How to Program the Student Growth Percentile (SGP) Data Set

The Student Growth Percentile (SGP) measures a students progress by comparing them to their academic peers nationwide. This information can be communicated in percentile terms that are familiar to most teachers and parents, helping them better understand a students progress and whether they are above or below their academic peers. SGP is calculated using a combination of Betebenner’s well-known catch-up and keep-up growth projections along with a number of other factors that help determine a student’s relative growth.

Educators can access SGP information for their students by looking at the “SGP Data” tab within a student’s assessment report. They can also view a more detailed spreadsheet of SGP information by selecting a student and choosing the “SGP Data” spreadsheet from the menu. This spreadsheet displays the SGP results for a student over a five year period.

While the SGP data set is available to educators through the SGP reports and sgpData spreadsheet, it requires some additional programming to make it useful. To begin with, it is necessary to create a database structure that can store the SGP results and projections for each student. Ideally, this should be created using the same database format that stores all the other student assessment data sets. This ensures that the SGP analyses and results are consistent with the rest of the school district’s data.

Once the data structure is in place, a series of python functions can be executed to perform the SGP analyses. This includes the lower level functions studentGrowthPercentiles and studentGrowthProjections as well as the higher level wrapper function analyzeSGP. The analysisSGP function “wraps” the six steps listed above into a single function call, making it easier to execute the SGP analyses on large datasets.

A final step is to merge the SGP analyses into the master longitudinal record, Demonstration_SGP@Data. This will require the use of the sgpData_INSTRUCTOR_NUMBER lookup table that is used to associate instructors with students test records. This is important because instructors may change over time, meaning that a student could have multiple teachers in a content area for the five years required for baseline-referenced SGPs.

As a result of these processes, educators should expect to see minor differences in the SGP reports and spreadsheets for each student. These differences can be caused by a variety of factors, including differences in the methodology for SGP calculations and slight variations in the design of the baseline cohort. However, these differences should not be significant enough to significantly affect educator interpretation of the SGP information. As such, educators should continue to use the SGP results to inform their classroom practice and provide students with the best opportunities for learning. As always, if you have any questions or concerns about the SGP data or the process of using it, please contact us. We are happy to assist! Thanks for reading! – SGP Team.