How to Use Data SGP
Data sgp leverages longitudinal student assessment data to produce statistical growth plots (SGPs) that provide visual evidence of student progress relative to their academic peers. These data products can be used by educators to inform classroom instruction, evaluate teachers, and support broader research initiatives. SGPs are constructed from students’ standardized test scores along with covariate information using an “academic peer group” established through their prior testing history. The resulting SGPs provide more accurate measurements of student progress than standard percentile scores do.
Data SGP provides educators and parents with an easy to understand measurement of their students’ relative performance. A student’s growth percentage indicates whether they are growing faster, slower, or about as fast as their academic peers. This metric can help educators identify underperforming students and target their instruction, differentiate classrooms for high performing students, and determine the effectiveness of accelerated programs.
SGPs also offer a unique advantage over other traditional metrics by providing educators with the ability to link teacher/student performance to official state achievement targets/goals. This capability is not possible with standard growth models or other methods and can be very powerful for motivating teachers by linking their performance to measurable goals.
How to use data sgp
Running SGP analyses requires access to longitudinal student assessment data in WIDE format. This type of data is typically stored with each row and column representing a single student over time. The SGPdata package provides an example WIDE format data set (sgpData) to simulate time dependent data for use with lower level functions like studentGrowthPercentiles and studentGrowthProjections as well as a LONG format data set (sgpData_INSTRUCTOR_NUMBER) to assist in converting existing longitudinal data sets into SGPdata format.
As a computationally intensive tool, running SGP analyses can require substantial computing resources and is not recommended for casual users or those without extensive computer science experience. The bulk of the analysis is spent on data preparation which requires a significant amount of memory and processing power.
SGP analyses can be performed on most major operating systems including Windows, OSX and Linux. However, we strongly recommend using a computer that has at least 8GB of RAM for optimal results. Additionally, SGP calculations can run much faster on computers with multiple cores.
Running SGP analyses requires the software R, which is available free of charge for most major operating systems and is widely considered to be the standard for open source scientific computation. If you are not familiar with R we highly encourage you to spend some time getting acquainted before diving into SGP analysis. You can find more information about R on CRAN and there are a number of great resources for new users on our Get Started page. We recommend you also read the SGP documentation and FAQs before you begin. This will help you better understand the SGP methodology and how to use the SGPdata package. We are always happy to answer questions and will assist you in any way that we can.