Student Growth Percentiles (SGP) measure how much a student has learned each year relative to his or her academic peers who scored similarly on previous assessments. SGP are vitally important metrics for teachers as it allows them to celebrate the progress their students are making even when they don’t yet meet grade level standards; plus it identifies which individuals require extra support and intervention so they can help ensure they remain on track with their studies.
SGPs are calculated with a statistical model that describes the relationship between students’ current test scores and previous assessments, which they had taken, and their subsequent MCAS test scores. They range from 1-99 with higher numbers indicating greater relative growth. As an example, students A and B scored identically on this year’s MCAS exam in one subject area yet have different SGPs due to having different academic peer groups assigned.
NJDOE uses NJ SMART district course roster submission data with teacher assessment data in order to calculate an mSGP, with each teacher’s total being calculated as the sum of individual student mSGPs associated with his or her class; this calculation occurs each summer following each statewide testing season.
Teachers can access student Growth Percentiles for their classes using the sgpData and sgptData_LONG data sets, both anonymized panel datasets that contain 5 years of annual, vertically scaled assessment data in WIDE format. The former serves as an example for how to prepare and load long formatted data into these analyses while the latter gives guidance for doing so.
In sgpData, ID contains unique student identifiers while the next five columns provide scale scores from assessments conducted over the past five years. For increased flexibility, sgptData_LONG contains these variables along with demographic and categorization data needed by higher level studentAggregates functions.
SGP rankings are calculated annually and should not be compared between years. For instance, a student with an SGP ranking of 50 in 2022 might have improved to 52 by 2023 without significant differences being noted between their ranks; these differences should not be reported as evidence of increased or decreased academic abilities.
As such, these data should not be reported within a comparison between districts or schools; rather, these insights can be used to examine how students are changing over time and across classrooms. Renaissance Star reports include both achievement and growth scores for Renaissance Star reports, in addition to SGPs. To interpret these results correctly, please review one of three stakeholder-specific guides outlining sample MCAS Growth reports for administrators, educators, or families. This page also links to technical resources related to these reports, including the MCAS Growth Wiki. This wiki offers examples on how to use the new MCAS Data Explorer for exploration of data. As it’s an open and collaborative resource, we welcome contributions from anyone looking for new ways to explore and explain various data visualizations available.