Data sgp is an open source software package that allows a user to run Student Growth Percentile (SGP) analyses. It is a statistical software tool that uses statewide data to determine the progress students are making in their education. It is available on CRAN, the main repository of free, publicly-available R packages, and can be installed on Windows, Linux or OSX. Data sgp is a tool for educators and administrators to help them understand and interpret the results of their students’ statewide assessment scores.
The SGP calculation compares a student’s score on a given assessment to the scores of academic peers who scored similarly on previous assessments. These “academic peers” are grouped by the grade level and assessment subject. In Renaissance Next for Leaders, the first fall test in a school year is used as the starting point to calculate a student’s SGP. This SGP is then updated throughout the year as additional test scores are reported to Renaissance.
A student’s SGP reflects how much their current assessment score is above or below the median of their academic peers in the same subject and grade. In other words, it reflects how well or poorly a student is growing on the path to proficiency.
SGPs are calculated for students enrolled in specific English language arts and mathematics classes who have been tested in those subjects in prior years. Teachers who meet certain criteria receive mSGP scores for their students in the fall of each year. The mSGP score is determined by comparing the student’s most recent assessment to the average of their two highest mSGP scores from the previous three years.
The sgpData_LONG data set contains the assessment score data needed for SGP analyses. It includes 7 required variables: VALID_CASE, CONTENT_AREA, YEAR, ID, SCALE_SCORE, and GRADE. These variables are used to create the sgpData object which is then used by the higher-level SGP analysis functions to generate SGP calculations.
There are several vignettes in the data sgp documentation that describe how to use sgpData for particular analyses. It is recommended that users consult these vignettes before using sgpData in their operations.
It is important to note that sgpData differs from full community databases like Genbank or EarthChem. These databases aggregate data from multiple research consortia and make it accessible to essentially all researchers. In contrast, the approach and goals of sgpData are to focus on answering specific research questions in Earth history that are of direct interest to individual researchers. As such, the end result of sgpData will be to assemble or generate multi-proxy sedimentary geochemical data across all of the Paleozoic and Neoproterozoic. This will eventually be migrated into fully integrated and accessible community databases.
