, as discussed and shown in the chart on the previous page. The six color-coded columns on the right side of the chart compare the actual SAT scores for the school to the predicted scores in the middle columns. Schools that scored ten or more points above predicted are coded with a bold green color, and those that scored from 1 to 9.9 points above predicted are coded with a light green color. Schools that are within one point of the predicted score are coded with a gray color. Schools that scored ten or more points below predicted are coded with a bold reddish color, and those that scored from 1 to 9.9 points below predicted are coded with a light reddish color.
