The Importance of Student Growth Percentiles
When it comes to assessing student learning, the most important measurement is not what students know, but how much they have grown. This is why student growth percentiles (SGP) are so important in interpreting test results, and why it is so important for schools to have access to high quality data that allows them to compare their students’ SGPs against those of their academic peers.
SGPs tell us how much a student has progressed from one year to the next on a state assessment, in comparison to other students taking that same assessment. They are a more reliable measure of student achievement than averages or percentages, as they account for the fact that different assessments use varying scales and may require students to start from different points on the curve.
For example, let’s say that a sixth grader in Wisconsin scored a 300 on this year’s ELA test and a 370 last year. A simple comparison of these two scores would ignore this variation and indicate that the student made no progress from one year to the next, despite the higher score being achieved this year. However, a SGP calculation will reveal that the student moved from a 57th percentile to a 60th in ELA, which is more accurate reflection of a student’s performance.
A SGP calculation also accounts for the effect of a particular teacher on a student’s performance in the same way that a student’s class schedule and other school related factors might influence their success. In other words, a student with a good SGP in ELA is likely to have an excellent SGP in Math and vice versa.
To understand this, the SGP package has been designed with a two step process for conducting operational analyses. The first step involves data preparation, the second step consists of performing the SGP calculations. The bulk of the work in this first step focuses on preparing and managing SGP data, particularly long formatted data such as that contained in sgpData and sgpData_instructor.
The prepareSGP function takes sgpData or sgpData_instructor and produces an SGP data object that can be used in the sgp function’s various analysis functions. This data object is also used to create the sgp_report, an output file that provides a clear and concise view of the results of SGP calculations.
Running SGP analyses requires a computer equipped with the R software environment. This software is free and can be installed on most operating systems, including Windows, OSX and Linux. The SGP package has been developed using the open source R language and therefore requires a working knowledge of this programming language to be used effectively.