Early promise for the Internet as a tool to make social research questionnaires easier and cheaper to deliver is not fully realized. This study reports a mixed-mode survey of 1,100 Australian researchers. When respondents were given the choice to complete either web- or paper-based versions of a questionnaire, the majority chose the paper-based mode. Web respondents were more likely to be young, male, middle ranking, and working in information technology-related sectors. The authors highlight the need to determine how far alternate delivery modes increase response rates. For mixed-mode surveys to be financially and methodologically worthwhile, the authors propose that the initial sample size be at least 1,000 individuals, this figure depending on the demographic characteristics of the sample.