The data sgp package provides an extensive library of functions for conducting student growth percentile analyses (SGP) from assessment data. These analysis tools allow school administrators to examine student achievement trends over time and identify underlying issues that may be impeding student performance.
The first step in using the data sgp package is preparing and organizing assessment data for SGP analyses. This process includes identifying appropriate prior assessment scores, creating student aggregates, and calculating SGPs. The second step is analyzing the results of the SGP analyses and using them to make informed decisions about educational practice.
SGPs are a measure of the current achievement of a student relative to students who have similar prior achievement (Betebenner, 2009). A student’s SGP is based on their current test score divided by their previous test score. This is a simple and intuitive measure that remains well-defined even if test scores are not vertically or intervally scaled. SGPs are also easily comparable across different test administrations and grade levels, making them a useful tool for assessing trends over time and identifying underlying issues in student learning.
While SGPs provide a useful snapshot of student performance, they can be misleading if used out of context. It is important to consider the context of the SGPs being reported as well as the limitations and assumptions associated with their use.
A number of factors can influence a teacher’s SGPs, including student background characteristics, the level of skill in the subject area, and other classroom factors. These factors can lead to significant variation in SGPs among teachers. In addition, SGPs can be correlated with student background variables; however, this correlation does not necessarily imply that these variables are driving the variability in true SGPs.
The sgptData_LONG data set contains an anonymized, panel data set of student assessment records in LONG format for 3 content areas (Early Literacy, Mathematics, and Reading). This data set can be used to conduct student growth percentageile analyses with the SGP package.
The sgptData_LONG file is structured with the following columns: ID, VALID_CASE, CONTENT_AREA, YEAR, SCALE_SCORE, and GRADE. The ID column identifies each unique student record in the data set, and the next 5 columns are the student assessment scores for the years 2013, 2014, 2015, 2016, and 2017. The scale scores and grades for the assessments are provided as the inputs to the studentGrowthPercentiles and studentGrowthProjections functions in the SGP package. For more details, please see the sgptData_LONG vignette for detailed documentation on how to use the data.