Integration of the National and International 2008 SDR: Bridging Effects and Expected Improvements to the Time Series Data

• Activity Number: 607
• Type: Topic Contributed
• Date/Time: Thursday, August 2, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Section on Survey Research Methods
• Author(s): Y. Michael Yang*+ and Wan-Ying Chang and Stephen Cohen and Rachel Harter
• Companies: NORC and National Science Foundation and National Science Foundation and RTI International
• Keywords: longitudinal survey ; bridging effect ; weighting adjustments ; variance estimation ; time series

Abstract

Traditionally, the Survey of Doctorate Recipients (SDR) is a longitudinal survey that collects information from U.S. residing individuals with a doctoral degree in a science, engineering, or health field (SEH) from a U.S. institution. Beginning with the 2003 cycle, the SDR added a new component, the International Survey of Doctorate Recipients (ISDR) to represent U.S.-trained doctorates living outside the U.S. Prior to the 2010 cycle, the traditional SDR, now named the National Survey of Doctorate Recipients (NSDR), and ISDR were implemented as two separate surveys. In 2010, the survey sponsor, the National Science Foundation (NSF), developed and implemented a methodology to integrate the sampling frame, sample design, weighting adjustments, and variance estimation procedures for the NSDR and ISDR. The integrated SDR, including both the NSDR and ISDR, covers the entire population of U.S. trained SEH doctorates. This paper discusses the integration methodology and explores the impact of integration on the survey weights and reported estimates on the 2008 SDR data. We compare the population estimates, the distribution of the weights, and the weighted estimates for a set of key variables under the integrated design to those from the traditional NSDR program.