As the fall of 2022 approaches we are continuing to provide districts and schools with the tools necessary to understand student growth percentiles (SGP) and how they relate to students’ M-STEP growth trajectories. SGP is a way to measure relative student performance, showing teachers and administrators whether students are advancing at a pace expected for their academic peers.
SGPs are calculated by comparing the current performance of a student to a cohort of students with similar score histories, taking into account previous test scores, grades, and subjects tested. Using this data we can determine where a student stands in comparison to their peers and provide guidance to educators on the next steps for helping students achieve at their potential.
While SGP calculations are complex and require an understanding of regression analysis, their interpretation is quite straightforward. Students with higher SGPs (in relation to the mean) are showing greater relative growth than students with lower SGPs, regardless of their current achievement levels.
This is why it is so important to provide SGP data for all students, including those that did not achieve proficient on a M-STEP. The SGP for these students still provides valuable information that can be used to guide instructional decisions and provide support for all students.
A few key points to note about SGP data are that students can have more than one teacher associated with their test record. This is due to the fact that each teacher can be assigned to multiple content areas and each content area can have different assessments administered. This will result in multiple teachers being assigned to a student for each year of data.
Also, SGP results do not follow a normal distribution and the number of students at each percentile can vary from year to year depending on the size of the cohort of students being measured. As a rough guide, you can expect to see approximately 10 students in each decile or grouping of percentiles across the state.
SGP analyses use two types of data, a student’s historical growth trajectories and their projected growth trajectories. While the sgpPercentiles and sgpProjections functions can run on WIDE formatted data, all of the higher level functions, including window specific SGP and projected SGP, require LONG formatted data.
The sgptData_LONG data set contains LONG formatted assessment data for 8 windows (3 windows annually) of content area data and is anonymized. To use this data you will need to have a valid sgpKey and a valid sgpInstructorNum for each student’s assessment record. This data will be utilized to create the SGP data used by the sgpPercentiles, sgpProjections, and studentSGPgraph functions. VALID_CASE, VALID_CONTENT_AREA, WINDOWS, ID, SCALE_SCORE, GRADE and ACHIEVEMENT_LEVEL are required variables for creating these functions. In addition, firstName and lastName are required if running student SGPgraphs. These are created from the sgptData_LONG dataset and will display the individual student’s data for each year of assessment.