Data SGP
Data sgp leverages longitudinal student assessment data to produce statistical growth plots (SGPs) for each student. These are a measurement of relative student progress against a growth standard established from prior test scores and student covariates. However, constructing SGPs from students’ standardized test score histories is a time-consuming process, and producing them requires complex calculations that are susceptible to large estimation errors.
The SGP Package provides tools for reducing these estimation errors by using least squares and Bayesian regression modeling to estimate latent achievement trait models, and by comparing these estimates against data collected from an identical baseline cohort of similarly-performing students. The result is a set of student aggregates that are comparable over time and across teachers, content areas, and windows of measurement. These aggregates are the foundation of operational SGP analyses and provide a reliable measure of student progression.
This software makes it easy for educators to compare SGP results between different time frames and report on the progress of individual students. SGP scores are reported on a 1-99 scale, with higher numbers indicating more relative growth. For example, a 75 SGP score indicates that the student has shown more growth than about 75% of their academic peers.
Macomb and Clare-Gladwin school districts have made their SGP data sets publicly available in formats compatible with these SGP analysis functions. The data sets include the sgptData_LONG dataset for SGP analyses, a set of LONG tables that contain student assessment records from 8 windows (3 windows annually) for 3 content areas, and a sgptData_INSTRUCTOR_NUMBER table for connecting students to instructors through unique IDs associated with their assessment records.
The SGP package provides the following 6 functions:
prepareSGP Takes exemplar SGPs from other schools and analyzes the data to identify common errors, such as over-estimating a student’s ability. This function is an essential step in ensuring that SGPs are valid and useable.
analyzeSGP Calculates student SGP percentiles and growth projections using the sgptData_LONG data set. This function also compares SGP percentiles and projections against teacher evaluation criteria.
updateSGP Adds new SGP estimates to the sgptData_LONG database, including current and lagged student growth percentiles and projections and teacher evaluation criteria.
combineSGP Merges the results derived from analyzeSGP into the master longitudinal record, Demonstration_SGP@Data. This function also merges the student aggregates calculated by summarizeSGP into a single table.
This function is an important component of the SGP Package because it allows users to easily visualize students’ progress over time. This tool is especially useful for identifying trends in student performance, evaluating teacher effectiveness, and providing a consistent basis for discussing and understanding students’ academic progress with parents. In addition, this tool is a valuable resource for educators as they plan and implement instruction for students. By connecting this information with other district-wide data, teachers can use it to make more informed decisions regarding their instructional practices. This can help them to identify strengths and weaknesses in their curriculum and provide the appropriate supports for struggling students.