Student Growth Percentiles (SGP) are a new statistic that provide teachers and administrators with information on the relative academic growth of their students. This data is intended to be considered alongside scaled scores and achievement levels as part of a holistic picture of student learning.

In calculating SGP, students are compared to other student in their grade with similar academic peer groups based on previous years’ performance on Star assessments. Academic peers are identified using a statistical procedure that identifies students with similar score histories, regardless of demographic groupings or special education needs (e.g., sheltered English immersion, multilingual learning).

While SGP is a useful indicator of student learning, it should be used in conjunction with other measures such as scaled scores and achievement levels to gain a fuller picture of student learning. In addition, the results should be viewed in context of the rigor and expectations for the particular subject area and school year.

The data sgp package provides 4 examplar data sets for use with SGP analysis. One of these, sgpData, provides the wide-format data that is passed to lower level SGP functions such as studentGrowthPercentiles and studentGrowthProjections. Two other data sets, sgpData_LONG and sgptData_LONG, provide the long format data needed by higher level SGP functions such as abcSGP and prepareSGP. The fourth data set, sgpData_INSTRUCTOR_NUMBER, provides an anonymized teacher-student lookup table that is used by the analyzeSGP function to produce teacher level aggregates.

Q. How should I interpret my student’s SGP?

SGP indicates how much a student’s performance improved from one year to the next compared to the average performance of their academic peers. A high SGP means that the student’s performance exceeded the average for their academic peers, while a low SGP indicates that the students’ performance was below the average for their academic peers.

SGPs are calculated based on trends in statewide performance over time, so they may shift slightly from year to year. This is particularly true during periods of time when a large number of students may be absent from schools due to the Covid-19 pandemic.

SGOs are computed by linking district course roster submission data with relevant mSGP data for each teacher. The NJDOE has provided a tool to convert mSGPs to a 1 to 4 teacher evaluation rating. The link can be found on the Teacher Evaluation Resources and Trainings webpage in the Documents for Download section.

The data sgp package is available for districts to use with their own customized SGP analyses. To ensure the best results, it is important to use the most recent data possible for each subject area and grade level. This includes using the most recent assessment, if available, as well as any assessments that have been administered in prior testing windows (i.e., Fall, Winter or Spring). It is also recommended that you read the data sgp documentation for further instructions on how to correctly use the various data sets.