The benefits and disadvantages of online surveys have been researched primarily in terms of response rates and sampling issues. Relatively little attention has been given to other factors associated with the overall quality of data from self-administered questionnaires distributed via the Internet. Addressing this point, this paper focuses on open-ended questions and looks specifically at the length of answers and the item non-response rates produced by online questionnaires when compared with more traditional paper-based questionnaires. Using closely matched groups of respondents (n=466) it compares responses to open-ended questions from near-identical versions of online and paper questionnaires. Findings from the research indicate that online questionnaires tend to produce longer answers to

open-ended questions, although the evidence for this is not strong. The evidence is more persuasive in relation to the item non-response rates where it has been foundthat, although the administration of questionnaires online might have relatively littleimpact on fixed-choice questions, it would seem to reduce item non-response rates significantly where the questions are open-ended and require respondents to provide unstructured text-based answers. Possible reasons for these findings will be considered and the implications for qualitative research will be discussed.