John Finamore, SRMS Publications Officer

The Survey Research Methods Section (SRMS) will sponsor the following three roundtables at the 2014 Joint Statistical Meetings. Consider signing up for these options when you register for the conference. Remember, space is limited, so sign up soon.

• Valid Analytic Properties and Disclosure Limitation for Microdata, led by William Winkler of the U.S. Census Bureau
  There is considerable interest in producing public-use data that allow analyses that reproduce a few analyses from original, confidential microdata. Several methods/tools facilitate the production of synthetic (or partially synthetic) data with valid analytic properties that allow provision of public-use data with reduced re-identification risk.
• Practical Tools for Designing and Weighting Survey Samples, led by Jill Dever of RTI International
  Survey sampling is fundamentally an applied field. During this roundtable, we will discuss techniques long used by experienced survey statisticians with little or no references in the literature. Although we will use the textbook Practical Tools for Designing and Weighting Survey Samples by Richard Valliant, Jill Dever, and Frauke Kreuter as the basis for discussion, participants are encouraged to share their experiences. This roundtable will benefit students seeking a more in-depth understanding of applied sampling, survey statisticians searching for practical guidance on sampling and weighting, and other survey practitioners who desire insight into the design and implementation of survey samples.
• Analyzing Survey Nonresponse, led by Polly Phipps of the Bureau of Labor Statistics
  Survey nonresponse and associated risks of nonresponse bias are major concerns for government agencies and other organizations conducting surveys to produce statistics. At the time of data collection, respondents may choose to decline survey participation (unit nonresponse) or participate selectively by responding to certain data items and not others (item nonresponse). This type of analysis often requires modeling the probability that a given unit will respond to the survey based on known unit characteristics. Often, the choice model method and the intended use of the model are quite different in this context than when adjusting estimates for nonresponse. We will discuss methods for modeling and identifying unit characteristics associated with nonresponse. We will emphasize methods designed to help understanding the effect these unit characteristics have on the response rate throughout the data collection process.

For more information about these roundtables, visit the JSM 2014 online program and search on SRMS-sponsored events.