Despite the widespread use of online panels in contemporary survey research there remain ongoing concerns that the incentivized nature of panels may encourage some people to go to elaborate lengths to maximize participation and therefore their incentives. This can include creating multiple false identities to increase their chances of qualifying for a survey. To address this issue, most online panel companies now use a validation processes to protect against false registrations. These processes validate the identify of every respondent by collecting basic identifying information (name, address, and date of birth) and comparing it to third party databases that claim to have data on a very high percentage of U.S. households. In 2011, Courtright and Miller conducted a study to assess the impact of validation on online samples. They found that respondents who could not be validated, either because they refused to supply the requested personal information or could not be matched in third party databases, were demographically, attitudinally, and behaviorally different from those who could be validated. In 2012, we replicated Courtright and Miller's design with some added questions about the frequency of Internet use and attitudes toward online privacy. As in the initial study, large numbers of potential respondents refused to provide the requested personal information, and those who refused were demographically different from those who complied. They reported spending less time on the Internet and had stronger concerns about online privacy than those who complied and were validated. However, these differences were not so great nor was the non-validated group so large that they created significant differences in the overall survey results.