How to Calculate Student Growth Percentiles (SGP)

data sgp

Student Growth Percentiles (SGPs) provide an indication of student achievement relative to their academic peers. Similar to standard percentiles, SGPs help identify students who fall above or below grade level peers as a way of differentiating learning for all learners–including those with special needs.

SGP (Student Growth Percentile) measures students’ academic growth relative to their peers with similar initial test scores, with higher numbers signifying greater growth. Teachers use this information to understand what it will take for students to reach their goals and meet them successfully.

ARM’s Student Growth Platforms (SGPs) offer educators tools that assist in accurately assessing student progress and pinpointing areas for improvement within classrooms. SGPs can be particularly beneficial in identifying how many students must make enough gains in order to “make it over the curve” and meet state or district goals.

SGPs are calculated using multiple years’ assessment data collected on students over time, typically organized in WIDE or LONG formats. When installing the SGP package, an exemplar data set such as sgpData becomes available that represents each case/row as one student and each column as information associated with that student at various points during their educational career.

The ID column in sgpData data set serves as a unique identifier for every student, while its five other columns — SS_2013, SS_2014, SS_2015, SS_2016 and SS_2017 — display each student’s scale score in each subject area in each year; furthermore, its column sgpData_INSTRUCTOR_NUMBER lists the instructor number assigned to teach that particular course during one of those years.

AVERAGE_SGP represents the average student growth over these years. An AVERAGE_SGP of 50 indicates that student’s growth over this time was equivalent to their academic peers with similar prior test scores; below or above 50 indicates their growth was above or below average respectively for this same group of students.

To accurately calculate a student’s Student Growth Percentile (SGP), teachers require at least three years of stable data and two different subject matter tests. As part of educator evaluation practices, using SGPs requires both more time and data than using baseline-referenced SGPs in some districts; furthermore, correlations between baseline SGPs and prior year scale scores will likely not be exactly zero, making interpretation of results even more complex.