With an expected mail survey response rate of between 25% and 40% (Newton &
Rudestam, 1999) between 700 and the entire population of 1,163 would have to be surveyed to generate a complete sample of 289. A survey this large can quickly consume available time and money. A realistic and affordable option was to survey 600 community colleges, using a web-based survey. In the original research in which the survey was used, Grandzol and Gershon (1998) achieved an initial response rate of 31% using a mail survey. Using follow up letters, they raised the response rate to 47%.