Extension professionals in agriculture increasingly rely on the Internet to survey growers, yet little is known how responses differ between electronic and traditional hardcopy approaches. We surveyed small grain growers in Idaho, Montana, and Wyoming to determine their nitrogen and crop management practices, understanding of urea volatilization, and preferred media for receiving research results. In Montana, a survey was sent to all 1063 Montana Grain Growers Association (MGGA) members. A mixed-mode data collection approach was utilized to maximize response rate, improve timeliness, and reduce coverage and non-response errors. Specifically, the web address of the online survey was emailed to the 716 members who had provided an email address to MGGA, and a hardcopy paper survey was mailed to the remaining 347 members. The online link and hardcopy surveys were also provided at several grower meetings. Grain grower associations in Wyoming and Idaho conduct their mailings only by regular mail; therefore, the online survey was advertised in mailings and hardcopy surveys were distributed at grower meetings. A total of 147 surveys were completed in the three states. In Montana, both online (n=41) and hardcopy (n=50) survey responses were sufficient for statistical analyses using Fisher’s Exact Test, whereas online response numbers were insufficient for Idaho and Wyoming. In Montana, responses to 13 of the 15 questions did not different significantly (P=0.05) between the online and hardcopy groups. The online group did have a better understanding of climatic conditions that exacerbate urea volatilization. The two groups also preferred to receive research results in different formats. The top media preferences of online respondents were email (38%) and webpage (20%), compared to newsletters (32%) and workshops (24%) for hardcopy respondents. Extension professionals relying on survey results to make decisions about their programs should choose the survey delivery method carefully to avoid the potential for sample selection bias.