Eleonore Stump argues that in some cases, God waits for us to ask for something before granting it in order to avoid spoiling or overwhelming us. We could be spoiled by God if God answered all of our prayers automatically, and we could be overwhelmed by God if God provided everything good for us without waiting for us to ask first (Stump 1979). In a similar vein, Michael Murray and Kurt Meyers argue that by making the provision of certain things dependent on petitionary prayer, God helps us to avoid idolatry, which is a sense of complete self-sufficiency that fails to recognize God as the source of all good things. They also say that requiring petitionary prayer in some cases helps us to learn about God's will as we recognize the patterns in prayers answered (and not answered: see Murray and Meyers 1994 and section 5 below).
A number of people have tried to conduct statistical studies to determine whether or not petitionary prayer is effective. These studies try to measure the differences between groups of people, one of which is the subject of petitionary prayers, and the other of which is not. But these studies seem to be flawed from the outset (see Brümmer 2008). First of all, there is no way to control the groups so that any group of people is the subject of no petitionary prayers, since it is impossible to prevent people from praying for any particular person. Second, God is assumed to be a free person, not a natural force that acts automatically in all similar cases, so we cannot assume that God will simply ignore those people for whom petitionary prayers have not been offered.