This study aims to improve survey response behavior by including schedules of reinforcement with financial incentives in a 12-week real-world field experiment. We found that lottery based-incentives produce more resistance to extinction and are also more cost-efficient than fixed honorariums, despite having the same expectations. We also found that immediate reinforcement and the high chance, low prize incentives reduce underreporting bias. In addition, we confirmed the resistance to extinction when switching incentives and sustained responses due to lottery-based incentives. Moreover, we evaluated the past reward winning experience and the habit of past responses, and found that experience influenced subsequent survey responses. Thus, participants’ responses are dependent on incentive programs.