Data sgp provides a variety of data sets and functions for analyzing educational assessment data. Its most widely used function is student growth percentiles which provide a measure of how much a students score has changed over time relative to their academic peers. The purpose of these percentiles is to identify if students are growing at an acceptable rate. If they are not, the student may need additional support.
Student growth percentiles can be analyzed for individual students or entire groups of students. In addition to student growth percentiles, data sgp can produce a number of other useful measures such as student group averages and trend lines. These reports can be a valuable tool for teachers to track student progress.
These reports can be accessed in the Star Reports menu by selecting the student/school/district tab and then choosing the Student Growth Percentiles option. The reports can be customized by changing the prior year and current window used for the analysis. Alternatively, the reports can be downloaded as PDFs or exported to a spreadsheet.
In addition to providing student growth percentiles, data sgp also produces growth projections for future years. These projections are based on the historical trajectories of student performance from past test scores and indicate what amount of student growth is required for a student to reach or maintain proficiency. This information can be particularly useful to educators when setting student learning goals (SGOs).
SGP projections are only available for ELA and mathematics. For grades 4 through 8, the projections compare the students current grade level test score with their score from a previous grade level test. For grades 9 and 10, the projections compare the students grade 10 test score with their score from the previous grade level test.
To create the projections, data sgp uses the current student score from the Star Report and the prior student score from either an earlier or subsequent test in the same testing window. This information is then converted to a percentage of the student population and placed on a graph. The percentage is then scaled by the mean of all other students and divided by the standard deviation to create a percentile rank.
The percentile ranks are then ranked by their location in the distribution, with the top 50 percent being at the 99th percentile and the bottom 50 at the 1st percentile. The sum of all the rank values is then summed to calculate the overall score.
Using the data sgp package to perform these analyses is relatively straight forward. In general, lower level functions such as studentGrowthPercentiles and studentGrowthProjections require WIDE formatted data whereas higher level functions such as studentScorePlots and studentSGPanalyzes require LONG formatted data. In order to use these longer formatted data sets, the sgpData_LONG package was developed which provides a set of variables that can be easily used with most SGP analyses. The sgpData_LONG package also includes an embedded vignette which provides more comprehensive documentation on how to work with this data set.