Student growth percentiles (SGP) describe relative performance in comparison to students with similar previous test scores (their academic peers). SGP measures a student’s progress over time and provides teachers and administrators information they can use to determine whether a student grew more than, less than or as much as expected.

To run SGP analyses, you need a computer with the R software environment installed. R is free and available for Windows, OSX and Linux. More information about the R software environment can be found at CRAN, the repository for open source software. You also need to be familiar with navigating the command line. The SGPdata package provides an interface to R and makes running SGP analyses very easy, but it still requires some knowledge of command line syntax.

The SGPdata package contains 4 examplar data sets that are used for SGP analyses. The first, sgpData, specifies data in the WIDE format that’s used with the lower level studentGrowthPercentiles and studentGrowthProjections functions. The second and third data sets, sgpData_LONG and sgptData_LONG, specify the LONG format that’s used by higher level functions like abcSGP, prepareSGP and analyzeSGP. The fourth data set, sgpData_INSTRUCTOR_NUMBER, is an anonymized teacher-student lookup table utilized to produce teacher-level aggregates.

A student’s SGP score is based on their performance in relation to academic peers from other schools in the state. This means that a student’s SGP score may be different from their score in another school because the growth model used to calculate SGP takes into account growth from all students in the state, not just those from one particular school.

SGPs are measured on a 1-99 scale; lower numbers indicate lower relative growth, and higher numbers indicate greater relative growth. For example, a student with an SGP of 85 would have scored higher on this year’s MCAS than about 75 percent of their academic peers.

It’s important to keep in mind that a SGP score is only valid for the current testing window. If a student’s SGP score changes in a future window, the new SGP score will be used to measure their growth in that time period.

In addition, differences in SGPs between years need to be interpreted with caution because SGPs are calculated anew each year. For this reason, a difference in SGP of fewer than 10 points between years should not be considered significant. For additional guidance on interpreting and reporting SGPs, see the technical resources on the Student Growth District and School Resources webpage.