Cross-cultural measures of superstition, part deux

Posted on October 20, 2008


A bit over a year ago I wrote a post in which I discussed the difficulties with comparing how superstitious two different societies are. The basic problem may be understood if we consider the following question: Do you believe that number eight is lucky? If this question were used to judge the superstitiousness of a society it would be found that the Chinese are the most superstitious nation in the world – it was no accident that this year’s olimpic games started in the eighth year, in the eighth month, on the eight day, at the eight hour, at the eighth minute and, yet, at the eighth second. Of course, the results would be very different if we’d asked: Do you believe that the number thirteen is unlucky? Quite simply, the content of superstitions tends to vary a lot between societies. The differences are no longer as great as they once were, of course – globalisation has seen to that. Still, a questionnaire that is adequate to the job of giving a rough measure of superstitious beliefs in one society is not appropriate to another. Still, this leads to an idea for providing a cross-cultural measure of superstition which requires a two-part method.

The first part of the research would require asking a group of people from a given society to write down ten superstitions that they know. They would have to be asked to do this without any prompting as to specific superstitions and it would have to be made clear in the question that they do not have to necessarily believe in the superstitions they list. The results could then be used to form a list of superstitions that are commonly recognised within the given society. This could be turned into a second questionnaire of, perhaps, ten entries that would be culture-specific. The second part of the research would consist in getting a different group of people from that same society to respond whether they, themselves, believe in those particular superstitions. The resulting numbers may then be compared to the numbers obtained from other societies without need to refer back to the specific superstitions that were espoused. More importantly, the resulting cross-cultural measures could then be correlated with other societal characteristics such as wealth.

Of course, the list of possible confounds is staggering, starting with all the usual problems related to using questionnaires. The exact method to be used may need to modified to deal with some of them.  Given my lack of training in psychology I fear I may have missed something obvious. However, the basic idea can’t be non-sensical since the methods to be used in both parts of the study have been used before but just not together in the one study. Indeed, having thought of this way of doing things I wonder how Gallop and others who try to measure the level of superstitious beliefs come up with their lists of questions. I suspect that all too often not enough thought goes into these lists.