Saturday, September 3, 2011

The Great Prayer Experiment

One of the topics Dawkins discusses in The God Delusion is the "great prayer experiment" of 2006. 

Two previous double-blind studies (1) (2) had shown that intercessory prayer had significantly improved the health of patients admitted to coronary units of hospitals. And a third study was funded by the Templeton Foundation to confirm or refute the results of the other two studies.

The design of the latest prayer study (3) was somewhat unusual. The researchers used three patient groups. Two groups were advised of the study, but were not told whether they were in the prayer group or placebo group. The third group knew that they were being prayed for. The study was performed at six hospitals. Out of 3295 eligible patients, 1493 (45%) refused to participate, which is very high, although they did not explain the reasons for non-participation. The intercessors were composed of three groups. Two were Roman Catholic and one was a Protestant group. Unlike in previous studies, the intercessors were not allowed to pray their own prayers. The prayers were given to them by the study coordinators to "standardize" the prayers. The discussion section of the paper suggested that at least some of the intercessors were dissatisfied with the canned nature of the prayers. In attempting to standardize prayer, I believe the study introduced a serious flaw, since most intercessors tend to pray as they are led by the Spirit, instead of praying prepared scripts. Jesus told His followers not to pray repetitiously, since God would not hear those kinds of prayers. (4)

Ultimately, the results showed that groups 1 (prayer) and 2 (no prayer) were identical, whereas group 3 (those who knew they were being prayed for) did worse than the other two groups. However, the lack of efficacy of intercessory prayer in this study could be due to theological problems associated with the study design.

Dawkins fails to cite prayer studies that showed a positive result, instead reporting only on the one that showed a negative result. In fact, a new meta-analysis, which takes into account the entire body of empirical research on intercessory prayer (17 major studies), shows that according to American Psychological Association Division 12 criteria, intercessory prayer is classified as an experimental intervention that, overall, shows a small, but significant, positive effect.(5)


  1. Prayer studies are useless both ways - trying to prove it to be effective or ineffective - as even the positive effects aren't necessarily "answers to prayer" (healing) - but rather the person was more upbeat, more positive, or whatever... which could result from knowing you have a community supporting you or whatever else. And of course the non-effects or negative effects could still be categorized as "answers to prayer" by Christians - with the answers being "no" or "not now". There are so many variables that these tests just don't cut it.

    ...of course the Bible offers a good way to test prayer in the story of Elijah and the Prophets of Baal (as I've mentioned in my blog). But no Christian would allow prayer to be tested that way today. However, just look at the results of Rick Perry's "prayer rallies" for rain in Texas. It has only got worse, and worse, and worse. Eventually it will rain (since that is bound to happen naturally) and it will be "an answer to prayer". Eventually, Texas will be totally recovered (since that is bound to happen naturally) and it will be a "miracle".

    There is simply no empirical evidence that prayer works - only confirmation bias and interpreting random events and coincidences as answers to prayer. On the other hand, it is also impossible to prove prayer doesn't work when any potential "answer" - yes, no, not now - can be considered acceptable regardless of how mind-boggling the response may seem. As long as we accept the answer that "God acts in mysterious ways", the results of prayer - no matter how strange - will always be acceptable to believers.

  2. I agree that doing studies to test whether or not prayer works are pretty much useless. I dislike that Dawkins used it as an example in The God Delusion as evidence that God does not exist.