In modern science, business and everyday life, the term ‘big data’ is frequently used to describe data that is too large for traditional analysis applications. However, compared to analyses of Facebook interactions and global climate changes, the data set for student growth percentiles (SGP) is relatively small.
Using the data sgp package, you can perform an operational SGP analysis with no need for sophisticated database software or complicated programming skills. The data sgp package is designed to be as user-friendly as possible. The majority of time spent conducting SGP analyses is spent on data preparation, which the package simplifies through its intuitive user interface.
The most important feature of the data sgp package is that it provides a single unified interface for performing all of your SGP analyses. This allows you to easily switch between analyses, and makes it much easier for educators and administrators to incorporate individual, classroom-level, and grade-level SGPs into their school improvement efforts.
SGPs are based upon comparisons of students with similar academic peer groups from previous MCAS assessments and determine relative performance. Students with the same prior score history have a similar SGP in the same subject. Similarly, students who have different prior scores and SGPs in the same subject indicate that they likely have academic peer groups with very different performance levels.
In addition to providing a clear and intuitive graphical interface, the data sgp package also provides a unified data set format for all of your SGP analyses. The sgpData data set contains the following columns: ID, prior_grade, current_grade, growth_period, subject, and SGP.
When importing the SGP data, users can choose whether to create a longitudinal data set by default based upon a stratified random sample of students or a state-specific cohort by setting the SGP_stateData argument. This is a critical decision for constructing confidence intervals. The use of a state-specific cohort enables the calculation of SGPs using actual student data, which is more accurate than assuming that the average value for all students in the state is equal to the value calculated for a specific sample.
The SGP report interface displays a student’s SGP in a table along with their overall MCAS score. Students can access their SGP by clicking on a column of the table. The report can also be sorted by various criteria.
When interpreting SGP results, it is important to remember that SGPs are calculated for each year of the assessment and may change from year to year. Differences in SGPs between years should be interpreted with caution, as they may reflect differences in the underlying methodology rather than differences in student performance. Ideally, SGPs should be viewed with the context of all other measures of student achievement and progress.
