An experimental test was made of whether completion rates for a census questionnaire could be improved by offering the option of calling a toll-free number and providing the requested information to an interviewer as an alternative to mailing it back. Conducted on a national probability sample of house holds, five treatment panels were created to examine the effect of delivering, in different ways, the invitation to respond by telephone or mail. Offering such invitations did not improve completion rates except when included with a follow-up letter that did not also include a replacement questionnaire.