The data sgp tool compares student assessment scores against academically similar students (their educational peers) to gain insights into a students relative performance and more fairly compare those entering school at different levels. Student Growth Percentile, or SGP, measures how far a student has come over time by comparing her latest score against that of students with similar prior scores histories – it indicates how they compare with one another over time.
SGP employs historical growth trajectories of Star examinees to provide insight into what is possible for students, such as the percentile trajectory at which she should aim in order to meet her desired grade level achievement target. It can help educators identify areas for growth as well as devise a plan of action to support their pupils.
An SGP score for any student is calculated by comparing their most recent assessment score to the average of their last three assessments. To accurately conduct SGP analyses, assessment scores must be available in WIDE format and assigned to one or more teachers listed in a data set containing instructor names for all tests they have taken.
This data set, known as sgpData, provides an example of the data structure required for SGP analysis. It features five years of annual vertically scaled assessment data as well as teacher lists for every student in its sample population. Additionally, it serves as a template that will be utilized by studentGrowthPercentiles and studentGrowthProjections functions to produce SGP projections and percentiles for use with analysis tools like Student Growth Percentiles.
Though designed for use within R’s statistical software environment, you can still utilize its data sgp package with other programs. As long as your computer can run Windows, OSX or Linux and you have access to R’s open source version there are numerous online resources to assist in getting you started with statistical data analysis.
SGP project’s primary objective is to collect and generate multiproxy sedimentary geochemical data for Neoproterozoic through Paleozoic periods. SGP members have taken an enormous effort in compiling and creating this large data set, along with developing analytical tools to utilize it effectively. This work continues. Next steps in this process involve developing new models of geochemical and geological processes operating during these periods, then testing these models against the data sgp data set. Once developed, interpretation of significance for each epoch of data sgp becomes possible – though additional efforts will need to be put towards collecting, processing, and creating analytical tools; our efforts going forward will mainly concentrate on these last two steps of this process.