Data sgp is an analysis tool for longitudinal student assessment data that creates statistical growth plots (SGP) which provide visual evidence of students’ academic progress relative to their peers. It does this by combining students’ standardized test scores with covariate information using an “growth standard”, which is established through the previous testing history of each individual student. As a result, SGPs offer more accurate measurements of student performance than traditional percentiles do.
Most errors encountered in conducting SGP analyses are associated with issues with data preparation rather than with the actual calculations. Therefore, the bulk of our time is spent on assisting users with proper data preparation and then to help them get up and running with their first SGP analyses. Once the data is prepared correctly, we believe that SGP analyses are relatively straightforward to conduct.
The SGPdata package provides both WIDE and LONG formatted data sets to assist with this process. However, we strongly recommend using the LONG formatted data set as most of the higher level functions (wrappers for the lower level SGP functions like studentGrowthPercentiles and studentGrowthProjections) require it. Furthermore, the management of long formatted data tends to be more straight forward than managing data in the Wide format.
When conducting SGP analyses, the first step is to prepare student assessment data in either the WIDE or LONG format. Almost all errors that are encountered with SGP analyses are associated with issues in data preparation so it is important that this be done properly.
For a typical student, their current SGP will be calculated from two tests taken during different testing windows (Autumn, Winter and Spring; the dates for these windows do not need to correspond to the school year). For students who are being held back one year, this means that they must take at least two tests in two separate testing windows in the prior school year before their SGP can be determined from their currently available data.
To ensure that the data is matched to the correct student, each SGP analysis should begin by creating a list of all students by their ID number in the Student View and then selecting the “Match Students” button next to each of their names. This will automatically select the matching student in the SGPData database for use with that particular SGP analysis.
Once the SGPData is populated with matching student data, each SGP analysis can be run by selecting the “Calculate” button next to the appropriate analysis. This will display the results of that SGP analysis in a table. The final column in the table indicates whether a student’s SGP is above or below their expected growth rate. If the SGP is above their expected growth rate, a positive number is displayed. If the SGP is below their expected growth rate, a negative number is displayed. If a negative number is displayed, the user should contact their local SGP administrator to determine what steps they may need to take to address the issue.
