Data sgp is an educational metric derived from longitudinal test score data that helps teachers and parents better understand student progress. By comparing students’ performance on standardized assessments against other similarly-advanced students, SGP provides educators insights into their students’ relative proficiency, allowing them to formulate plans for enhancing students’ learning.
This metric is an alternative to more traditional metrics such as mean and median scores that do not provide sufficient information on how well individual students are performing over time. Growth percentiles allow educators to identify students who require additional support while differentiating classroom instruction for high performing students and identifying the progress of those in the lower proficiency range. SGP also allows educators to make predictions about future student achievement.
Educators use SGP data to inform their instruction, evaluate their students/teachers and to support educator evaluation systems. While SGP analyses can be complex, they are useful tools for assessing and supporting the academic performance of all students. SGP is particularly valuable because it measures student progress in percentile terms that are more familiar to teachers and parents than standard deviations or other statistical metrics.
SGP uses the open source R software environment. R is available free-of-charge for Windows, OSX and Linux. The sgp package includes advanced functions that assume some familiarity with the program and its methodology, however, numerous resources exist on CRAN that can assist those new to SGP analysis.
As with all data analysis, proper preparation of the assessment data is critical to ensuring that the results are valid and reliable. The bulk of the time spent on SGP analyses is devoted to this step. Once the data is prepared correctly, the calculations and reporting are relatively straightforward. As a result, any errors encountered during the analysis process generally revert to issues in the data preparation process.
The data set sgptData_LONG is an anonymized panel data set of 8 windows (3 windows annually) of assessment data in LONG format for 3 content areas (Early Literacy, Mathematics and Reading). The sgpData_LONG file contains the 7 required variables for SGP analyses: VALID_CASE, CONTENT_AREA, YEAR, ID, SCALE_SCORE, GRADE and ACHIEVEMENT_LEVEL.
For those who are unfamiliar with longitudinal data analysis the vignette below walks through the basic steps for creating a SGP plot from the sgpData_LONG data set.
A growing number of educators are implementing SGP analyses and reporting. This is due in part to a greater awareness of the utility of this metric and the need for a more meaningful, transparent measure of student performance. While SGP is a step in the right direction it is important to remember that it should be only one piece of the educational data aggregation puzzle. To ensure that SGP is meaningful, it should be analyzed in conjunction with other data sources such as classroom observations and standardized tests. In addition, it is essential to examine the implications of the various data aggregations and to understand the limitations of each. To further discuss these limitations see the SGP Data Analysis Vignette.