Data sgp is an R package for generating student growth percentiles and percentile growth projections/trajectories from large scale, longitudinal education assessment data. It uses quantile regression to estimate the conditional density of a students achievement history and derives coefficient matrices that can be used to generate percentile growth scores.

A key problem faced by many organizations is the issue of data sprawl. This refers to the excessive number of siloed and disparate datasets within an organization that make it difficult to gain a single, consistent view of a company’s data. This often leads to a situation where IT resources are spent managing these disparate sources instead of using them for strategic business initiatives. Data sprawl is a major impediment to the success of a data-driven culture.

In the field of education, there is a strong need to provide more useful and actionable information about student progress. Student growth percentiles (SGP) are a way to communicate this information in terms that are easily understood by teachers and parents. SGP scores compare a student’s performance on a given assessment to the performances of academic peers who have similar previous test scores.

The data sgp package provides an efficient way to create, analyze and share SGP scores. It also includes features to facilitate the comparison of student performance across schools and districts. These capabilities are particularly valuable in a time when states and districts are seeking to improve educational outcomes by making data-driven decisions.

Data sgp is an open source, platform independent application that runs on the statistical software R. This package is designed to be used with long term, longitudinal data sets such as those found in a school district’s student database or a state education agency’s longitudinal system. The package also allows for the creation of a wide range of visualizations and statistical output.

Unlike full community databases like Genbank and EarthChem which aggregate and store essentially all data, research consortia assemble and share a limited subset of data for specific scientific questions. The resulting data is still massive in scale but in comparison to the amount of data generated by a typical social network it is relatively small potatoes.

To use the data sgp package, users must have an R installation on their computer. R is a free, open source program that can be compiled for any operating system. If you do not already have it installed, there are several resources on CRAN to get started. There are also several online tutorials available. These tutorials cover both the basic programming skills needed to use R as well as the more advanced topics required for running SGP analyses. These online courses are an excellent resource for learning the syntax of the language and getting up to speed quickly. Ideally, it is recommended that you take one of these courses before you begin working with the data sgp package. This will help ensure that you have a firm understanding of the tools and techniques required to run your SGP analyses.