Arbitron’s paper and pencil radio listening diary has proven itself as a reliable and cost-effective survey instrument for the past 40 years. Recent declines in response rates, along with increased Internet penetration and usage, however, have prompted Arbitron to test supplementing the paper diary with an Internet version, resulting in a mixed-mode design. Two recent experimental design tests together included 118,000 randomly selected households in 63 metropolitan areas across the U.S. During recruitment phone calls, the test groups were told they would receive paper diaries in the mail but would also have the choice to complete the diary online. These households received paper diaries along with a special insert explaining the online option and instructions for logging on to the eDiary website. All subsequent phone and mail contacts also mentioned the eDiary option to test group respondents. Control group respondents were not told of the eDiary option in our phone calls or survey materials, and were mailed only paper diaries. This presentation will discuss use patterns of the online response option observed in these two tests, including the percent of those offered the option who registered for the eDiary and the percent of total completed diaries made up of online submissions. Differences in these metrics by demographic characteristics will be presented. Also discussed will be the percent of eDiary registrants who submitted a completed eDiary, who submitted a paper diary instead, and who registered but did not return either type of diary. We will present results of callback studies that provide insight into why respondents chose a given survey mode and how offering a choice of mode may affect a respondent’s propensity to participate in a survey.