Evaluation and Assessment Capability

Evaluation and Assessment Capability (EAC) supports NSF decision makers by studying programs and activities and their impact on the people who participate in and benefit from NSF investments. EAC also provides centralized resources and guidance for data collection, analysis, evaluation design, surveys, and enterprise analytics. EAC engages stakeholders in identifying learning priorities which drive a broad portfolio of projects that use qualitative and quantitative methods, from simple descriptive analysis to advances in modeling, network analysis, and machine learning.

           Featured Evaluations and Studies

Convergence Accelerator Summary Report: Descriptive Characteristics from 2019 to 2021

April 2023

Student researchers, wearing goggles and lab coats, work at a laboratory bench.

This report provides descriptive statistics on applicants, grantees, and projects for the 2019, 2020, and 2021 cohorts in the Convergence Accelerator program using NSF administrative data. Full Report (PDF, 1.35 MB)


Study of the Established Program to Stimulate Competitive Research (EPSCoR)

June 2022

This report presents an exploratory analysis of academic research excellence and competitiveness for EPSCoR using a variety of publicly available data sources.  Abstract (PDF, 106.41 KB) | Full Report (PDF, 20.52 MB)  

NSF Rotator Study: Analysis of the Impact of IPAs on NSF Strategic Goals

June 2022 

A researcher looks at a semiconductor

This report provides a descriptive portrait of Intergovernmental Personnel Act assignees and an assessment of IPA costs compared with federal employee costs.  Full Report (PDF, 867.41 KB)


Anti-Harassment Policy, Guidelines, and Communications

June 2022

Building and sky view

This report investigates and responds to questions in NSF's learning agenda related to recent anti-harassment policies to promote safe, harassment-free environments for the practice of science.  Executive Summary (PDF, 170.61 KB) | Full Report (PDF, 1.18 MB) | Technical Appendix (PDF, 817.54 KB)