Data Sgp is a software package that provides the classes, functions and data used to calculate student growth percentiles and percentile growth projections/trajectories using large scale, longitudinal education assessment data. SGP is designed to help educators gain a better picture of student progress to more accurately identify areas for improvement within classrooms and to assist in meeting the needs of all students effectively.
SGP is working to assemble an unprecedented amount of data but in comparison to the data being gathered by research consortia and full community databases such as Genbank or EarthChem it remains quite modest in size. This means that, for most analyses conducted with SGP, the bulk of the work is in data preparation and the analytical process itself is relatively straightforward.
The SGP analysis pipeline is built on top of the Python data management and processing framework, Scipy. The pipeline takes in data from multiple sources, performs preprocessing and transformation, and then outputs a series of output files for further processing. This pipeline is used in conjunction with other tools, including scripts written in R and python, to perform various analysis tasks. These scripts can be found in the sgp_scripts github repository.
Students are rated on their current grade-level test score and compared to other students with similar scores and history in order to determine how far they have progressed toward reaching official state achievement targets or goals. Educators use this information to see whether students are performing inside or outside of the curve and then adjust their instruction accordingly.
In addition to the SGP report that is provided for each individual student, many states will include a data dashboard for their districts which can be accessed by all educators. The dashboard allows educators to view a summary of all students at the district level and to select subsets of students by specific characteristics. This allows for more focused attention to students in need of assistance.
Aside from being a useful teacher tool, the SGP system has a number of unique features that set it apart from competing methods. One key advantage is that it can identify which students are close to passing a subject by calculating the percentage of students who have scored above or below a given cutoff point in the most recent test section.
Another important advantage of the SGP system is that it uses actual performance data from the MCAS exam, rather than estimates based on past performance or predictive models. This is a significant improvement over previous systems that have been plagued by false positives due to biases in baseline cohort design and school/teacher characteristics.
As mentioned above, the SGP analysis workflow is very simple and the system is flexible enough to allow for a wide range of possible analysis. To get started, simply download and open sgpData for a particular student. The first column, ID, identifies the student by their unique identifier and the subsequent 5 columns, GRADE_2013, GRADE_2014, GRADE_2015, GRADE_2016 and GRADE_2017, provide the grade levels of the student assessment scores from each of the last five years.