Data sgp leverages longitudinal student assessment data to produce statistical growth plots (SGPs). These are graphs that show students’ relative progress compared to their academic peers. SGPs can be used to gauge whether or not a student is on track to reach an established achievement target, such as being 75% of their academic peers.
However, creating SGPs is a complicated process that requires significant computation and has large estimation errors. These errors can make it difficult to interpret the results of an SGP analysis. In addition, SGPs are only meaningful for those students who have a sufficiently long history of test scores to provide meaningful information about their performance. Consequently, the SGPs provided by DESE are only useful for students in grades 4 through 8 and 10 and only compare student performance with their prior grade-level tests.
SGP analyses convert raw student test scores into scaled scores for comparison with an average of scaled scores from the previous year and subject area. This average is calculated from the scaled scores of all students who have tested in that particular grade and subject. The SGP program then determines if each student’s scaled score is above, below or at the same level as this average.
The SGP program also generates projections of future student test scores based on a variety of potential growth trajectories. These projections are shown for each student in the Star Growth Report. These projections are based upon the selected time frame in the report customization dialog box. The projections displayed for the current school year are based on student test score data from the Badger and 2015-16 school years.
Each year, the SGP data set is updated to include all available student test score data from past years. In addition, the SGP data set is extended with an additional year in order to provide the most accurate and complete SGP analysis possible. The resulting dataset is called the SGPdata and contains 4 examplar data sets for use with SGP analyses. The first, sgpData, specifies the WIDE format for use with lower level SGP functions such as studentGrowthPercentiles and studentGrowthProjections. The remaining two, sgptData_LONG and sgpdata_INSTRUCTOR_NUMBER, specify the LONG format for use with higher level SGP functions such as prepareSGP and analyzeSGP.
Most errors associated with SGP analyses can be traced back to issues related to data preparation. As such, it is important to understand the SGPdata data structure and how to correctly prepare it for use with SGP analyses. The SGPdata package contains extensive documentation on how to properly prepare SGP data for use with its higher level functions. Using the SGPdata, it should be relatively straightforward to run and interpret SGP analyses. This vignette provides a basic outline of the steps needed to do this. For more detailed examples, refer to the SGPdata documentation or request a specific SGPdata vignette from our team. We’re always happy to assist with SGPdata related requests.
