Data sgp is an open source software package for the R statistical environment that facilitates running student growth percentile (SGP) analyses. It consists of a set of lower level functions (studentGrowthPercentiles and studentGrowthProjections) and higher level function wrappers that simplify the code required for running these analyses. The package requires the free and open source R software which is available for Windows, OSX, and Linux. As a software package that is designed to run within the R software environment, it assumes some familiarity with R and some time spent on the software to become comfortable using it.
Aside from the SGP functions, the data sgp package also includes a number of additional tools for performing a variety of analyses. This includes a function that produces a table report for comparing sites or teachers based on their multi-year weighted average SGP or overall SGP rating score. This provides a side by side comparison of NWEA SGP data that can be helpful for identifying trends over time.
Another useful function is one that converts a range of student growth trajectory points into a graph of those trajectories. This can be used to identify students who are close to proficiency and help to target interventions for those students. It also allows for the identification of a range of growth patterns that may not be consistent with proficiency or other desirable characteristics for a specific group of students.
A final important function is one that utilizes the historical growth trajectories of Star examinees to calculate projections of what their future scores might be given current performance levels. This can be used to predict the likelihood that a student will pass or fail a particular subject, as well as to determine what the minimum amount of future growth is needed for a student to reach or maintain proficiency.
This package also contains 4 example data sets for demonstrating how to use these SGP functions. The first, sgpData, specifies data in the WIDE format that’s used by the lower level SGP functions studentGrowthPercentiles and studentGrowthProjections. The next two data sets, sgpData_LONG and sgptData_LONG, specify data in the LONG format used by higher level functions like abcSGP, prepareSGP and analyzeSGP. The last data set, sgpData_INSTRUCTOR_NUMBER, is a teacher-student lookup file that’s utilized by the teacher level aggregate functions prepareSGP and analyzeSGP.
SGP analysis is generally straightforward following proper data preparation. However, many errors that may arise during the process can be traced back to problems in the data or its preparation. It is recommended that those who wish to conduct SGP analyses review the SGP data analysis vignette and the data preparation steps listed above before beginning to analyze any real data. In addition, any errors that may be encountered while analyzing SGP data should be carefully documented and reviewed to determine the root cause of the error. This will allow for more thorough troubleshooting and help to reduce the potential for error in future analyses.
