What is Data SGP?
Data sgp is the tool used by teachers, school districts and states to analyze longitudinal student assessment data. This type of data contains important information on how students are performing in their classes and can be analyzed to identify trends over time as well as by gender, race and socioeconomic status – a very useful tool for improving student performance.
Using this information, educators can make informed decisions about which instructional practices to implement and which students may need extra support in the classroom. This can help narrow the achievement gap between high and low performers and ensure that all students are receiving a quality education and that resources are being spent wisely.
One of the most important things to remember when working with this data is that aggregating growth scores in a school or group of schools using mean rather than median results yields more stable results. This is a result of the fact that standard deviations for mean SGPs are lower than those for school-level medians when using the same set of students. The Department recently compared mean school-level SGPs for 4th grade math to medians by school and found that the school mean SGPs had much less volatility than the school-level medians.
SGP analyses are generally straightforward to run with proper data preparation. In fact, any errors that arise during SGP analysis almost always revert back to data preparation issues. Therefore, it is recommended that users read the SGP Data Preparation Vignette before conducting any analysis.
The SGPdata package installed with R includes exemplar WIDE and LONG formatted data sets that can be used for SGP analyses. The sgpData_WIDE file provides anonymized panel data where each row represents a unique student and its columns represent different variables associated with that student at different times. The sgpData_LONG file is similar except that the variables are spread across multiple rows of the file.
To use the sgpData data sets, users will need to have access to a computer running the free program, R. This is a powerful statistical computing environment that can be downloaded and installed from the R Project website. SGPdata also includes a tutorial on how to get started with this software.
While there are many ways to perform SGP analyses, the two most common involve either importing the raw assessment data or creating an object from the raw assessment data that can then be passed to the appropriate function for calculations and visual presentation. The SGPdata tutorial provides detailed instructions on how to import and work with raw data. The lower level functions in the sgpData package (studentGrowthPercentiles, studentGrowthProjections) require WIDE formatted data and the higher level functions are wrappers for these functions. In general, most analyses will require the WIDE data set while more complicated analyses are best run on the LONG formatted data.