In the context of education research, student growth profiles (SGPs) are a powerful statistical measure of student progress and achievement. They can provide more accurate information about student performance than traditional measures such as mean differences and average gains, and they are well suited to accountability systems that place value on student growth rather than test scores alone. However, constructing SGPs requires access to longitudinal student data in WIDE or LONG format. The sgp package provides exemplar WIDE and LONG format data sets (sgpData and sgpData_LONG) that allow researchers to conduct SGP analyses.
SGPs are also less susceptible to error and bias compared with conventional standardized test score comparisons. For these reasons, SGPs are becoming increasingly important in educational measurement. The sgp package includes tools for constructing, analyzing, and visualizing SGPs.
The goal of data sgp is to assemble a comprehensive dataset that is tractable to multivariate statistical analyses and accounts for multiple geological and methodological biases. The current phase of data collection is focused on the Neoproterozoic and Paleozoic, and future phases will extend coverage to more geologic time periods.
In addition to enabling the development of SGP models and analysis, the data sgp project will provide an archival copy of the resulting database for use in research and teaching. The data sgp project is structured as an open-source collaboration, and researchers who wish to use the database may request access through a standard peer review process.
Data sgp is a domain specific project that differs from other databases in the way it collects and stores its data and in the tailored way in which it presents it to the research community. The primary objective of data sgp is to build a well vetted, comprehensive, and scalable longitudinal data set that can be used for research and analysis of student achievement.
The sgpdata repository contains a comprehensive, longitudinal student assessment dataset in the WIDE format. Each row in the sgpdata contains a unique student ID and 5 columns representing the student’s grade level for the previous assessment year. The final column, SS_2013, SS_2014, SS_2015, and SS_2016, provides the student’s scale score in each of the previous assessments.
To conduct SGP analyses, you need access to a longitudinal dataset in the WIDE or LONG format that contains raw student assessments and their associated percentiles. The minimum dataset required is sgptData_LONG which provides 8 windows (3 annually) of assessment data in the LONG format for content areas in Early Literacy, Mathematics and Reading. In addition, you will need access to state level student aggregates provided in sgpstateData.