SGP Data Pipeline Discussion
NHSAS and ACCESS
Damian Betebenner
Center for Assessment
August 21st, 2025
SGP Analysis Data Workflow
• Receive annual text (currently pipe “|”) delimited file from New Hampshire Department of Education
(both SAS and ACCESS)
• Lauren sends highly consistent layouts every year so she knows the extraction process from the
NHDOE side.
• Clean/Prepare data followed by SGP analysis
• Cleaning/preparing involves making at most minor modifications to labels (e.g., ethnicity labels),
checking for duplicate records, etc. and saving the clean results in an .Rdata format.
• With clean data in .Rdata format run SGP analysis using R software environment and the SGP
package. Analysis calculates SGPs (cohort and baseline referenced), straight and lagged projections
(cohort and baseline referenced), and exports of files that combine initial data with SGP data.
• All source code for data prep and analysis (no data) are located on the Center’s GitHub repo for
New Hampshire
• Export results as pipe delimited files and upload to the Center’s secure share file site for transfer back to
New Hampshire DOE
About this copy
SGP Data Pipeline Discussion — New Hampshire Department of Education, August 21, 2025.
This is an image-based copy of the original Keynote slides; builds and animations are flattened.