The goal of data sgp is to leverage longitudinal student assessment data to create statistical growth plots (SGP) for individual students. These SGPs provide a way for teachers and administrators to compare and report student growth as measured by academically similar students. In addition, a student’s SGP can serve as an indicator of whether they are meeting an established achievement target, or SLO.
To achieve this, SGP analyses combine multiple years of student test score data and student characteristics to calculate students’ relative progress compared with academically similar students. SGPs are measured on a 1 to 99 scale, and the lower the number, the less a student has grown compared to academically-similar students. The higher the number, the more a student has grown.
Unlike standard student assessment scores, SGPs are based on the results of multiple years of testing and take into account student characteristics that are predictive of student performance on assessments, such as prior achievement levels, gender, racial/ethnic identity, and disability status. This is why they are more useful than student averages in identifying areas where students may need more support to accelerate their progress.
SGPs can be calculated for up to five years of student test score history, including the Badger Exam year. However, because statewide Badger Exam performance was so different from the WKCE, which preceded it, and the Forward Exam, which followed it, this year’s SGPs were not able to be compared with the previous two years.
Therefore, if you are looking for an SGP indicating that a student met or exceeded the Badger Exam SLO, we encourage you to check out the 2015-16 SGPs in our online dashboard. Please note that these SGPs do not include a column for the 2014-15 Badger Exam, and we will be recalculating them with the next round of data.
sgptData_LONG contains student assessment scores in LONG format for each of the 8 windows (3 windows annually) for which SGP analyses were run. In addition to a row for each student containing the identifier, grade level and assessment date, the file includes rows for each of the five previous years of test score data for that student. The data also contains additional columns for the VALID_CASE, CONTENT_AREA, YEAR and ID variables required for SGP analyses.
These files can be used to run individual and aggregate SGP analyses using the prepareSGP function. The prepareSGP function can also produce a variety of other variables often used in SGP analyses, such as HIGH_NEED_STATUS, which identifies the top and bottom quartile of students by year, content area, and grade grouping. The prepareSGP function also has a boolean argument that determines how the function deals with duplicate records based upon a key of VALID_CASE, CONTENT_AREA, and YEAR. The default value is TRUE, which warns of duplicate records but doesn’t attempt to modify the data.
Using sgptData_LONG and the prepareSGP function to perform SGP analyses is relatively straightforward. More detailed guidance on the process can be found in our SGP analysis vignette, and we will continue to update it as more information becomes available.