Extended abstract
For believers, God's answer to their petitionary prayers is important both for the fulfillment of their requests and for the God's special attention to them by doing so. Facing with general and specific guarantees offered by God in religious traditions for granting prayers, believers often consider the obtaining of the state of affairs they petitioned for as a definite evidence of the prayer being answered by God. Some philosophers of religion, however, have doubted that we can certainly know a prayer is answered and have raised the epistemological problem of prayer. Here, the basic question is how we can know that the obtaining of the petitioned state of affairs after the prayer, was only due to the servants' prayers delivered to God, not other reasons such as God's previous plan. Some theologians like Davison are skeptical about petitionary prayer, while theologians like Choi hold that, based on some evidence, it is possible for us to be certain about petitionary prayers being granted. The main purpose of this article is to compare, with a descriptive-analytical method, Davison's objections and Choi's answers concerning the epistemological problem of prayer. Criticizing both views, an attempt has been made to introduce the kind of certainty required in the theology of prayer.
 
Davison's Objections
Considering the traditional theism according to which an all-knowing, all-powerful, and perfectly good God has created the world, sustains it in being from moment to moment, and in accordance with a plan, providentially guides all events toward a good ending, Davison examines and criticizes the three accounts he postulates for answered prayer in response to the epistemological problem: "pure correlation", "modified correlation" and "counterfactual dependence". According to the "pure correlation" account, if a person prays that a certain event happen, and that the event in question actually happens, then we can know that the prayer has been answered. Davison argues that this account is not correct because if people pray for bad things to happen, and they actually do happen, then we should not conclude that the prayers in question were answered; God's moral perfection prevents Him from cooperating in the production of evil. Therefore, the occurrence of such events should be deemed as coincidences, not as answered prayers. Due to this restriction, Davison's other account, "modified correlation", is activated, according to which, the divine response is exclusive to the petitions for good. So, whenever a person prays for something good to happen, and it happens, then God has answered a prayer. Davison, however, is not satisfied by this suggestion, either; For it could also happen to be a coincidence and, for example, the event in question happens to be just a part of God’s plan for the world, in such a way that God’s bringing it about had nothing to do with anyone’s prayers at all. The third account of effective prayer, "counterfactual dependence", says that for a prayer to be answered, it must be the case that if the person had not prayed for the event in question, then it would not have occurred. This view is not satisfactory for, on the one hand, it demands too much in the sense that it makes the divine response dependent upon one's request, and on the other hand, it is too weak because on the assumption that prayer is necessary for the realization of an event, there are other possible ways in which it might be true that the event in question would not have happened if one had not prayed for it to happen. For example, it might be the case that the event in question was somehow caused by one's very act of praying and was only a natural result of that person's prayer, not an answer from God.
 
Choi's answers
According to Choi, if we consider a set of evidence along with the occurrence of the event in question, we can have good reasons to believe that the prayer was answered. Some of these evidences include: 1. Timing: the precise timing between our prayer and the obtaining of the state of affair in question can serve as a good reason; 2. Specificity: this involves prayers that ask for very specific things and are followed by precisely those things happening; the more detailed and specific the request, the more clearly its obtaining can prove the efficacy of prayer; 3. Statistical Inference: the large number of cases of provision after prayer gives the petitioners more assurance that there is an actual relationship between their prayer and God's provision.; 4. Internal Assurance: one's own internal assurance and personal intuition can be a direct evidence.
 
Discussion and Conclusion

Davison's objections to the pure correlation account seems to be based on a view that sees the occurrence of moral evil in the world and the existence of God incompatible. If one can be justified in accepting the existence of both moral evil and God, they can also be justified to believe that such prayers could be answered, and hence, there would be no need for the modified correlation account.
Although it can hardly be claimed with logical certainty that only our prayers have been effective in the occurrence of events in question, considering some evidence can increase the plausibility of the belief that God has granted them because of our prayers.
Although it is very difficult to achieve logical certainty in most fields of human knowledge, including the theology of prayer, one can be satisfied with a kind of psychological certainty regarding the problem of epistemology of prayer; a psychological certainty that can essentially improve the believers' faith and serenity.
In solving the epistemological problem of prayer, along with ways such as direct revelation, as suggested by Davison, we can benefit from reasoning through the best explanation, as Choi did. In this proposal, by considering a set of evidence in favor of answered prayer, we may improve the plausibility of the belief that the prayer was answered.