The Data Sgp provides educators with a new way to view student achievement information. Instead of tying their results to one assessment or grade, it shows growth over a student’s entire educational career. It also helps to identify students who are making great strides and those that need additional support. Educators can use this information to tailor their teaching practices to fit the needs of each student.
The underlying math used by the Data Sgp is complex, but the good news is that it’s a very intuitive system for those who understand how it works. As such, most errors that arise when using the Data Sgp often revert back to data preparation issues. The Data Sgp is easy to use, and a variety of online resources are available to help beginners get started.
To use the data sgp, you will need to have access to a computer running Windows, OSX, or Linux and an open source version of the R statistical software program. While designed specifically for longitudinal student assessment data, it can be used with any type of time dependent data. It can be run as a script from the command line or as an interactive user interface.
Before diving into the details of the Data Sgp, it’s important to understand a few key concepts. First, SGPs are based on a statistical model, which means that their accuracy depends on how well the data is prepared. This includes setting up the correct variables, converting data to the proper format, and selecting the appropriate scale for analysis.
Next, it’s important to understand the distribution of SGP scores across the state. SGPs are not expected to follow a bell-shaped curve; instead, they should cluster around the median, meaning that most students have similar scores. The diagram below demonstrates this clustering, and the circled points are the median scores for each grade level.
While it may seem tempting to make direct comparisons between SGP scores, this is a dangerous practice. Different assessments require different starting points, and differences in grade levels can lead to dramatically different degrees of achievement. For example, Simon in sixth grade scored a 370 on this year’s statewide assessment in English language arts (ELA). This year’s score is a 70 point increase over his fifth-grade performance. But if we were to directly compare his scale score gains with those of his academic peers, we would find that his gains are less impressive.
To address this issue, the SGPdata package includes an example WIDE format dataset that can be used to simulate the time dependent data required by lower level functions like studentGrowthPercentiles and studentGrowthProjections. Additionally, it includes a LONG format dataset to facilitate the conversion process. Both are provided on the GitHub page for this tool.
