How to Interpret Student Growth Profile (SGP) Data

An SGP (Student Growth Profile) is a time series of data that illustrates student progress over time, providing educators with information to better understand differences in performance between children. Data SGP may also be aggregated for reporting or research efforts across schools or districts.

Data SGP utilizes longitudinal student assessment data to generate statistical growth plots that measure students’ relative progress relative to their academic peers. Students are organized according to prior test scores and this can be used to ascertain how many of them can expect certain levels of progress – for instance a student expected to gain 75% or more than their academic peers may be considered having made satisfactory progress.

SGP includes two data sets in WIDE and LONG formats to assist with setting up student growth analysis. sgpData_LONG provides anonymized long-term assessments in 8 window (3 windows annually). Meanwhile, sgpData_WIDE contains similar information but with individual student identifiers and separate variables per assessment item that allow you to create student achievement plots across time windows by analyzing individual cases.

SGP provides its most valuable feature to most users when it comes to projecting percentile growth trajectories for individual students using historical data, and displaying them in Star Growth Report under Window Specific SGP when customizing reports (in the Timeframe drop-down menu) when choosing either previous or current year when customizing reports (Star Growth Report under Window Specific SGP). These graphs can help users quickly spot any issues with specific students while developing instructional plans to address any potential concerns that may exist in class settings.

Percentile growth projections/trajectories provide more accurate results than student growth measures alone, because they take into account both individual differences as well as test score variability. Students with similar scale scores may still have very different SGPs due to differences in underlying ability that cannot be captured with just one test score.

As such, it is vital that high quality data be utilized when developing SGPs. Unfortunately, interpreting such complex and large estimation errors can make interpretation challenging; fortunately there are ways to mitigate risk when creating SGPs from standard testing data by making sure it has been designed well and analyzed appropriately.