Data SGP

Data sgp is an analysis tool built for the statistical software environment R; to use it requires a computer running Windows, OSX or Linux and access to the free open source version of R. It also requires familiarity with the process of performing statistical data analysis; several online resources are available to assist users new to this approach.

SGP leverages longitudinal student assessment data to produce statistical growth plots (SGP). A SGP indicates a students relative progress, measured as the distance from their academic peers on a common scale; higher numbers indicate more than, less than or about as much progress as expected. Educators and parents use SGPs to gauge whether students are making reasonable progress toward achievement targets/goals and to inform instructional decisions. Administrators use them to support the development of Student Learning Objectives (SLOs) and guide school/district improvement efforts.

Unlike other commonly used measures of student progress such as standard score averages, SGPs take into account the varying sizes and levels of student populations and therefore provide more precise estimates of a students true level of performance. SGPs are also capable of adjusting for the effects of covariates such as gender, socioeconomic status and grade level. In addition, the SGP methodology allows for the creation of growth standards based on official state achievement targets/goals; a unique feature that facilitates communicating to stakeholders that proficiency must be reached within a given time frame.

The SGP Package provides both basic and advanced growth projection analyses that can be run simultaneously using various combinations of the studentGrowthPercentiles and studentGrowthProjections functions. These higher level functions combine multiple lower level function calls, simplifying the source code associated with operational analyses. In addition, the SGP Package features the sgpData_INSTRUCTOR_NUMBER lookup table containing instructor details associated with each student test record; these are used by the studentGrowthProjections and sgpData_GroPercentiles functions to adjust student group means and proportions for the effect of individual instructors.

The SGP Package can be used with either LONG or WIDE format data for operational analyses. We recommend the LONG format for managing data; all of the higher level functions are designed to be used with this format. Managing LONG format data is substantially simpler than managing wide format data since all of the higher level functions share an identical data structure and depend on the embedded SGPstateData meta-data for their state specific calculations. Using the sgpData_LONG data set for your operational analyses will provide the most flexibility and simplest data management procedure. To use the sgpData_LONG dataset for your SGP analyses, you will need to provide the following variables: VALID_CASE, CONTENT_AREA, YEAR, ID, YEAR, SCALE_SCORE, GRADE and ACHIEVEMENT_LEVEL. The sgpData_LONG data sets are available on CRAN and can be downloaded using the sgpData.R download link or by visiting the sgpdata github repository.