Department of Statistics, The Ohio State University
Statistics and Biostatistics Colloquium Series

Web surveys have become very popular in recent years because they are inexpensive relative to phone or mail surveys and because they are very fast. The main drawbacks of web surveys are the challenge to draw representative samples of the general population, whether and how to deal with respondents without Internet access and how to mitigate nonresponse bias. I will discuss some proposed solutions including adjustments for selection bias using propensity scoring, the use of WebTV's for people without Internet access, and web survey panels. Examples are drawn from work with the American Life Panel, the Health and Retirement Study, the Harris Interactive Panel and the Knowledge Network panel.

Meet the speaker in Room 212 Cockins Hall at 4:30 p.m. Refreshments will be served.