Data Sgp is an analysis tool for longitudinal student assessment data that creates statistical growth plots (SGPs) to show students’ relative progress compared to academic peers. These SGPs are calculated by combining students’ standardized test scores and covariate information with a growth standard established through prior testing history. This allows for more accurate measurements of student performance than traditional percentile scores do.
Students can use SGPs to track their own growth over time, and educators can use them as tools for assessing students/teachers, informing instructional practice, supporting classroom research initiatives, and evaluating schools/districts. Students can also share their SGPs with teachers and families to provide them with a picture of their growth.
SGP analyses require longitudinal student assessment data in either WIDE or LONG format. This data is typically stored in a database system such as MySQL, PostgreSQL, or Microsoft SQL Server. The SGPdata package provides example data sets for each of these systems and assists with converting from a WIDE to a LONG format.
The SGPdata package also provides a series of functions to calculate student growth projections/trajectories. These calculations use a method called quantile regression to estimate conditional density curves from student achievement histories and then uses these curves to project or trace the student’s future percentile score or trajectories.
In addition to creating student growth percentiles and growth projections/trajectories, the SGPdata package contains a number of other statistical methods useful in education research such as standard deviation analysis, group mean comparison, and the likelihood ratio test. The SGPdata package is available as a stand-alone application, as well as a part of the statistical software package R.
SGP requires a computer with the free, open source programming language R installed. R is compatible with Windows, OSX and Linux and many resources on CRAN provide help for getting started. Running SGP analyses uses advanced R functions so familiarity with the program is required. If you’re not familiar with R, it may be helpful to read the SGP Data Analysis Vignette before beginning. This will give you an idea of the kinds of data that will be needed to run the analysis and the kind of results you can expect. You can also contact a member of the SGP Development Team for more assistance. They are happy to help!
