Student Paper Competition Award Winners Announced

Nancy Clusen, 2011 SSS Program Chair

Each year, the Social Statistics Section cosponsors a student paper competition with the Government Statistics Section and Survey Research Methods Section. The overall quality of submissions this year was excellent, and we found ourselves having to decide between papers that were extremely close in terms of quality. This year’s winners are the following:

• Brady Thomas West, University of Michigan, “Bayesian Analysis of Between-Group Differences in Variance Components in Hierarchical Generalized Linear Models”

  The paper was the final project for a course in applied Bayesian inference. The general concept came from Thomas’s interest in applications of multilevel modeling for studies of interviewer variance and a lack of small sample methods in the literature for comparing models with different variance components for different groups of higher-level units (e.g., interviewers).

• Qi Dong, University of Michigan, “Combining Information from Multiple Complex Surveys”

  The paper is the first of three in Dong’s dissertation. The other papers are about the nonparametric method to generate synthetic populations and an application of this new method when some variable of interest is missing in one survey.

• Dan Liao, RTI International, “Variance Inflation Factors in the Analysis of Complex Survey Data”

  This paper is part of Liao’s dissertation, which adapted and extended the conventional collinearity diagnostics techniques to the survey setting for generalized linear regression analysis of complex survey data, under both the model‐based and design‐based methods of inference.

• Joseph Sakshaug, University of Michigan, “Synthetic Data Generation for Small-Area Estimation in the American Community Survey”

  This paper is based on Sakshaug’s dissertation research.

• Anna Sikov, Hebrew University of Jerusalem, “Imputation and Estimation Under Nonignorable Nonresponse for Household Surveys with Missing Covariate Information”

  The paper, part of Sikov’s dissertation, proposes a new approach to handling nonignorable nonresponse and shows that the approach can be applied to real survey data.

The authors will present their papers at JSM in Miami Beach, Florida, and be recognized at the sections’ business meetings. A subsidy of up to $800 was provided to each winner to cover JSM 2011 expenses.