Data SGP

data sgp

Data sgp is the collective of aggregated student performance data collected over time that teachers and administrators use to make instructional decisions. It includes individual-level measures like test scores and growth percentiles, as well as aggregated measures at the school/district level such as class size, attendance rates, and graduation rates. Data sgp can help educators identify areas for improvement, inform classroom practices, evaluate schools/districts, and support broader research initiatives.

Student growth percentiles (SGPs) are an important measure of students’ relative progress compared to their academic peers. They are calculated by comparing students’ prior test scores to a growth standard – which is determined by taking a subset of their prior score distribution and scaling them up/down to the average value of this distribution. The results are then plotted on a chart to show the distribution of this subset.

Creating SGPs from longitudinal student assessment data is a complex process. The resulting estimates are often unreliable and have large estimation errors. In addition, generating these plots is labor intensive and time consuming. The SGPdata package makes this work easy and provides results that are reliable and comparable.

SGPdata provides a series of functions for SGP analyses: prepareSGP Takes an exemplar wide format data set, sgpData, and a student-instructor lookup table, sgpData_INSTRUCTOR_NUMBER, to generate the master longitudinal record, Demonstration_SGP@Data. analyzeSGP Performs SGP analyses for all content areas and years, generating student growth percentiles, student growth projections, and lagged student growth projections. combineSGP Merges the results derived from analyzeSGP back into the master longitudinal record, Demonstration_SGP@Data, and calculates/creates scale scores associated with SGP targets.

For each SGP analysis, SGPdata also creates a table of results. This table shows the percentiles for each student in a particular content area, the mean of these percentiles, and the median of these percentiles. The SGPdata table can be sorted in various ways to facilitate exploration of the data.

The SGPdata package is freely available on GitHub and requires Python 2.7 or later. Please see the documentation for more information about using SGPdata.

The sgpdata package is intended to be used by anyone analyzing student assessment data and wanting to produce SGP graphs. The goal is to make the process of creating these graphs as easy and consistent as possible. We do this by providing a series of functions that automate much of the work. This way users can focus on the analysis and interpretation of the results rather than the tedious task of preparing the data. If you have questions or comments about the SGPdata package, feel free to contact us by submitting an issue on GitHub. We will respond as quickly as we can.