According to a new analysis of data from the US National Survey of Family Growth, there is no relationship between giving blood any facet of religiosity. Neither the religion in which the person was raised (versus none), nor religious service attendance, nor the importance of religion in daily life, were related to whether the person had given blood in the past.
In terms of raw numbers, women raised as mainline protestants were slightly more likely to have given blood than people from other religions or none (see figure), but these effects disappeared once the statistics were adjusted for other factors (for example, people who are born in the USA are more likely to give blood than are immigrants). There was no effect among men, even in the unadjusted data.
By cutting and dicing the statistics, the authors (Frank Gillum at Howard University and Kevin Masters at Syracuse) were able to find occasional groups that seemed to be more generous donors (Catholic men aged 35-44, for example), but I think they have fallen foul of the problem of multiple comparisons. If you make enough groups (nearly 50 in this case) some are going to come out high just by random chance.
This is a surprising result, given that the religious are supposed to be more charitable and pro-social than the non-religious. If nothing else, you would expect religious service attenders to donate more, simply because (in the US at least) conscientious and dutiful people are expected to do both.
The authors think that it might be because religiously-motivated charity is primarily directed to people of the same ‘tribe’. The problem with blood donations is that anyone could benefit – even people who are outsiders.
That would certainly fit with other research into charitable giving (some 80% of evangelical charitable giving goes to other evangelicals), and charitable giving to co-religious is inversely related to support for broad-based state welfare. It also fits with theories that explain religion as an invented (or socially evolved) tool to increase group solidarity.
But I can’t help thinking that another process might also be at work here. One of the interesting things about religion is that, although religious people tend to report that they are more pro-social, when tested in controlled conditions they are not (except when previously primed with religious messages).
Now, blood donations are all traceable. That means you might think twice to make sure that you give an accurate report of your donation history, and resist the temptation to inflate your contributions to match your self image!
Gillum RF, & Masters KS (2010). Religiousness and blood donation: findings from a national survey. Journal of health psychology, 15 (2), 163-72 PMID: 20207660