Data sgp is a software package used to calculate student growth percentiles and projections/trajectories using large scale, longitudinal education assessment data. The data can be from standardized tests, portfolios, grading scales and/or other sources. It is an open-source package that was developed by Adam Van Iwaarden, Daniel Aguilar and Michael Kiesling. It can be used to perform a wide variety of analyses, including the comparison of student performance across different time periods, as well as the evaluation of educational policies and practices.
SGPs describe how much a student’s performance improved over time, relative to other students with a similar achievement history. The higher the SGP, the more progress a student made. SGPs can be helpful for assessing student learning, providing educators and parents with a better understanding of how a child is doing at school. They can also be useful in identifying students who are struggling and may need additional support.
In addition, SGPs can help to determine whether a student has achieved academic growth that is appropriate for their age and grade level. However, they do not provide information on whether the growth level would be considered adequate by stakeholders in the educational system or whether a particular level of achievement is sufficient for a student to meet their potential.
As a result, SGPs cannot provide the same kind of individualized information that VAMs can. They are not a good fit for use with current accountability systems that emphasize test score-based measures. However, they could be a good fit for future accountability systems that focus more on student growth and development.
The most important step in using SGPs is preparing the data to be analyzed. This involves converting the data to the right format. There are two common formats: WIDE and LONG. In the WIDE format, each case/row represents a single student, while in the LONG format, each time dependent variable is associated with a unique student at multiple times. The SGPdata package installed when you install the SGP package includes exemplar WIDE and LONG formatted data sets (sgpData_WIDE and sgpData_LONG, respectively) to assist in setting up your data for analysis.
The previous section established that true SGPs for a student are correlated across math and ELA and that they are related to student background characteristics. In this section, we investigate how well these relationships can be exploited to improve the accuracy of the SGPs estimated for individual students. The figure below shows the error distribution for a number of estimators of e4,2,i, conditioned on varying amounts of data. The RMSE decreases as the reliability increases.