Data SGP

Data SGP, or Student Growth Percentiles, are a way of quantifying relative student growth using large scale longitudinal education assessment data. Students’ growth percentiles can be used by teachers and administrators to better understand student progress, inform instruction, evaluate schools/districts, and support broader research initiatives.

SGP is calculated by comparing a students’ current score to their performance relative to a group of their academic peers. The higher the SGP number, the more a student has grown in comparison to their peers. SGP scores can also be used to communicate a school’s progress toward official achievement targets/goals to stakeholders.

The SGP package, which is available via CRAN, uses the R statistical programming language to conduct analyses. This means that to run SGP analyses users will need a computer with the R software installed on it. R is available for Windows, OSX, and Linux and is open source so it can be downloaded and used at no cost. While SGP is a relatively simple analysis method, it requires careful attention to the details of data preparation in order to produce meaningful results. As such, we recommend that users spend time reviewing the documentation and vignettes before running any SGP analyses. Any errors that arise during the course of an SGP analysis generally revert back to issues in data preparation.

In addition to providing the lower level functions (studentGrowthPercentiles and studentGrowthProjections) that calculate SGP, the SGP package contains higher level function wrappers such as abcSGP and updateSGP that simplify source code associated with operational analyses. The SGP package also includes the sgpData_INSTRUCTOR_NUMBER lookup table that provides instructor information associated with each student test record.

Finally, the SGP package provides a set of graphical tools for viewing/displaying SGP analyses and generating reports that can be distributed to stakeholders. We encourage users to use these tools to help communicate their results and explain the significance of their findings.

The SGP package is intended to be used with the sgptData_LONG data set. This is an anonymized data set consisting of 8 windows (3 windows annually) of assessment data in LONG format for 3 content areas. The sgptData_LONG includes 7 required variables: VALID_CASE, CONTENT_AREA, YEAR, ID, SCALE_SCORE, GRADE and ACHIEVEMENT_LEVEL. These variables are necessary to create aggregated student data sets required by the sgpsummarizeSGP and sgpprojections functions. The sgptData_LONG also includes the demographic variable FIRST_NAME and LAST_NAME that are required for creating student-level student growth/achievement plots by the sgpsummarizeSGP function. This data set also includes a teacher indicator variable, sgpData_INSTRUCTOR_NUMBER, that is only present in the LONG version of the data set.