Release Date: 8/10/2015
Raphael Nishimura, an expert in sampling, non-response bias, and missing data, has joined the Abt SRBI Advanced Methods Group, having just completed his Ph.D. at the University of Michigan Program in Survey Methodology.
His doctorate research at Michigan examined optimal ways to reduce survey non-response bias through multiple imputation, calibration and modeling.
His most recent published research, in the International Statistical Review, also focuses on reducing non-response bias. In this just published study, Raphael and his co-authors used a large-scale simulation to explore how alternative indicators perform as indicators of non-response bias. They examine alternative indicators because response rate is a poor indicator of non-response bias.
Raphael has also conducted research on responsive design tools, item non-response in weighting covariates, estimation of weighting and design effects, and Bayesian foundations in survey sampling.
“Raphael’s explorations into procedures to reduce non-response bias represent progress in a critical area of survey research – how survey respondent answers may differ from those who did not participate,” said Mark Schulman, Abt SRBI Co-Founder and Research Chief. “His addition to our Advanced Methods Group gives us a strong advantage in this vital area.”
He holds a B.S. in Statistics at the University of São Paulo, Brazil and holds a Masters in Statistics from the Institute of Mathematics and Statistics, Universidade de São Paulo, São Paulo, Brazil. Prior to beginning his doctoral work at Michigan, he worked for four years as a survey statistician at Ipsos Public Affairs in Brazil.
Abt SRBI’s Advanced Methods and Sampling Group has a staff of nationally recognized sampling and statistical experts known for their ability to meet clients’ needs for reliable and representative samples. The group works closely with all Abt SRBI Practice Groups.
Nishimura, R., Wagner, J. and Elliott, M. (2015), Alternative Indicators for the Risk of Non-response Bias: A Simulation Study. International Statistical Review, doi: 10.1111/insr.12100.