Improving Surveys with Paradata: Analytic Uses of Process Information

CHAPTER 6

DESIGN AND MANAGEMENT STRATEGIES FOR PARADATA-DRIVEN RESPONSIVE DESIGN: ILLUSTRATIONS FROM THE 2006--2010 NATIONAL SURVEY OF FAMILY GROWTH

NICOLE G. KIRGIS and JAMES M. LEPKOWSKI

Survey Research Center, Institute for Social Research, University of Michigan–Ann Arbor

6.1 INTRODUCTION

Survey design requirements have increased substantially over the last several decades. Today, scientists and policy makers require more extensive measures on larger samples, and to collect those measures on a more frequent basis from more specific segments of a population in greater geographic and demographic detail than ever before. At the same time, survey researchers have found that the environment for conducting surveys is one of increasing uncertainty regarding successful implementation.

Surveys on families and family growth, for example, clearly show the increasing demands, and uncertainties, of survey research. Family growth surveys have been conducted in the United States since 1955. The first such survey was conducted over the course of several months using paper and pencil methods and a population of Caucasian women aged 18--39 years who were currently married (Mosher and Bachrach, 1996). The most recently completed survey in the subsequent U.S. fertility survey series, the 2006–2010 National Survey of Family Growth (NSFG), obtained interviews from women and men 15--44 years of age regardless of race, ethnicity, or marital status (Groves et al., 2009). The 2006–2010 interviews were ...