Data SGP is a free, open source software package designed to facilitate the processing of longitudinal education assessment data. The package provides classes, functions and data that allow users to calculate student growth percentiles and project/create student growth trajectories using large scale, longitudinal educational assessment data. The package uses quantile regression to regress the conditional density of each student’s assessment record against the student’s current achievement level and then based on the regression results, produces the resulting coefficient matrices that are used for calculating student growth percentiles.
Educators can access the SGP software and download individual student reports by visiting their state’s website. Each report includes detailed information on a student’s performance relative to their academic peers across the nation and is based on the students most recent Star assessment combined with one prior test from another testing window (Fall, Winter or Spring). In addition to providing an accurate picture of each students performance, SGP helps teachers identify low achieving students so that they can provide the necessary assistance. It also enables educators to assess accelerated programs so that they can be sure that most students are making enough progress towards meeting their achievement goals.
SGP data is provided by state departments of education and by large community databases such as those hosted by the National Center for Education Statistics. However, it is important to note that while research consortia and full community databases aggregate data for many purposes and share it, the approaches and goals of these initiatives are different. Research consortia collect and make accessible only a small subset of their data, whereas full community databases seek to store and provide essentially all available data.
When analyzing SGP data, it is important to ensure that the proper data preparation steps have been taken. Almost all errors that are encountered during analysis of SGP data revert back to problems with the original data or data preparation.
In order to perform SGP analyses, a data set must be in one of two common formats for longitudinal (time dependent) student assessment data: WIDE or LONG format. The SGPdata package, installed when you install the SGP package, contains exemplar WIDE and LONG data sets to assist in setting up your own data sets. The lower level functions in the SGP package, such as studentGrowthPercentiles and studentGrowthProjections, require WIDE formatted data whereas the higher level functions (wrappers for these lower level functions) are intended to be used with LONG formatted data.
For more information about SGP and a demonstration of its use, please visit the SGPdata homepage. Additionally, the SGPdata user manual, written by Adam Van Iwaarden and Jason A. Hoefle, is a great resource for anyone interested in using this powerful tool. The manual describes the data structure required to perform SGP analyses as well as the methods for preparing and analyzing SGP data. It is a must-read for anyone planning to use SGP in their work. The manual is also available in pdf format.