Data SGP – A Basic Overview
Data sgp is an analysis tool for the statistical software environment R. It provides classes, functions and data for estimating student growth percentiles and percentile growth projections/trajectories using large scale, longitudinal education assessment data. The methodology uses quantile regression to estimate the conditional density of a students achievement history and then projects their progress towards future learning goals. This is done by calculating the conditional densities of their current and target scores with the help of derived coefficient matrices.
SGP analyses are intended to be straightforward and simple. They can be done on a desktop computer with minimal programming knowledge and the software is free of charge with many online resources available to help get users started. However, there are a few important concepts that newcomers to SGP should be familiar with before attempting to use it. The bulk of the time spent on a SGP project is spent on data preparation and analysis. This article will provide a basic overview of the required software/hardware, data preparation and analysis steps involved in running SGP analyses.
Unlike most statistical software applications, which require a subscription or a license to operate, the open source statistical program R can be used on any computer without a cost or limit on the number of users. While the software is designed for statistical analysis, it has also been used by educators to create SGP analysis tools that have broad applicability in the educational setting. This makes it a unique and valuable resource for those interested in assessing the impact of their programs on students’ academic performance.
To begin using data sgp, users must have access to a statistical software application and a set of longitudinal educational assessment (SETA) records. There are several ways to obtain these records, including a searchable data portal that allows educators to find and download individual student data sets. In addition, most state education agencies have websites that provide educators with aggregated student data. Educators can also create customized reports with student data by selecting individual students in the searchable database and choosing the SGP Data tab.
The SGP Data tab in a student report displays the SGP data for that student over five years. The first column, ID, provides the student’s unique identifier. The next five columns, SS_2013, SS_2014, SS_2015, SS_2016, and SS_2017, provide the assessment score associated with each of the student’s five years of testing.
Using wide format data such as sgpData with the higher level SGP functions is relatively straight forward, but we recommend that users consult the sgpdata vignette for more comprehensive documentation on how to manage this type of data set with the SGP package. Specifically, note that the SGP functions often assume the presence of state specific meta-data embedded within the sgpdata. This is because the function’s internal mathematical models rely on this information to calculate students’ growth percentiles and projections. This is why we recommend that users always use SGP’s sgpdata exemplar data sets for operational analyses.