How to Analyze Data SGPHow to Analyze Data SGP
The Student Growth Percentile (SGP) analyses compare a student’s academic performance to that of his or her academic peers nationwide. These peers are students in the same grade with a similar achievement history on Star assessments. SGP is a longitudinal analysis – it projects a student’s potential growth trajectories over time, allowing the user to see how much growth a student will need to make to reach proficiency and/or maintain proficiency.
Data SGP can be analyzed in two ways: WIDE and LONG formats. The WIDE format consists of a single table where each case/row represents a unique student and the columns represent variables associated with the student at different times. The LONG format consists of multiple tables where each row represents a different time point for the same student. The SGPdata package, installed when you install the SGP package, provides exemplar WIDE and LONG formatted data sets (sgpData, sgpData_LONG, and sgptData_LONG) for use in setting up your own longitudinal data for SGP analyses.
Both the WIDE and LONG formats are used by the lower level SGP functions studentGrowthPercentiles and studentGrowthProjections – however, for operational analyses, it is recommended that you format your data in the LONG format. This will provide numerous preparation and storage benefits compared to using the WIDE format.
Whether you choose to utilize the WIDE or LONG format is determined by your specific needs and the size of your data set. We recommend that for most users of the SGP package, you begin with a LONG data set to gain the greatest advantages from the SGP analyses.
SGP is a free, open source software application that calculates a student’s expected progress along their educational pathway by comparing the student to their academic peers nationwide. It does this by comparing the student’s current test results with the performance of their academic peers. SGP analyses use historical growth trajectories for each student, and project what the expected progression is over time, based on those trajectories, as well as the amount of additional growth needed to meet or exceed the national average for proficiency.
The SGPdata packages include a number of higher level wrapper functions that simplify the source code for operational SGP analyses. These are useful for constructing larger models that are more complex than what can be accomplished with the lower level SGP functions alone. These wrapper functions are also useful for creating customized SGP reports that can be generated for specific subsets of students or for different time periods. These functions are available to all users of the SGP package and may be downloaded from the SGP website. We encourage you to provide feedback on these functions, and are happy to answer any questions. Please contact us via our SGP forum. Thank you!