Utilizing longitudinal student assessment data for SGP analyses requires careful and deliberate attention. Most errors associated with such analyses result from issues in data preparation, so it is crucial that it is prepared appropriately. While the SGP package supports both wide and long formatted data sets for operational analyses, for operational use we advise using only long formatted sgpData_LONG data sets due to their ease of management compared with wide formatted versions; all higher level functions developed specifically for use with SGP analyses assume there will be SGPstateData meta-data embedded within long formatted long formatted data sets for optimal use when managing SGPstateData meta-data will exist within them.
The ID column in sgpData_LONG provides a unique identifier for every student in this dataset. GRADE_2013, GRADE_2014, GRADE_2015 and GRADE_2016 display the grade level associated with their assessment score for each year since 2013. Finally, SCALE_SCORE provides scale scores associated with each year of assessment results.
These variables are used by the SGP function studentGrowthPercentiles to calculate an appropriate percentile of a given student’s growth rate based on how many years of data have been compiled, while studentGrowthProjections generates projections of expected achievement levels based on current performance levels and potential enhancement with additional effort invested into individual students.
For projections to be accurate and valid, it is crucial that one knows how many years of data a given student will have available for future testing. Unfortunately, schools or districts don’t usually provide this information and it may even be impossible to calculate from existing data alone. To address this issue, the SGP package offers an estimate for how long one student might remain at their school by using current test results as well as length of enrollment and student tenure data.
It can be especially useful for educators and administrators looking to forecast students’ growth rates and expected achievement levels well into the future. It can also provide important guidance when considering implementing SGP in classrooms or schools. Remember, however, that SGP should not be seen as an alternative to more traditional statistical methodologies and techniques. Before implementing SGP in their schools, educators should make sure they have an in-depth knowledge of its assumptions and limitations – this will prevent making false projections based on incomplete or inaccurate data. It will lead to more accurate and meaningful analyses of their data and more informed decisions about students’ learning, which in turn will enhance teaching and learning in our schools. Now is an exciting time for education; innovations are endless! Let’s work together toward making improvements!