The sgpData package is designed to provide a model for the format of data used in the SGP (Student Growth Project) function. This package is based on the WIDE (Wide Format) format and is a panel data set containing five years of annual assessment data for students. As such, the format of the data is important in that it is used by the SGP function. The data is used to generate a variety of analyses and student growth plots.
The first column in the sgpData table contains a unique student identifier (ID). This ID is required to run the SGP analyses. There are a number of other variables that are necessary to conduct the SGP analyses, including a first name, last name, and grade. All these variables are demographic variables used to generate student aggregates by the summarizeSGP function.
The next set of columns is where the numeric scores associated with the assessment occurrences are stored. Typically, these scores are for the scale score. A numeric score is a measure of a student’s performance on a given measure. If a student does not have a score, the missing value (NA) is displayed. When the sgpData table is loaded, the user must also provide the grade level and/or time associated with the assessment occurrences.
The LONG format is another form of longitudinal data used in the SGP package. Unlike the WIDE format, which uses a single row to represent a student, the LONG format distributes the data over multiple rows, spreading out time-dependent data across many rows. It has several preparation benefits, as well as storage benefits. Most analyses can benefit from using the LONG format.
The sgpData table has five columns. Each column represents a variable associated with a student at a different time in a student’s life. The columns are grouped by year and are each composed of two parts: a numeric scale score and a missing value (NA) for data that is not available. In addition to the scale score, the sgpData table will display the first name, last name, and grade for each year. These columns must be accurate. Otherwise, the sgpData table will show a site error or an unintentional false positive.
The SGP package provides comprehensive documentation on the sgpData functions. It also includes a set of exemplar longitudinal data sets. They include the Master 2022 and the Master 2012 sgp data sets. Both of these are very valuable for those who are interested in running the various analyses and growth plots that are included in the SGP package.
Using sgpData for analyses is quite simple. To begin, the user must first load the appropriate data sets. For the master 2022 data set, the user must have a first name, last name, and ID. Additionally, the user must provide the year and grade of the student. Since the sgpData table is a panel data set, the user must be careful to identify a student’s occurrences for a specific year and a particular grade.