Data sgp is a data set that contains the aggregated results of student growth percentile analyses for up to two years. These data sets provide valuable information about a student’s performance over time, including their ability to make progress towards proficiency and what is needed to achieve it. These data are used to generate the student growth projections displayed in Star’s Student Growth Reports.
The data sgp are based on statistical models that compare students with similar score histories and determine relative performance. For example, although Students A and B have identical MCAS scaled scores for this year’s assessment in a subject area, their SGPs may be different because they were placed into different academic peer groups when their previous score history was assessed.
A student’s SGP is calculated by comparing their current performance to the average SGP of their academic peers from past MCAS administrations in the same subject area. These academic peers are identified based on demographic characteristics (e.g., gender, income) and educational programs (e.g., sheltered English immersion, special education). A statistical process called quantile regression is then used to place the academic peers’ current performance on a normative scale–making it possible for the SGPs to be determined.
SGPs can be analyzed by student group or grade, as well as by individual student. These reports allow educators to explore trends in student growth within their schools, districts or regions. They also help to identify areas where student groups may need additional support. The SGPs for all students in a particular school can be found on the MCAS Results by Student Group page of School and District Profiles.
When using the sgptData_LONG data set for student growth percentiles and student growth projections, there are seven required variables: VALID_CASE, CONTENT_AREA, YEAR, ID, SCALE_SCORE, GRADE and ACHIEVEMENT_LEVEL. The last two variables are only required if running student growth projections (as opposed to individual student growth and achievement plots).
SGP analyses can be run on WIDE or LONG format data, but for operational use, the latter is recommended as it provides numerous preparation and storage benefits. The lower level functions that create the SGP plots and projections use the WIDE data format, but the higher level wrapper functions utilize the LONG data format. Therefore, you are likely to want to format the data in LONG format if you plan to run these analyses repeatedly over an extended period of time. The sgptData_LONG file is available for download on the Downloads page. A detailed description of the SGP data is also included in the readme.txt file in the archive. This readme describes the contents of the sgptData_LONG and WIDE data sets, as well as how to prepare and use them for various analyses. The sgptData_LONG files can be downloaded using the links below. The readme includes instructions for creating the data sets, sample graphs and a tutorial on using the SGP analysis tools. It is recommended that you read this readme before using the data sets.