Implications of Using Survey Routers in Marketing Research

• Activity Number: 395
• Type: Topic Contributed
• Date/Time: Tuesday, August 2, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and Marketing
• Author(s): Shon Magnan*+
• Companies: GfK Custom Research North America
• Keywords: marketing research ; sampling ; routers

Abstract

Survey routers in online research are used to assign incoming respondents to open surveys for which they are likely to quality. Router benefits touted include enhanced efficiency, support for the fielding of surveys to low-incidence populations and respondent experience by minimizing the number of screen-outs or "quota full" messages. Routers however also can introduce sampling bias especially if priority routers are used instead of random routers. Priority routers assign respondents who quality for multiple surveys to one survey in preference over all others. Prioritizing surveys may be efficient but can also create selection bias as the set of respondents sent to unprioritized surveys differ from the potential respondents who enter the router. Recent research suggests that although there is indeed bias introduced, the bias does not affect the outcome of the research. This presentation discusses the issue of survey routers, the potential bias introduced and explores the affect of this bias via a series of simulations.