Statistical Growth Points (SGPs) leverage longitudinal student assessment data to produce a measure of relative progress for students compared with their academic peers. However, creating SGPs from standardized test score histories is challenging due to the large estimation errors that result from complex calculations and the reliance on covariates (i.e., students’ prior test scores and other factors that may affect their performance).
Consequently, the SGP reports released by DPI for the 2014-15 school year did not include the option to view student-level growth information for that year. In order to have the full range of SGP analyses available, it is necessary for educators to have access to data that includes the 2014-15 year.
In response to this need, the sgp package contains wide and long format data sets that contain longitudinal student assessment data in the form of SGPs. The lower level functions that do the SGP calculations, studentGrowthPercentiles and studentGrowthProjections, require WIDE formatted data whereas the higher level functions (wrappers for the lower level functions) require LONG formatted data. For the most part, the lower level functions are designed to be used with data that has been properly prepared so it is not unusual for errors to occur during the analysis phase that revert back to problems with the data preparation process.
Both the wide and long format data sets include an anonymized lookup table, sgpData_INSTRUCTOR_NUMBER, that associates a teacher to each student’s test record. This data is helpful in analyzing SGPs by teacher since a single teacher can be assigned to multiple students within a content area for a given year.
The sgp package also provides a number of pre-made tables to display the SGP data in a variety of ways. These tables can be downloaded from the sgp website at https://www.mde.state.md/data/sgp/index.html. Currently, the following tables are available for download:
The tables provide a summary of the SGP data by grade level, content area and teacher for each of the four years included in the SGP report. For each of these tables, the data is organized by SGP level, showing low, typical, and high levels. This data is helpful in identifying the level of growth that each student is exhibiting which may have implications for their overall achievement. For example, a student who is experiencing low growth may be struggling to maintain their current level of achievement while a student who is demonstrating typical or high growth will be making academic improvements.