140 – Address-Based Telephone RDD and Cell Phone Sampling
Listening Analysis of Personally Recruited Panelists Using Wilcoxon Tests
Kelly Dixon
Arbitron, Inc.
William Waldron
Arbitron, Inc.
Arbitron implemented In-Person Recruitment (IPR) of survey respondents in July 2010. The purpose was to improve the response rate and sample representation for certain demographic groups that were difficult to engage using traditional RDD or mail recruitment methods. This mode of selecting panelists poses a question, however. Does the new method unduly influence the listening patterns of panelists selected via IPR with that of traditionally sampled panelists? In order to investigate this question, we implemented a three-pronged analysis of the radio listening data: running linear regressions to control for various demographic factors, directly comparing the mean listening levels, and a two-sample Wilcoxon test. Furthermore, we repeated much of the tests under various stratifications, including different months, genders, age groups, races, and markets. The following paper details an applied utilization of the third prong: the two-sample rank sum (Wilcoxon) test.
