Data sgp is a collection of software functions that enable the rapid preparation and analysis of student growth percentile (SGP) and growth projection data. These calculations are used to provide insight into student achievement progress over time and can be used by educators and administrators in informing instruction, assessing teacher/student performance, and supporting educator evaluation systems.
The SGP calculations are very different from most assessment metrics as they do not depend on an absolute score scale or fixed value and instead report a student’s raw score growth in relation to the scores of students with similar prior test results (their academic peers). Educators can then use this information to inform instructional practices, support classroom research initiatives and evaluate teachers/students.
SGP analyses require large amounts of longitudinal assessment data and a lot of computing power. Therefore, we need a database that allows for the storage of these large datasets and easy access to data for analyses. While a database is not an ideal solution for all data management issues, it can serve as a centralized repository for many of the SGP calculations.
GitHub is a popular platform for collaborative work on projects and provides an ideal platform to host a database that can be easily accessed by researchers. The GitHub SGP database will be a collection of metadata, legacy data and analytical results that will be curated over time to help researchers access the information they need. It will also act as a prototype for other geochemical databases that may need to be created in the future.
The data set sgpData is an anonymized, panel data set that contains 5 years of annual, vertically scaled assessment data in WIDE format. This exemplar data set models the format of the data used by the lower level SGP functions, studentGrowthPercentiles and studentGrowthProjections.
SGP uses a statistical model to estimate a student’s raw test score growth and projections over time. The model is fit using a linear regression algorithm with student covariates including gender, race, socioeconomic status and grade level. This calculation can then be compared to the state average student score growth and projections to determine if a school is on track for its desired learning outcomes.
The higher level SGP function, studentAchievementPlots provides a variety of visual representations of the student’s SGP and projection data. These plots can be displayed as pdf’s, png’s or JSON files and can be customized by user preferences. The plots can be sorted by year or by student to identify schools where further intervention is needed. The output can be saved to a folder in the /Visualizations/growthAchievementPlots directory. Arguments are available to specify a print or presentation style, and to set the number of years to display for each student in the plot. In addition, an argument can be set to use either the default order of coefficient matrices or to utilize the maximum order that is available with the coefficient matrices. Arguments can also be set to zip the school folders containing the studentAchievementPlots.