Social network research often depends on the willingness of respondents to provide personal information about themselves and alters. Survey design strategies that increase willingness to share this information are necessary for social network research to be feasible, especially when name generators are used for sampling because rosters are unavailable. We conducted an experiment in which one group of respondents (n = 94) received an online survey that included an example network map and the other group (n = 100) received one that did not. Results show the map did not increase or decrease provision of network contacts nor influence the types of ties reported. Furthermore, respondents were reluctant to provide names and contact information of alters. Our study demonstrates the difficulty of collecting network information without a previously defined roster and an attempt to improve data collection through strategic survey design.