Data SGP is a collection of aggregated student growth and achievement data that can be used by educators and parents to better understand how a child is progressing over time. It provides individual student information including test scores and growth percentiles, along with aggregated data at the school and district levels such as class size attendance rates and graduation rates. This data can help shape classroom practices and evaluate school/district performance while supporting broader research efforts.
Student growth percentiles indicate a students current performance relative to their academic peers. They are calculated by comparing a student’s assessment score to the assessments of all other students in their grade level. Therefore, a student can have a high growth percentile even if their assessment score is low compared to other students in their grade. Conversely, a student can have a low growth percentile even if their assessment score exceeds the mean assessment score for the grade.
The sgpData tables provide valuable information for educators and researchers but do not contain all the contextual information needed to understand how a child is progressing over the course of their education. To fully understand how a student is performing, it is important to know which percentage of their grade level peers are at or above the same proficiency level as them. This information can be found by examining the sgpProgress table.
SGP analyses can be conducted in a number of ways with several variations. The SGP package contains wrapper functions, abcSGP and updateSGP, which simplify the source code associated with conducting these analyses.
An exemplar data set sgptData_LONG is available to assist with setting up a data frame for SGP analysis. This data set includes anonymized longitudinal (time dependent) student assessment data in long format for 3 content areas. The sgptData_LONG data set also contains a student demographic variables lookup table sgpData_INSTRUCTOR_NUMBER that can be used to associate teacher names with students by using their ID. This is required if running student growth projections with the summarizeSGP function.
To utilize the sgpData_LONG data set for SGP analysis the following 6 steps must be taken:
The first step is to create a long file that has a unique student identifier in the first column and then add assessment scores for each year of interest. Next a scale score must be calculated for each assessment by dividing the total assessment score by the sum of all the scale scores across each of the windows. Finally, student growth percentiles and progress reports can be generated for each individual student.
There are a variety of bubblePlot styles that can be used to display SGP information. The choice of style should be based on the desired outcome. For example, some bubblePlot styles accept specification for schools which can be useful if you want to compare student performance across schools. Other styles, such as the trendline bubblePlot, can produce a linear progression of student performance over time. This progression can be analyzed in more detail by using the progress report data that is generated for each student.