Data SGP is an information database about student development that can be used to help teachers pinpoint issues with learning. However, many educators find that performing analyses on this extensive dataset is both time consuming and complex. The good news is that this need not be the case. With the right understanding of how the database works and the proper tools, these analyses can be done efficiently.
The sgpData package contains classes and functions that provide a powerful, user-friendly framework for calculating student growth percentiles and projections/trajectories from large scale longitudinal education assessment data. The package uses a method known as quantitative regression to estimate the conditional density associated with each student’s achievement history, then applies this matrix to calculate growth projections/trajectories showing what percentile of their peers a student needs to reach to achieve future performance goals.
There are two primary types of longitudinal data that can be used with the SGP package: WIDE and LONG format. WIDE data is arranged in rows, each row representing a unique student, and columns represent variables at different times for the same student. The sgpData package includes exemplar WIDE format data sets (sgpData) and long format data sets (sgpData_LONG) to assist in getting started with operational analyses. The sgpData_LONG data set, in particular, is a great starting point because all higher level SGP functions are designed to work with this data format.
sgpData_LONG contains all of the student assessment data for each student over their entire school career. The first column, ID, provides the unique student identifier. The next five columns (Grade 2013, Grade 2014, Grade 2015, Grade 2016, and Graduation Year) contain the student’s scale score for each of these content areas, respectively. If a student has no assessments in any given year, the missing value is indicated by a ‘NA’ value.
SGP scores are based on relative student growth, so the first and last columns indicate the student’s SGP score relative to the 75th percentile of their academic peers. The middle column shows the current SGP score for that student, which indicates how much that student has grew or deteriorated since their last assessment.
The sgpData_INSTRUCTOR_NUMBER column in sgpData_LONG is an anonymized teacher lookup table that links insturctors with each students test record. For each student, there is an entry in the column indicating which teachers taught that student in each content area during the given year. In some cases, a single teacher may have had more than one student in their classroom for a given year, so multiple teachers will be listed for each of these tests. This is why the number of entries in the column is not equal to the number of assessments for that student in a given year. However, these entries are needed for each of the student’s assessments in order to generate SGP projections/trajectories. This is why these rows are nested within the student growth percentiles tables, so that the projections can be calculated correctly.