National Science Foundation

NISS and Duke University are collaborating on the Triangle Census Research Network (TCRN), one of eight research nodes that are working on the National Census Research Network (NCRN). NISS and Duke are looking at ways to improve how federal statistical (“FedStats”) agencies disseminate data to the public and to researchers.

Official statistics surveys face a constant tradeoff between data quality and cost, which is dictated by current and anticipated budget pressures.   The World’s Simplest Survey Microsimulator (WSSM) was developed to work as a laboratory for survey scientists so they can try various scenarios out in order to make decisions.NISS sponsored an interdisciplinary Workshop on Microsimulation Models for Surveys. Simulation to explore three issues: need, utility and feasibility—surrounding simulation models for Federal surveys.

Two postdoctoral fellows will work on research aspects of longitudinal analyses using the Survey of Doctoral Recipients and small domain estimation for several SRS surveys.According to the NSF website, "The Survey of Doctorate Recipients is a longitudinal biennial survey conducted since 1973 that provides demographic and career history information about individuals with a research doctoral degree in a science, engineering, or health (SEH) field from a U.S. academic institution.

The goal of this project is to reconcile two methods for modeling change in social networks over time: a class of statistical models and intelligent agent models. The research contrasts the properties of these two approaches, exploring what qualitative dynamic behaviors in social networks are captured usefully and interpretably by each. The primary tools are latent variable representations and dynamical systems analysis.

Software engineering researchers use machine learning techniques to model and predict program execution behaviors. NISS helped software engineers by designing, implementing and evaluating light weight instrumentation for software testing and profiling to best optimize their performance.

NISS conducted research in data confidentiality, data quality, and data integration. Prototypes were built which could scale to operate on large sets of Federally held data. Researchers partnered with several large Federal Government statistical agencies. This topic was of particular importance given the balance of these agencies must strive for, in terms of their dual missions to collect and keep private confidential data, while at the same time making that data accessible for research and policy issues.

A critical problem for users of the World Wide Web is that many sites are difficult to navigate, hard to use and have confusing structure. Users may become lost, and they may make large leaps within a website (for example,

Though inherently statistical, model evaluation lacks a unifying statistical framework. NISS was hired to help find a overlying system to help with model evaluation. The research team used Bayesian techniques to measure the degree to which a model captures the underlying reality; theory and methods that allowed dual use of data in both estimation of model inputs and evaluation of outputs. SFCME also involved selection of evaluation functions by which a model and reality are compared. It also looked at design for determining what field or computer simulation data to collect.