The data sgp package offers an efficient method for organizing longitudinal (time dependent) student assessment data into statistical growth plots. This package uses R software environment which is available for Windows, OSX and Linux operating systems. While its use requires some familiarity with its programming language R programming environment there are numerous resources available to get you going quickly and efficiently.
The median SGP serves as a useful summary statistic to illustrate student growth across subgroups, classes, schools and districts; however it only takes into account the middle value of all student results; mean SGPs provide more accurate representations of performance; they therefore are the more popular statistic used in Star Growth Reports.
SGPs may not provide a perfect indicator of student performance, but they can still assist decision makers in their deliberations process. For example, SGPs can help identify accelerated programs with few students that struggle to keep pace with the rest of their class; this may be especially relevant when starting out from similar knowledge, skills and abilities.
An SGP is typically comprised of student test score data from multiple assessment years. The first column, id, provides unique student identifiers; subsequent columns GRADE_2013 through GRADE_2017 provide scale scores; while GROUP_SCORE displays aggregate group scores across schools or districts.
Integrating SGP analyses is straightforward when using either of the sgpData_WIDE or sgpData_LONG data sets. The former contains an exhaustive list of variables used for analysis while the latter distributes time dependent variables across multiple rows for each student and provides demographic/student categorization variables that can be used with SGP to generate student aggregates.
At present, the sgpData_LONG dataset includes assessment results for eight windows (3 per year) of content areas that use LONG format for assessments. The sgpData_LONG format includes variables for valid_case, content_area, year, ID number, SCALE_SCORE GRADE grade level projections as well as ACHIEVEMENT_LEVEL for student growth projections. Information on these formats can be found in the SGP Data Analysis Vignette. Adding variables is simple by creating an issue on GitHub. Issues should include details regarding variable names, data sets from which variables will be extracted and usage in an SGP environment. Once an issue has been generated, developers can review it and respond appropriately.