Data sgp is an R package that offers an efficient means for organizing longitudinal (time dependent) student assessment data into statistical growth plots. Specifically, it provides class and function support for calculating student growth percentiles and projections/trajectories using large scale, longitudinal education assessment data. SGP’s classes and functions use quantile regression to estimate conditional density matrices for each student, and then use these matrices to calculate their percentile rank.
While it can be challenging to understand the meaning of student growth percentiles and projections/trajectories, once these have been calculated they can provide useful insights into student progress. Teachers and families can use these reports to quickly gauge a students’s performance against similar-aged peers, helping them to tailor instruction. For example, if a student’s mathematics score falls below the 3rd percentile, it could be a sign that they need additional assistance in this area.
This article will describe the classes and functions in the data sgp package, as well as provide examples of how to use them. Additionally, we will provide a brief overview of the SGP methodology and the underlying assumptions behind its calculations. We will also discuss the benefits and limitations of SGP. Finally, we will provide some examples of the graphical displays that can be produced with this data set and function.
The data sgp package contains the sgpData_WIDE and sgpData_LONG data sets. The WIDE data format allows each case/row to represent one student, and its columns to represent variables associated with the student at various times; whereas LONG spreads time dependent variables across multiple rows per student. The higher level wrapper functions that make up the sgpData_PERCENTILES and studentGrowthProjections functions utilize the LONG data format, which offers numerous preparation and storage advantages over WIDE data.
If you are interested in exploring the details of this package we encourage you to download and play with sgpData_PERCENTILES. This is an easy-to-use, interactive tool that can produce many different graphical displays of your student’s growth and achievement. The sgpData_PERCENTILES function takes the following data as input: sgpData_WIDE – a data set that includes student achievement results from at least two tests taken in different windows. To create the charts displayed below, the sgpData_PERCENTILES data set must include a minimum of one test from a prior window and a maximum of three tests from a current window. If you do not have this data available you can use a default year for the charts or choose not to generate any graphs at all. Alternatively, you can also generate a single chart that contains all of your student’s data by entering the name of your assessment data in the sgpData_PERCENTILES parameter. The function will then automatically display all of your student’s data on a single chart. You can then highlight individual students and their scores to see a visual representation of your data. You can also change the type of graph to a scatter plot or a bar chart by changing the chartType parameter.