As more and more undergraduate students act as voluntary tutors to rural pupils after school, there is a growing need for a supporting environment to facilitate adaptive instruction. Among others, a teaching method retrieval system is intended to help tutors find relevant teaching methods for teaching a particular concept. However, teaching methods demanded by tutors might not be in well-organized forms or even unwritten. Thus, the technology of conventional information retrieval cannot be simply applied to retrieve teaching methods. Therefore, we propose to invite experts (teachers) to join the support environment and provide ad hoc teaching methods. Nevertheless, it is difficult for tutors to decide how long they will wait for replies from remote teachers, which is called a due time setting problem. Furthermore, a trade-off inherent in the problem is that an early reply does not necessarily mean a good one. Our idea is to propose an interactive approach to deal with the uncertainty during retrieving teaching methods. In this study, we define this problem and evaluate several waiting strategies. A prototype has been designed and implemented, and experiments have been conducted to evaluate the performance of the four waiting strategies. The results show that waiting strategies have a significant impact on retrieval performance. (Contains 4 tables and 13 figures.)