Data SGP – Calculating Student Growth Percentiles and Percentile Growth Projections/Trajectories

The data sgp package contains classes, functions and data that are used to calculate student growth percentiles and percentile growth projections/trajectories using large scale, longitudinal education assessment data. This is accomplished by performing quantile regression to estimate the conditional density associated with each student’s achievement history and then utilizing a set of derived coefficient matrices to determine the student’s future performance targets.

A student’s growth percentile indicates how well they are progressing compared to their academic peers. The calculations for determining this are quite complex, but the concept is fairly intuitive: a student’s current test score is compared to the average test score of students with similar prior test scores. This allows for fair comparisons of students across schools and districts and is an important measure of student achievement.

However, as the percentage of the population that a student is ranked relative to increases, so too will their potential for future success. The higher the student’s growth percentile, the more likely they are to achieve their educational goals and realize their full potential.

In order to perform SGP analyses, users will need to have longitudinal student assessment data in either WIDE or LONG format. Most errors that are encountered in SGP analyses can be traced back to issues related to data preparation and, for this reason, it is critical that accurate and complete data preparation be performed before running any analysis with the SGP package.

Ideally, the data set will have 5 years of testing results for each student. The first column of the data should contain the unique student identifier (ID). The next five columns, GRADE_2013, GRADE_2014, GRADE_2015, GRADE_2016 and GRADE_2017, will provide the scale score associated with each student for each of these years. Finally, the last row should contain the student’s current test score.

Using the data sgp package for this purpose is, in general, relatively straightforward. For more detailed information on how to prepare student assessment data for use with the SGP package, please consult the SGP data analysis vignette and the documentation of the sgpData and sgpData_LONG data sets provided with the SGP package. If you have any questions, do not hesitate to contact the SGP development team. We will be happy to help!

Categories: Gambling Blog