Data SGP leverages longitudinal student assessment data to produce statistical growth plots (SGPs), which measure students’ relative progress compared to academic peers. This metric, which is typically reported as a percentile, is useful for gauging whether students are growing more than, less than or about as much as their academic peers, and can be used to inform classroom practices, evaluate educators, and support broader research initiatives.
To generate SGPs, educators must have access to high quality longitudinal student assessment data that includes student identifiers and grade level and content area associated with each student’s individual assessments. SGPs are only valid if they are created from longitudinal student assessment data and if the data has been vertically scaled to produce an accurate, meaningful measure of student growth. The sgpData data set includes a variety of variables that meet these requirements, and an analysis function that allows users to create SGPs from this data.
Currently, many districts use data sgp to identify areas for improvement in their instruction and to provide feedback to teachers. While SGPs can be an important tool to improve student achievement, there are a few key issues that must be addressed before this method can become more widely adopted.
The most significant issue is the need to have access to longitudinal student data in formats compatible with SGP analyses. While some districts, such as Macomb and Clare-Gladwin ISDs, have made their SGP data publicly available in formats compatible with operational SGP analyses, the vast majority of districts do not have access to such data.
In addition to having access to longitudinal student data, districts need a system of identifying instructors for each student. Currently, some districts assign students to multiple instructors for each content area in which they are assessed, and determining which instructor’s SGPs to use is a time-consuming process. The sgpData_INSTRUCTOR_NUMBER field in the data sgp table addresses this issue by allowing users to connect student records with instructors using unique identifiers associated with each student’s assessment occurrences.
Finally, if the goal of using SGPs is to evaluate teachers, it is essential to use baseline-referenced SGPs rather than cohort-referenced ones. While cohort-referenced SGPs are more comparable across teacher evaluation systems, they are also less accessible to districts, as they require three years of stable assessment data and a correlation between students’ current year test scores and the previous year’s scale score. Such a correlation is likely to be imperfect, potentially introducing substantial bias into interpretation of the resulting SGPs.