The data sgp package provides classes, functions and data for calculating student growth percentiles (SGPs) and percentile growth projections/trajectories from large scale, longitudinal education assessment data. These percentiles compare students’ current performance to their academic peers nationwide based on their MCAS score histories. Students’ academic peers are grouped by grade level, and they may or may not be in the same subgroup as the student (e.g., special education, race/ethnicity). The SGPs and projections are calculated using a technique called quantile regression.
The SGP analyses in this package require a computer running the free R software environment. R is available for Windows, OSX and Linux and can be downloaded from CRAN. If you are not familiar with R, there are several resources to help you get started.
In order to use the SGP package, you will need to create an exemplar LONG dataset sgpData. Then, you will need to prepare the sgpData for SGP analysis by creating the SGP object Demonstration_SGP with the appropriate input data. The preparation process is a little involved, but it is relatively straightforward. Detailed instructions can be found in the SGP data analysis vignette.
Once the preparation is complete, you can conduct your SGP analyses. You can either directly call the lower level functions studentGrowthPercentiles and studentGrowthProjections, or you can use higher-level wrapper functions, such as prepareSGP and updateSGP. The wrapper functions simplify the steps and provide many benefits for conducting operational SGP analyses.
How are SGPs interpreted?
SGPs are reported on a 1-99 scale. Higher numbers indicate that the student’s score relative to the academic peer group is higher. For example, a student’s SGP of 75 indicates that the student scored higher than 75% of their academic peers on the most recent MCAS test section. This information can be used by teachers and administrators to evaluate the progress of their students.
A student’s SGP does not necessarily mean that they will achieve a particular proficiency level on the next MCAS test section. However, it does suggest that the student is likely to continue their current performance level or possibly improve. This information can be useful in planning instruction for the future. SGPs are also useful for evaluating whether a particular school is making progress toward its goals. In addition, SGPs can be compared to previous years’ SGPs for the same student to identify patterns of growth or decline. However, differences in SGPs from year to year should be interpreted with caution. A change in a student’s SGP relative to an academic peer group can result from a change in the mix of students in the group or a difference in the way the data are analyzed.