Data SGP provides state educators with an opportunity to gain insights into students’ progress by revealing growth in relation to academically-similar peers. This unique feature distinguishes SGP from standard growth models and other methods in that it allows districts to link student/teacher performance against official state achievement goals/targets.
SGP analyses take the form of latent achievement trait models estimated from historical test score data and student covariates and compared against growth standards established by teacher evaluation criteria. This helps reduce estimation error and increase validity of comparisons.
In order to run SGP analyses, districts need access to longitudinal (time dependent) student assessment data. This type of data is typically stored in WIDE format with each case/row representing a single student and columns providing information associated with the student at different points in time. The SGP package offers wide (sgpData) and long format data sets (sgpData_LONG) for use with operational analyses; sgpData contains variables like VALID_CASE, CONTENT_AREA YEAR ID SCALE_SCORE and GRADE while sgpData_LONG includes additional columns such as SS_2013, SS_2014, SS_2015, SS_2016, and SS_2017 that provide information about the student’s assessment scores over the course of 5 years. sgpData_LONG also contains a lookup table (sgpData_INSTRUCTOR_NUMBER) that associates an instructor with every test record for a particular content area.
To run a SGP analysis, users will need to have the open-source R software installed on their computer. This is available for Windows, OSX and Linux and can be downloaded free-of-charge through the CRAN website. Running SGP analysis requires some familiarity with the R environment and a thorough review of its documentation is recommended. Additional guidance can be found in the SGP Data Analysis Vignette.
Assuming a district has the necessary data and software, the next step is to prepare the student data for analysis. The most important step is to make sure that all of the tests a student has taken are included in their record. This is critical for ensuring that the growth model can accurately capture the student’s progression and that the resulting SGP is meaningful.
Consider an example: Simon achieved a scale score of 370 on this year’s statewide assessment in sixth grade. SGPs place Simon’s score growth in context by comparing him with students that had a similar history of test results – i.e., with students whose fifth grade scale score was similar to his.
By using SGPs, Simon’s teachers can understand how well he is making progress and identify areas where further instruction might be needed. This will allow them to focus their efforts on the students that need it most while keeping the entire class on track to meet state learning expectations. To learn more about how SGPs work, check out this video.