This study compared a national RDD telephone survey (by the Center for Survey Research at Ohio State University) with national Internet surveys (by Knowledge Networks and Harris Interactive) done before and after the 2000 U.S. Presidential Election using the same questionnaires. The samples recruited by the Center for Survey Research and Knowledge Networks were more representative of the nation than was the volunteer Harris Interactive sample in terms of demographics and electoral participation. The Harris Interactive sample was more engaged in and knowledgeable about politics than were the other two samples. The Harris Interactive sample manifested less random measurement error and less survey satisficing than did the other two samples. Data collected via the Internet manifested less social desirability response bias than the telephone data after controlling for differences between the samples in terms of demographics and political engagement.