Activity Number:
|
284
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Survey Research Methods
|
Abstract - #303909 |
Title:
|
One-Time Contact for a Census Internet Survey: Is It Sufficient? Washington State Professional Health Worker Census Internet Mode Effect and Response Rate Study
|
Author(s):
|
Danna L. Moore*+ and Bruce Austin and Kent Miller
|
Companies:
|
Washington State University and Washington State University and Washington State University
|
Address:
|
Social & Economic Sciences Research Center, Pullman, WA, 99164-4014,
|
Keywords:
|
census survey ; mode effects ; propensity scores ; mixed mode ; health policy ; bias
|
Abstract:
|
Can a one contact produce defensible survey results in a census of 184,000 licensed professional health workers? In response to a legislative mandate health workers were contacted by letter and asked to respond to a web survey. The survey was limited to a one time contact only for the majority of respondents. Anticipating a concern for low response and representativeness, an experiment was implemented for 32 groups to allow evaluation of the effect on response rates, data quality, and characteristics. One third were randomly assigned a multi-contact mixed mode treatment for comparison. License characteristics were available for analysis and comparison to survey completions for 32 professions. Results show significant differences in response to the "one time only" compared to "multi-contact mixed mode" surveys.
|