Data SGP is an educational measurement system that employs longitudinal student data to produce statistical growth plots (SGPs). SGPs are percentile rank measures that compare current test scores of individual students with others who started at similar levels; their aim is to be more accurate than unadjusted growth scores while providing a more holistic picture of student performance by taking into account an academic journey rather than current year performance alone. Furthermore, SGPs serve as powerful tools for evaluating education programs, teachers, and schools.
SGPs can be an invaluable educational measure, yet misuse may mislead users into misinterpretations of them. To avoid confusion it’s essential that we understand how these SGPs are calculated from student data and what their limitations are.
This article serves to provide an introduction to SGPs and their uses in school accountability systems, while outlining their structures within an SGPdata database and detailing some higher level functions designed for use with it.
The SGPdata database offers a wealth of data regarding student growth. It includes students’ prior and current test scores as well as Student Growth Parameters (SGPs) in math and English language arts (ELA). Furthermore, this comprehensive resource also features detailed student background characteristics such as race/ethnicity/gender statuses; eligibility statuses (free lunch eligibility / reduced lunch eligibility); special education status and free and reduced lunch eligibility status reports submitted by states as accountability reports. The primary use for SGPdata database reports mean and median SGPs reported by states for accountability reports is this database which supplies mean and median SGP values reported by states in accountability reports submitted by states to states accountability reports.
Importantly, SGPs cannot be directly compared across different assessment administrations due to varying scales for these assessments. Therefore, SGPs must be adjusted in order to account for these differences, by multiplying each scale-weighted student achievement gains (e.g. e4,2,i) with its appropriate scale weightings for each test administration.
One final point about SGPs: it is important to be mindful that SGPs may be affected by the performance of individual students who either can’t keep up with or progress as quickly as the rest of the class, particularly those in accelerated programs. SGPs for these individuals will likely be lower than for other students in their grade as they will not make as much progress than their counterparts who started from similar knowledge, skills, and ability levels.
To address this problem, the SGPdata database offers an anonymous sgpData_INSTRUCTOR_NUMBER data set that contains instructor details associated with each test record of students taking exams. This allows for identification of teachers responsible for SGP calculations; then an exploration can be made between true SGPs and teacher influence in calculating individual SGPs – something particularly pertinent since individual student SGPs tend to correlate closely with instructors’ instructional practices.