Data sgp leverages longitudinal student assessment data to produce statistical growth plots which reveal insights into a students relative progress compared to their academic peers. These growth percentiles are derived from standardized student test score history and can provide useful information for informing instruction, assessing teacher/student performance, and supporting educator evaluation systems. However, creating these growth percentiles from standardized student assessment data is a complicated process that often yields large estimation errors which render them unusable for measurement purposes. Data sgp provides a computationally simple and statistically sound alternative that avoids these errors.

SGP is an evaluation metric that compares a student’s current grade-level test score to the score of students with similar prior test scores (their academic peers). It is reported on a scale of 1 through 99, with higher numbers indicating more relative growth than lower ones. This metric can be used to identify underperforming students and target instruction, as well as to determine whether accelerated programs are effective at improving students’ learning outcomes.

To use data sgp, educators need to register on the state’s website and then access a more detailed sgpData spreadsheet for each student in their class. This spreadsheet displays a student’s SGP data over the course of five years. The first column, ID, provides the student’s unique identifier. The next five columns, SS_2013, SS_2014, SS_2015, and SS_2016 provide the student’s assessment score for each of these years. If a student doesn’t have 5 years of data, the spreadsheet will show missing values (NA).

The final two columns, sgpData_INSTRUCTOR_NUMBER and sgpData_STUDENT_PERCENTILES, provide the insturctor number and student percentage associated with each student’s SGP percentile estimate. The INSTRUCTOR_NUMBER value is anonymized to ensure confidentiality and the STUDENT_PERCENTILES value is a rounded up version of the student’s percentile score.

Educators can run SGP analyses on sgpData files using the low level function studentGrowthPercentiles and the higher level wrapper functions, sgpProjections and sgpDivisors. The choice of formatting sgpData files in WIDE or LONG format is largely driven by the nature of the SGP analyses to be performed; in general, the lower level functions require WIDE format while the higher level wrapper functions, which utilize the same data set, prefer LONG format due to its numerous preparation and storage benefits over WIDE.

Using the SGP analysis tools requires familiarity with the R software environment. This is a free, open source statistical package that can be downloaded for Windows, OSX or Linux and has extensive online resources available to assist newcomers getting started with its use. Running SGP analyses also requires a computer with at least 4GB of memory and at least one core. For larger SGP data sets, it is recommended that you use a dedicated machine to run your analyses. If you are not familiar with R, we recommend reading the SGP vignette for further documentation on how to get started with the package.