Dual Frame Theory Applied to Landline and Cell Phone Surveys
Nov 10 2009
The SRMS will host the first webinar at 1-3PM EST on Tuesday, 10 November 2009. Mike Brick will present "Dual Frame Theory Applied to Landline and Cell Phone Surveys."
About SRMS Webinars
At each 2 hour session webinar, participants register for a modest fee. Fees may vary from webinar to webinar depending on the length of the presentation and expected audience.
Each registration is allowed one web connection and one audio connection. The section encourages multiple persons to view each registered connection. For example, if your department or agency has a conference room with a computer, projector, screen, and a speakerphone, you can pay for one registered connection and have all of your staff attend the webinar in the conference room.
The access information and presentation materials will become available to registered participants two business days before the scheduled webinar time. The presenter controls the flow of the presentation. When the presenter run software or show Web pages, the audience will see this on their computer screens in real time. Generally the audience does not ask questions on phone. Questions for the presenter are submitted using the chat feature found on the webinar web page. The presenter can see the questions as they come in and choose to answer.
Abstract
As the number of households that have only cell phones has increased dramatically over the past 5 years, telephone surveys have addressed this problem by sampling from both landline and cell phone numbers. One of the issues emerging from these dual frame surveys is that the theoretical foundation for these surveys largely ignores nonsampling errors. Because these errors may be large and result in biases, they must be considered in dual frame telephone surveys. This Webinar begins with a review of dual frame theory with particular attention to surveys that sample landline and cell phone numbers. It then examines the effect of nonsampling errors when surveys are conducted without considering these errors. In particular, we describe the potential effect of nonresponse and measurement error using data from surveys of landlines and cell phone numbers. We discuss both practical sample design issues such as whether to screen for cell-only households, and weighting methods to reduce the effects of the errors. The advantages and disadvantages of different sample designs and estimation methods are discussed. The examples are from actual dual frame telephone surveys.
Links to website:
This article has been read 5214 times
