A student growth percentile (SGP) is a measure of student achievement that compares a student’s current assessment score with their prior assessment scores, often reported in terms of a percentile rank. Using SGPs, educators can share information about student learning in terms that are familiar to most teachers and parents. SGPs can be calculated with a variety of assessment data, including the results from standardized tests and classroom-based assessments.
SGPs can be computed for individual students or for groups of students, such as entire schools, districts, or state. They can also be used to produce a variety of reports and graphs that highlight trends in student performance. These reports and graphs can be used for teacher evaluation systems, parent conferences, and other school-wide decision making. SGPs can be either cohort-referenced or baseline-referenced. Cohort-referenced SGPs are more reliable and valid than baseline-referenced SGPs. Baseline-referenced SGPs are more difficult to interpret, and they require more student and testing data than cohort-referenced SGPs.
In addition to defining the SGP calculation model, SGPdata contains 4 examplar data sets to help you get started with your own SGP analyses. One data set, sgpData, specifies data in the WIDE format used by lower level SGP functions such as studentGrowthPercentiles and studentGrowthProjections. Another data set, sgpData_LONG, specifies data in the LONG format used by higher level SGP functions such as abcSGP, prepareSGP, and analyzeSGP. Finally, the data set sgpData_INSTRUCTOR_NUMBER is a teacher-student lookup table that’s utilized to produce teacher level aggregates.
The SGPdata package can be downloaded from GitHub. Once you have installed the package, open the SGPdata folder in your Python environment and follow the instructions to load the sgpdata.py file in order to begin a new SGP analysis.
To get the best results from your SGP calculations, take some time to format your data set. For example, spend a few minutes to make sure that the statistic category columns display decimal points, and that all of the columns have the same column headers. You should also be sure to select the correct denominators for your variables (e.g., choose a number format that allows you to easily see the numbers in each row).
Then, start by loading your data set and running a few basic SGP calculations. Once you have a good understanding of the basics of SGP calculations, you can move on to more sophisticated models. The SGPdata package provides many examples of advanced SGP models, and we will continue to add additional example in the future. If you have any questions or suggestions, don’t hesitate to contact us. We’re always happy to help!