Data SGP – Using the Data SGP Package to Analyze Longitudinal Education Data

The data sgp package, built on the R software environment, provides classes and functions used to conduct Student Growth Percentile analyses using large scale, longitudinal education assessment data. The package calculates students’ current achievement status, growth projections/trajectories to reach future achievement targets and a comparison of the student to their academic peers nationwide. The growth models used by the SGP package rely on quantitative regression analysis to estimate conditional density and derive student-level coefficient matrices which are then used to produce percentile growth projections/trajectories for each student.

The SGP package is available on CRAN and requires the free and open source software R to run. R is available for Windows, OSX and Linux and can be installed via a package manager. It is recommended that you become familiar with R prior to beginning your SGP analyses. It is also necessary to have a computer capable of running the R software. While it is possible to conduct SGP analyses on a standard desktop or laptop, it is not advisable due to the computational demands of the software.

Depending on your needs and the size of the data set being analyzed, it may be more appropriate to format the data in WIDE or LONG formats. The WIDE format is used by the lower level functions, studentGrowthPercentiles and studentGrowthProjections whereas higher level wrapper functions, such as abcSGP and updateSGP, require the LONG format. For operational analyses, we recommend using the LONG data format as this makes managing the data more straightforward in the long term.

In addition to the basic functions in the SGP package, the prepareSGP function provides a number of additional options which can be used to enhance the utility of the package for your purposes. This includes the option to use a student-instructor look up file, sgpData_INSTRUCTOR_NUMBER, to produce teacher level summaries in place of individual student reports. The prepareSGP function also supports the use of a custom trajectory variable to provide unique trajectories for each student.

The final argument, sgpData, specifies the location of the school folders where the studentGrowthPlots are produced (default is “Visualizations/studentGrowthPlots”). Using this argument allows users to zip their schools’ folders for faster access to studentGrowthPlots. A boolean argument indicating whether to anonymize studentGrowthPlots by removing the names of students, teachers and schools. This is a significant performance improvement and we are currently working on parallelizing this functionality to further reduce report production time.