Data SGP
Data SGP is a collection of student growth and progress data aggregated over time, used to better understand student achievement and learning. It includes individual-level measures like test scores and growth percentiles, and it also includes a set of aggregated school and district level reports that help educators, parents, and policymakers track student progress over time and across schools.
The SGP package contains classes and functions for calculating student growth percentiles (SGP) and percentile growth projections/trajectories using large scale, longitudinal education assessment data. It uses quantile regression to estimate the conditional density associated with each students achievement history, and to construct coefficient matrices from which SGP can be calculated.
SGP compares a students relative performance to that of academic peers nationwide. The SGP score for a given student is the percentage of his or her academic peers who scored higher on the same assessment. Teachers and administrators use SGP to determine whether a student has grown more, less, or at the same rate as their peers, as well as to predict the likelihood that a student will reach future achievement targets.
Located on 160 acres of cattle pasture and wheat fields, the core of the SGP observatory is the heavily instrumented Central Facility southeast of Lamont. In addition to monitoring continuous observations, scientists conduct a variety of research projects at the site including single observation analyses, multi-observation process studies, and assimilation into Earth system models. Researchers supplement the ongoing observational effort at SGP by using guest instruments during field research campaigns and by requesting increases in the frequency of existing instruments.
In addition to a standard set of aggregations and reports, the SGP package also provides four exemplar data sets: sgpData, sgpData_LONG, sgptData_LONG and sgpData_INSTRUCTOR_NUMBER, that serve as templates for users to setup their own data for SGP analyses. The data set sgpData is a WIDE formatted dataset that serves as an exemplar for how to set up your own SGP analysis, the lower level functions studentGrowthPercentiles and studentGrowthProjections require a WIDE formatted data set while the higher level functions, which act as wrappers for the lower level function require a LONG formatted data set.
When used properly, SGP analyses are straightforward and nearly all problems that arise while analyzing SGP data revert back to issues with the data itself. In addition, the SGP team is always happy to provide help and answer questions on the github issue tracker. We love feedback and welcome suggestions for additional vignettes and analyses. Please don’t hesitate to contact us!