The classes, functions and data within the SGP package are used to calculate student growth percentiles and percentile growth projections/trajectories using large scale, longitudinal education assessment data. The SGP methodology utilizes quantitative regression to estimate conditional density and derived coefficient matrices to create student growth trajectories showing what growth is needed to reach or maintain proficiency.
Students are compared to academic peers with similar MCAS score histories when calculating SGPs. Academic peers can be within a student’s school, district or across the state. They can also be in specific subgroups such as race/ethnicity, special education or multilingual learning.
A student’s SGP is based on the performance of the student relative to academic peers in their grade who have similar MCAS score histories. This means that two students who have different MCAS score histories and do not belong to the same academic peer group can have the same SGP. For example, Student A and Student B scored the same on this year’s MCAS test in a subject area. However, because they had different MCAS scaled scores in previous administrations of the same test, they had different SGPs. This is because SGP is based on comparing students with similar score histories rather than being tied to the same academic peer group from previous MCAS tests.
SGPs are calculated for each individual student by analyzing their results in the context of the student’s prior performance. This is what differentiates SGPs from traditional standardized test scores. The goal of SGPs is to provide a more accurate measure of student achievement that takes into account the impact of prior performance on current achievement.
The average SGP, which is reported on the Star Growth Report, is a measure of the average student’s growth from their previous test to the current one. This growth is determined by averaging a student’s SGP from each of their testing windows. A student’s SGP from the Autumn window will be added to their SGP from the Winter and Spring windows. If a student was held back a year, their SGP will be based on their results from the previous school year’s testing.
SGPs can also be aggregated to analyze growth at the school, district and student subgroup levels. Because averaging involves a larger sample of available data, the average SGPs for schools, districts and subgroups can fluctuate more than individual SGPs. However, these aggregate measures are useful because they can inform about the typical level of growth for a particular type of school or group of students.
SGPs are calculated using longitudinal (time dependent) data and therefore require a large amount of data to provide an accurate picture of a student’s growth over time. The sgpData package, installed when you install the SGP package, contains exemplar WIDE format and LONG format data sets (sgpData and sgpData_LONG) to assist you with setting up your student assessment data for SGP analyses. Please see the SGP data analysis vignette for more comprehensive documentation on how to use these data sets.